Editorial

10 pages, 249 KiB

Open AccessEditorial

by Gniewko Niedbała and Sebastian Kujawa

Agriculture 2023, 13(9), 1686; https://0-doi-org.brum.beds.ac.uk/10.3390/agriculture13091686 - 26 Aug 2023

Cited by 1 | Viewed by 1246

Digital agriculture, defined as the analysis and collection of various farm data, is constantly evolving [...] Full article

(This article belongs to the Special Issue Digital Innovations in Agriculture)

Research

Jump to: Editorial, Review, Other

17 pages, 4448 KiB

Open AccessArticle

Deep CNN-Based Planthopper Classification Using a High-Density Image Dataset

by Mohd Firdaus Ibrahim, Siti Khairunniza-Bejo, Marsyita Hanafi, Mahirah Jahari, Fathinul Syahir Ahmad Saad and Mohammad Aufa Mhd Bookeri

Agriculture 2023, 13(6), 1155; https://0-doi-org.brum.beds.ac.uk/10.3390/agriculture13061155 - 30 May 2023

Cited by 1 | Viewed by 1302

Abstract

Rice serves as the primary food source for nearly half of the global population, with Asia accounting for approximately 90% of rice production worldwide. However, rice farming faces significant losses due to pest attacks. To prevent pest infestations, it is crucial to apply [...] Read more.

Rice serves as the primary food source for nearly half of the global population, with Asia accounting for approximately 90% of rice production worldwide. However, rice farming faces significant losses due to pest attacks. To prevent pest infestations, it is crucial to apply appropriate pesticides specific to the type of pest in the field. Traditionally, pest identification and counting have been performed manually using sticky light traps, but this process is time-consuming. In this study, a machine vision system was developed using a dataset of 7328 high-density images (1229 pixels per centimetre) of planthoppers collected in the field using sticky light traps. The dataset included four planthopper classes: brown planthopper (BPH), green leafhopper (GLH), white-backed planthopper (WBPH), and zigzag leafhopper (ZIGZAG). Five deep CNN models—ResNet-50, ResNet-101, ResNet-152, VGG-16, and VGG-19—were applied and tuned to classify the planthopper species. The experimental results indicated that the ResNet-50 model performed the best overall, achieving average values of 97.28% for accuracy, 92.05% for precision, 94.47% for recall, and 93.07% for the F1-score. In conclusion, this study successfully classified planthopper classes with excellent performance by utilising deep CNN architectures on a high-density image dataset. This capability has the potential to serve as a tool for classifying and counting planthopper samples collected using light traps. Full article

(This article belongs to the Special Issue Digital Innovations in Agriculture)

► Show Figures

Figure 1

16 pages, 6373 KiB

Open AccessArticle

Vegetation Indices for Predicting the Growth and Harvest Rate of Lettuce

by Ana Luisa Alves Ribeiro, Gabriel Mascarenhas Maciel, Ana Carolina Silva Siquieroli, José Magno Queiroz Luz, Rodrigo Bezerra de Araujo Gallis, Pablo Henrique de Souza Assis, Hugo César Rodrigues Moreira Catão and Rickey Yoshio Yada

Agriculture 2023, 13(5), 1091; https://0-doi-org.brum.beds.ac.uk/10.3390/agriculture13051091 - 19 May 2023

Cited by 3 | Viewed by 1690

Abstract

Urbanization has provided greater demand for food, and the search for strategies capable of reducing waste is essential to ensure food security. Lettuce (Lactuca sativa L.) culture has a short life cycle and its harvest point is determined visually, causing waste and [...] Read more.

Urbanization has provided greater demand for food, and the search for strategies capable of reducing waste is essential to ensure food security. Lettuce (Lactuca sativa L.) culture has a short life cycle and its harvest point is determined visually, causing waste and important losses. Using vegetation indices could be an important alternative to reduce errors during harvest definition. The objective of this study was to evaluate different vegetation indices to predict the growth rate and harvest point of lettuce. Twenty-five genotypes of biofortified green lettuce were evaluated. The Green Leaf Index (GLI), Normalized Green Red Difference Index (NGRDI), Spectral Slope Saturation Index (SI), and Overall Hue Index (HUE) were calculated from images captured at 1, 8, 18, 24, and 36 days after transplanting (vegetative state). The diameter and average leaf area of plants were measured using QGIS software. Green mass, number of leaves, and plant and stem diameter were measured in the field. The means were compared using the Scott–Knott test (p ≤ 0.05) and simple linear regression models were generated to monitor the growth rate, obtaining R² values ranging from 62% to 99%. Genetic dissimilarity was confirmed by the multivariate analysis presenting a cophenetic correlation coefficient of 88.49%. Furthermore, validation between data collected in the field versus data obtained by imaging was performed using Pearson’s correlations and showed moderate to high values. Overall, the vegetation indices SI, GLI, and NGRDI were efficient for monitoring the growth rate and determining the harvest point of different green lettuce genotypes, in attempts to reduce waste and losses. It is suggested that the definition of the harvest point based on vegetation indices are specific for each genotype. Full article

(This article belongs to the Special Issue Digital Innovations in Agriculture)

► Show Figures

Figure 1

13 pages, 25790 KiB

Open AccessArticle

Machine-Learning Approach to Non-Destructive Biomass and Relative Growth Rate Estimation in Aeroponic Cultivation

by Oskar Åström, Henrik Hedlund and Alexandros Sopasakis

Agriculture 2023, 13(4), 801; https://0-doi-org.brum.beds.ac.uk/10.3390/agriculture13040801 - 30 Mar 2023

Cited by 1 | Viewed by 1287

Abstract

We train and compare the performance of two machine learning methods, a multi-variate regression network and a ResNet-50-based neural network, to learn and forecast plant biomass as well as the relative growth rate from a short sequence of temporal images from plants in [...] Read more.

We train and compare the performance of two machine learning methods, a multi-variate regression network and a ResNet-50-based neural network, to learn and forecast plant biomass as well as the relative growth rate from a short sequence of temporal images from plants in aeroponic cultivation. The training dataset consists of images of 57 plants taken from two different angles every hour during a 5-day period. The results show that images taken from a top-down perspective produce better results for the multi-variate regression network, while images taken from the side are better for the ResNet-50 neural network. In addition, using images from both cameras improves the biomass estimates from the ResNet-50 network, but not those from the multi-variatemultivariate regression. However, all relative growth rate estimates were improved by using images from both cameras. We found that the best biomass estimates are produced from the multi-variate regression model trained on top camera images using a moving average filter resulting in a root mean square error of 0.0466 g. The best relative growth rate estimates were produced from the ResNet-50 network training on images from both cameras resulting in a root mean square error of 0.1767 g/(g·day). Full article

(This article belongs to the Special Issue Digital Innovations in Agriculture)

► Show Figures

Figure 1

19 pages, 2416 KiB

Open AccessArticle

Prediction of Pea (Pisum sativum L.) Seeds Yield Using Artificial Neural Networks

by Patryk Hara, Magdalena Piekutowska and Gniewko Niedbała

Agriculture 2023, 13(3), 661; https://0-doi-org.brum.beds.ac.uk/10.3390/agriculture13030661 - 12 Mar 2023

Cited by 7 | Viewed by 1919

Abstract

A sufficiently early and accurate prediction can help to steer crop yields more consciously, resulting in food security, especially with an expanding world population. Additionally, prediction related to the possibility of reducing agricultural chemistry is very important in an era of climate change. [...] Read more.

A sufficiently early and accurate prediction can help to steer crop yields more consciously, resulting in food security, especially with an expanding world population. Additionally, prediction related to the possibility of reducing agricultural chemistry is very important in an era of climate change. This study analyzes the performance of pea (Pisum sativum L.) seed yield prediction by a linear (MLR) and non-linear (ANN) model. The study used meteorological, agronomic and phytophysical data from 2016–2020. The neural model (N2) generated highly accurate predictions of pea seed yield—the correlation coefficient was 0.936, and the RMS and MAPE errors were 0.443 and 7.976, respectively. The model significantly outperformed the multiple linear regression model (RS2), which had an RMS error of 6.401 and an MAPE error of 148.585. The sensitivity analysis carried out for the neural network showed that the characteristics with the greatest influence on the yield of pea seeds were the date of onset of maturity, the date of harvest, the total amount of rainfall and the mean air temperature. Full article

(This article belongs to the Special Issue Digital Innovations in Agriculture)

► Show Figures

Figure 1

19 pages, 1178 KiB

Open AccessArticle

Development and Evaluation of a Deep Learning Based System to Predict District-Level Maize Yields in Tanzania

by Isakwisa Gaddy Tende, Kentaro Aburada, Hisaaki Yamaba, Tetsuro Katayama and Naonobu Okazaki

Agriculture 2023, 13(3), 627; https://0-doi-org.brum.beds.ac.uk/10.3390/agriculture13030627 - 06 Mar 2023

Cited by 4 | Viewed by 1874

Abstract

Prediction of crop yields is very helpful in ensuring food security, planning harvest management (storage, transport, and labor), and performing market planning. However, in Tanzania, where a majority of the population depends on crop farming as a primary economic activity, the digital tools [...] Read more.

Prediction of crop yields is very helpful in ensuring food security, planning harvest management (storage, transport, and labor), and performing market planning. However, in Tanzania, where a majority of the population depends on crop farming as a primary economic activity, the digital tools for predicting crop yields are not yet available, especially at the grass-roots level. In this study, we developed and evaluated Maize Yield Prediction System (MYPS) that uses a short message service (SMS) and the Web to allow rural farmers (via SMS on mobile phones) and government officials (via Web browsers) to predict district-level end-of-season maize yields in Tanzania. The system uses LSTM (Long Short-Term Memory) deep learning models to forecast district-level season-end maize yields from remote sensing data (NDVI on the Terra MODIS satellite) and climate data [maximum temperature, minimum temperature, soil moisture, and precipitation (rainfall)]. The key findings reveal that our unimodal and bimodal deep learning models are very effective in predicting crop yields, achieving mean absolute percentage error (MAPE) scores of 3.656% and 6.648%, respectively, on test (unseen) data. This system will help rural farmers and the government in Tanzania make critical decisions to prevent hunger and plan better harvesting and marketing of crops. Full article

(This article belongs to the Special Issue Digital Innovations in Agriculture)

► Show Figures

Figure 1

19 pages, 4170 KiB

Open AccessArticle

Multi-Node Path Planning of Electric Tractor Based on Improved Whale Optimization Algorithm and Ant Colony Algorithm

by Chuandong Liang, Kui Pan, Mi Zhao and Min Lu

Agriculture 2023, 13(3), 586; https://0-doi-org.brum.beds.ac.uk/10.3390/agriculture13030586 - 28 Feb 2023

Cited by 5 | Viewed by 1424

Abstract

Under the “Double Carbon” background, the development of green agricultural machinery is very fast. An important factor that determines the performance of electric farm machinery is the endurance capacity, which is directly related to the running path of farm machinery. The optimized driving [...] Read more.

Under the “Double Carbon” background, the development of green agricultural machinery is very fast. An important factor that determines the performance of electric farm machinery is the endurance capacity, which is directly related to the running path of farm machinery. The optimized driving path can reduce the operating loss and extend the mileage of agricultural machinery, then multi-node path planning helps to improve the working efficiency of electric tractors. Ant Colony Optimization (ACO) is often used to solve multi-node path planning problems. However, ACO has some problems, such as poor global search ability, few initial pheromones, poor convergence, and weak optimization ability, which is not conducive to obtaining the optimal path. This paper proposes a multi-node path planning algorithm based on Improved Whale Optimized ACO, named IWOA-ACO. The algorithm first introduces reverse learning strategy, nonlinear convergence factor, and adaptive inertia weight factor to improve the global and local convergence ability. Then, an appropriate evaluation function is designed to evaluate the solving process and obtain the best fitting parameters of ACO. Finally, the optimal objective function, fast convergence, and stable operation requirements are achieved through the best fitting parameters to obtain the global path optimization. The simulation results show that in flat environment, the length and energy consumption of IWOA-ACO planned path are the same as those of PSO-ACO, and are 0.61% less than those of WOA-ACO. In addition, in bump environment, the length and energy consumption of IWOA-ACO planned path are 1.91% and 4.32% less than those of PSO-ACO, and are 1.95% and 1.25% less than those of WOA-ACO. Therefore, it is helpful to improve the operating efficiency along with the endurance of electric tractors, which has practical application value. Full article

(This article belongs to the Special Issue Digital Innovations in Agriculture)

► Show Figures

Figure 1

26 pages, 12720 KiB

Open AccessArticle

Real-Time Plant Health Detection Using Deep Convolutional Neural Networks

by Mahnoor Khalid, Muhammad Shahzad Sarfraz, Uzair Iqbal, Muhammad Umar Aftab, Gniewko Niedbała and Hafiz Tayyab Rauf

Agriculture 2023, 13(2), 510; https://0-doi-org.brum.beds.ac.uk/10.3390/agriculture13020510 - 20 Feb 2023

Cited by 15 | Viewed by 7056

Abstract

In the twenty-first century, machine learning is a significant part of daily life for everyone. Today, it is adopted in many different applications, such as object recognition, object classification, and medical purposes. This research aimed to use deep convolutional neural networks for the [...] Read more.

In the twenty-first century, machine learning is a significant part of daily life for everyone. Today, it is adopted in many different applications, such as object recognition, object classification, and medical purposes. This research aimed to use deep convolutional neural networks for the real-time detection of diseases in plant leaves. Typically, farmers are unaware of diseases on plant leaves and adopt manual disease detection methods. Their production often decreases as the virus spreads. However, due to a lack of essential infrastructure, quick identification needs to be improved in many regions of the world. It is now feasible to diagnose diseases using mobile devices as a result of the increase in mobile phone usage globally and recent advancements in computer vision due to deep learning. To conduct this research, firstly, a dataset was created that contained images of money plant leaves that had been split into two primary categories, specifically (i) healthy and (ii) unhealthy. This research collected thousands of images in a controlled environment and used a public dataset with exact dimensions. The next step was to train a deep model to identify healthy and unhealthy leaves. Our trained YOLOv5 model was applied to determine the spots on the exclusive and public datasets. This research quickly and accurately identified even a small patch of disease with the help of YOLOv5. It captured the entire image in one shot and forecasted adjacent boxes and class certainty. A random dataset image served as the model’s input via a cell phone. This research is beneficial for farmers since it allows them to recognize diseased leaves as soon as they noted and take the necessary precautions to halt the disease’s spread. This research aimed to provide the best hyper-parameters for classifying and detecting the healthy and unhealthy parts of leaves in exclusive and public datasets. Our trained YOLOv5 model achieves 93 % accuracy on a test set. Full article

(This article belongs to the Special Issue Digital Innovations in Agriculture)

► Show Figures

Figure 1

16 pages, 3323 KiB

Open AccessArticle

Automatic Classification of Bagworm, Metisa plana (Walker) Instar Stages Using a Transfer Learning-Based Framework

by Siti Nurul Afiah Mohd Johari, Siti Khairunniza-Bejo, Abdul Rashid Mohamed Shariff, Nur Azuan Husin, Mohamed Mazmira Mohd Masri and Noorhazwani Kamarudin

Agriculture 2023, 13(2), 442; https://0-doi-org.brum.beds.ac.uk/10.3390/agriculture13020442 - 14 Feb 2023

Cited by 3 | Viewed by 1936

Abstract

Bagworms, particularly Metisa plana Walker (Lepidoptera: Psychidae), are one of the most destructive leaf-eating pests, especially in oil palm plantations, causing severe defoliation which reduces yield. Due to the delayed control of the bagworm population, it was discovered to be the most widespread [...] Read more.

Bagworms, particularly Metisa plana Walker (Lepidoptera: Psychidae), are one of the most destructive leaf-eating pests, especially in oil palm plantations, causing severe defoliation which reduces yield. Due to the delayed control of the bagworm population, it was discovered to be the most widespread oil palm pest in Peninsular Malaysia. Identification and classification of bagworm instar stages are critical for determining the current outbreak and taking appropriate control measures in the infested area. Therefore, this work proposes an automatic classification of bagworm larval instar stage starting from the second (S2) to the fifth (S5) instar stage using a transfer learning-based framework. Five different deep CNN architectures were used i.e., VGG16, ResNet50, ResNet152, DenseNet121 and DenseNet201 to categorize the larval instar stages. All the models were fine-tuned using two different optimizers, i.e., stochastic gradient descent (SGD) with momentum and adaptive moment estimation (Adam). Among the five models used, the DenseNet121 model, which used SGD with momentum (0.9) had the best classification accuracy of 96.18% with a testing time of 0.048 s per sample. Besides, all the instar stages from S2 to S5 can be identified with high value accuracy (94.52–97.57%), precision (89.71–95.87%), sensitivity (87.67–96.65%), specificity (96.51–98.61%) and the F1-score (88.89–96.18%). The presented transfer learning approach yields promising results, demonstrating its ability to classify bagworm instar stages. Full article

(This article belongs to the Special Issue Digital Innovations in Agriculture)

► Show Figures

Figure 1

16 pages, 9154 KiB

Open AccessArticle

Automatic Disease Detection of Basal Stem Rot Using Deep Learning and Hyperspectral Imaging

by Lai Zhi Yong, Siti Khairunniza-Bejo, Mahirah Jahari and Farrah Melissa Muharam

Agriculture 2023, 13(1), 69; https://0-doi-org.brum.beds.ac.uk/10.3390/agriculture13010069 - 26 Dec 2022

Cited by 14 | Viewed by 2327

Abstract

Basal Stem Rot (BSR), a disease caused by Ganoderma boninense (G. boninense), has posed a significant concern for the oil palm industry, particularly in Southeast Asia, as it has the potential to cause substantial economic losses. The breeding programme is currently [...] Read more.

Basal Stem Rot (BSR), a disease caused by Ganoderma boninense (G. boninense), has posed a significant concern for the oil palm industry, particularly in Southeast Asia, as it has the potential to cause substantial economic losses. The breeding programme is currently searching for G. boninense-resistant planting materials, which has necessitated intense manual screening in the nursery to track the progression of disease development in response to different treatments. The combination of hyperspectral image and machine learning approaches has a high detection potential for BSR. However, manual feature selection is still required to construct a detection model. Therefore, the objective of this study is to establish an automatic BSR detection at the seedling stage using a pre-trained deep learning model and hyperspectral images. The aerial view image of an oil palm seedling is divided into three regions in order to determine if there is any substantial spectral change across leaf positions. To investigate if the background images affect the performance of the detection, segmented images of the plant seedling have been automatically generated using a Mask Region-based Convolutional Neural Network (RCNN). Consequently, three models are utilised to detect BSR: a convolutional neural network that is 16 layers deep (VGG16) model trained on a segmented image; and VGG16 and Mask RCNN models both trained on the original images. The results indicate that the VGG16 model trained with the original images at 938 nm wavelength performed the best in terms of accuracy (91.93%), precision (94.32%), recall (89.26%), and F1 score (91.72%). This method revealed that users may detect BSR automatically without having to manually extract image attributes before detection. Full article

(This article belongs to the Special Issue Digital Innovations in Agriculture)

► Show Figures

Figure 1

21 pages, 3202 KiB

Open AccessArticle

Prediction of Protein Content in Pea (Pisum sativum L.) Seeds Using Artificial Neural Networks

by Patryk Hara, Magdalena Piekutowska and Gniewko Niedbała

Agriculture 2023, 13(1), 29; https://0-doi-org.brum.beds.ac.uk/10.3390/agriculture13010029 - 22 Dec 2022

Cited by 6 | Viewed by 2687

Abstract

Pea (Pisum sativum L.) is a legume valued mainly for its high seed protein content. The protein content of pea is characterized by a high lysine content and low allergenicity. This has made consumers appreciate peas increasingly in recent years, not only [...] Read more.

Pea (Pisum sativum L.) is a legume valued mainly for its high seed protein content. The protein content of pea is characterized by a high lysine content and low allergenicity. This has made consumers appreciate peas increasingly in recent years, not only for their taste, but also for their nutritional value. An important element of pea cultivation is the ability to predict protein content, even before harvest. The aim of this research was to develop a linear and a non-linear model for predicting the percentage of protein content in pea seeds and to perform a comparative analysis of the effectiveness of these models. The analysis also focused on identifying the variables with the greatest impact on protein content. The research included the method of machine learning (artificial neural networks) and multiple linear regression (MLR). The input parameters of the models were weather, agronomic and phytophenological data from 2016–2020. The predictive properties of the models were verified using six ex-post forecast measures. The neural model (N1) outperformed the multiple regression (RS) model. The N1 model had an RMS error magnitude of 0.838, while the RS model obtained an average error value of 2.696. The MAPE error for the N1 and RS models was 2.721 and 8.852, respectively. The sensitivity analysis performed for the best neural network showed that the independent variables most influencing the protein content of pea seeds were the soil abundance of magnesium, potassium and phosphorus. The results presented in this work can be useful for the study of pea crop management. In addition, they can help preserve the country’s protein security. Full article

(This article belongs to the Special Issue Digital Innovations in Agriculture)

► Show Figures

Figure 1

23 pages, 3057 KiB

Open AccessArticle

A Decision-Making Capability Optimization Scheme of Control Combination and PID Controller Parameters for Bivariate Fertilizer Applicator Improved by Using EDEM

by Yugong Dang, Gang Yang, Jun Wang, Zhigang Zhou and Zhidong Xu

Agriculture 2022, 12(12), 2100; https://0-doi-org.brum.beds.ac.uk/10.3390/agriculture12122100 - 08 Dec 2022

Cited by 8 | Viewed by 1239

Abstract

The fertilization rate is adjusted through the regulation of opening length and the rotational speed for bivariate fertilizer applicators. It is essential to optimally determine the control combination according to the target fertilization rate and further improve the control performance of fertilization operation [...] Read more.

The fertilization rate is adjusted through the regulation of opening length and the rotational speed for bivariate fertilizer applicators. It is essential to optimally determine the control combination according to the target fertilization rate and further improve the control performance of fertilization operation in precision agriculture. In this study, a novel decision-making capability optimization scheme of control combination and PID controller parameters is proposed to improve the feasibility and practicability of variable fertilizer applicators. Firstly, EDEM is adopted to acquire the minimum allowable opening length and the proper gap between the spiral blades and the discharge cavity wall, and then calibration experiments are implemented to establish the fitting model of fertilization rate using polynomial fitting. Secondly, the modified sparrow search algorithm (SSA) with chaotic operator and mutation section of the DE algorithm is used to optimize the control combination utilizing the accuracy, uniformity, and adjustment time as the evaluation criteria. Moreover, the tent mapping bat algorithm (TBA) is applied to tune the PID controller parameters for enhancing the accuracy and response speed of the fertilization-rate control system. Compared to the PID controller based on the bat algorithm (BA), traditional PID controller, and fuzzy PID controller, the rise time of the PID controller improved by TBA decreases by 0.018 s, 0.09 s, and 0.038 s, respectively, and the average steady-state deviation of that drops by 0.02 kg ha⁻¹, 1.45 kg ha⁻¹, and 0.19 kg ha⁻¹, respectively. In addition, under the condition of the same controller, compared with SSA, GA, and MOEA/D-DE, the average accuracy of the proposed decision-making algorithm decreases from 1.9%, 2.5%, and 3.5% to 1.8%, the average uniformity drops from 0.52% and 0.48% to 0.47%, and the average adjustment time declines from 0.99 s, 1.48 s, and 1.34 s to 0.5 s. It can be concluded that the method proposed in this study performs better in terms of accuracy and adjustment time but exhibits no apparent effect on the improvement of uniformity. Full article

(This article belongs to the Special Issue Digital Innovations in Agriculture)

► Show Figures

Figure 1

27 pages, 1044 KiB

Open AccessFeature PaperArticle

Prediction of Blueberry (Vaccinium corymbosum L.) Yield Based on Artificial Intelligence Methods

by Gniewko Niedbała, Jarosław Kurek, Bartosz Świderski, Tomasz Wojciechowski, Izabella Antoniuk and Krzysztof Bobran

Agriculture 2022, 12(12), 2089; https://0-doi-org.brum.beds.ac.uk/10.3390/agriculture12122089 - 06 Dec 2022

Cited by 10 | Viewed by 3083

Abstract

In this paper, we present a high-accuracy model for blueberry yield prediction, trained using structurally innovative data sets. Blueberries are blooming plants, valued for their antioxidant and anti-inflammatory properties. Yield on the plantations depends on several factors, both internal and external. Predicting the [...] Read more.

In this paper, we present a high-accuracy model for blueberry yield prediction, trained using structurally innovative data sets. Blueberries are blooming plants, valued for their antioxidant and anti-inflammatory properties. Yield on the plantations depends on several factors, both internal and external. Predicting the accurate amount of harvest is an important aspect in work planning and storage space selection. Machine learning algorithms are commonly used in such prediction tasks, since they are capable of finding correlations between various factors at play. Overall data were collected from years 2016–2021, and included agronomic, climatic and soil data as well satellite-imaging vegetation data. Additionally, growing periods according to BBCH scale and aggregates were taken into account. After extensive data preprocessing and obtaining cumulative features, a total of 11 models were trained and evaluated. Chosen classifiers were selected from state-of-the-art methods in similar applications. To evaluate the results, Mean Absolute Percentage Error was chosen. It is superior to alternatives, since it takes into account absolute values, negating the risk that opposite variables will cancel out, while the final result outlines percentage difference between the actual value and prediction. Regarding the research presented, the best performing solution proved to be Extreme Gradient Boosting algorithm, with MAPE value equal to 12.48%. This result meets the requirements of practical applications, with sufficient accuracy to improve the overall yield management process. Due to the nature of machine learning methodology, the presented solution can be further improved with annually collected data. Full article

(This article belongs to the Special Issue Digital Innovations in Agriculture)

► Show Figures

Figure 1

22 pages, 5880 KiB

Open AccessArticle

DFCANet: A Novel Lightweight Convolutional Neural Network Model for Corn Disease Identification

by Yang Chen, Xiaoyulong Chen, Jianwu Lin, Renyong Pan, Tengbao Cao, Jitong Cai, Dianzhi Yu, Tomislav Cernava and Xin Zhang

Agriculture 2022, 12(12), 2047; https://0-doi-org.brum.beds.ac.uk/10.3390/agriculture12122047 - 29 Nov 2022

Cited by 14 | Viewed by 2009

Abstract

The identification of corn leaf diseases in a real field environment faces several difficulties, such as complex background disturbances, variations and irregularities in the lesion areas, and large intra-class and small inter-class disparities. Traditional Convolutional Neural Network (CNN) models have a low recognition [...] Read more.

The identification of corn leaf diseases in a real field environment faces several difficulties, such as complex background disturbances, variations and irregularities in the lesion areas, and large intra-class and small inter-class disparities. Traditional Convolutional Neural Network (CNN) models have a low recognition accuracy and a large number of parameters. In this study, a lightweight corn disease identification model called DFCANet (Double Fusion block with Coordinate Attention Network) is proposed. The DFCANet consists mainly of two components: The dual feature fusion with coordinate attention and the Down-Sampling (DS) modules. The DFCA block contains dual feature fusion and Coordinate Attention (CA) modules. In order to completely fuse the shallow and deep features, these features were fused twice. The CA module suppresses the background noise and focuses on the diseased area. In addition, the DS module is used for down-sampling. It reduces the loss of information by expanding the feature channel dimension and the Depthwise convolution. The results show that DFCANet has an average recognition accuracy of 98.47%. It is more efficient at identifying corn leaf diseases in real scene images, compared with VGG16 (96.63%), ResNet50 (93.27%), EffcientNet-B0 (97.24%), ConvNeXt-B (94.18%), DenseNet121 (95.71%), MobileNet-V2 (95.41%), MobileNetv3-Large (96.33%), and ShuffleNetV2-1.0× (94.80%) methods. Moreover, the model’s Params and Flops are 1.91M and 309.1M, respectively, which are lower than heavyweight network models and most lightweight network models. In general, this study provides a novel, lightweight, and efficient convolutional neural network model for corn disease identification. Full article

(This article belongs to the Special Issue Digital Innovations in Agriculture)

► Show Figures

Figure 1

21 pages, 2935 KiB

Open AccessArticle

An Improved Intelligent Control System for Temperature and Humidity in a Pig House

by Hua Jin, Gang Meng, Yuanzhi Pan, Xing Zhang and Changda Wang

Agriculture 2022, 12(12), 1987; https://0-doi-org.brum.beds.ac.uk/10.3390/agriculture12121987 - 23 Nov 2022

Cited by 3 | Viewed by 1748

Abstract

The temperature and humidity control of a pig house is a complex multivariable control problem. How to keep the temperature and humidity in a pig house within a normal range is the problem to be solved in this paper. The traditional threshold-based environmental [...] Read more.

The temperature and humidity control of a pig house is a complex multivariable control problem. How to keep the temperature and humidity in a pig house within a normal range is the problem to be solved in this paper. The traditional threshold-based environmental control system cannot meet this requirement. In this paper, an intelligent control system of temperature and humidity in a pig house based on machine learning and a fuzzy control algorithm is proposed. We use sensors to collect the temperature and humidity in the pig house and store these data in chronological order. Then, we use these time series data to train the GRU model and then use the GRU model to predict the temperature and humidity change curve in the pig house in the next 24 hours. Finally, the mathematical model of the pig house and related equipment is established, and the output power of the related equipment is calculated based on the prediction results of GRU so as to effectively regulate the indoor temperature and humidity. The experimental results show that compared with the threshold-based environmental control system, our system reduces the abnormal temperature and humidity by about 90%. Full article

(This article belongs to the Special Issue Digital Innovations in Agriculture)

► Show Figures

Figure 1

13 pages, 883 KiB

Open AccessArticle

Associations of Automatically Recorded Body Condition Scores with Measures of Production, Health, and Reproduction

by Ramūnas Antanaitis, Dovilė Malašauskienė, Mindaugas Televičius, Mingaudas Urbutis, Arūnas Rutkauskas, Greta Šertvytytė, Lina Anskienė and Walter Baumgartner

Agriculture 2022, 12(11), 1834; https://0-doi-org.brum.beds.ac.uk/10.3390/agriculture12111834 - 02 Nov 2022

Cited by 3 | Viewed by 1395

Abstract

In the present study, we hypothesize that an automated body condition scoring system could be an indicator of health and pregnancy success in cows. Therefore, the objective of this study is to determine the relationship of the automated registered body condition score (BCS) [...] Read more.

In the present study, we hypothesize that an automated body condition scoring system could be an indicator of health and pregnancy success in cows. Therefore, the objective of this study is to determine the relationship of the automated registered body condition score (BCS) with pregnancy and inline biomarkers such as milk beta-hydroxybutyrate (BHB), milk lactate dehydrogenase (LDH), milk progesterone (mP4), and milk yield (MY) in dairy cows. Indicators from Herd NavigatorTM were grouped into classes based on their arithmetic means. Values were divided into various classes: MY: ≤31 kg/day (first class—67.3% of cows) and >31 kg/day (second class—32.7%); BHB in milk: ≤0.06 mmol/L (first class—80.7% of cows) and >0.06 mmol/L (second class—16.9%); milk LDH activity: ≤27 µmol/min (first class—69.5% of cows) and >27 µmol/min (second class—30.5%); milk progesterone value: ≤15.5 ng/mL (first class—28.8% of cows) and >15.5 ng/mL (second class—71.2%); and BCS: 2.5–3.0 (first class—21.4% of cows), >3.0–3.5 (second class—50.8%), and >3.5–4.0 (third class—27.8%). According to parity, the cows were divided into two groups: 1 lactation (first group—38.9%) and ≥2 lactations (second group—61.1%). Based on our investigated parameters, BCS is associated with pregnancy success because the BCS (+0.29 score) and mP4 (10.93 ng/mL) of the pregnant cows were higher compared to the group of non-pregnant cows. The MY (−5.26 kg, p < 0.001) and LDH (3.45 µmol/min) values were lower compared to those in the group of non-pregnant cows (p < 0.01). Statistically significant associations of BCS and mP4 with the number of inseminations were detected. The number of inseminations among cows with the highest BCS of >3.5–4.0 was 42.41% higher than that among cows with the lowest BCS of 2.5–3.0 (p < 0.001). BCS can also be a health indicator. We found that the LDH content was greatest among cows with the highest BCS of >3.5–4.0; this value was 6.48% higher than that in cows with a BCS of >3.0–3.5 (p < 0.01). The highest MY was detected in cows with the lowest BCS of 2.5–3.0, which was 29.55% higher than that in cows with the highest BCS of >3.5–4.0 (p < 0.001). BCS was the highest in the group of cows with mastitis (4.96% higher compared to the group of healthy cows), while the highest statistically significant mean differences in body condition score (9.04%) were estimated between the mastitis and metritis groups of cows (p < 0.001). Full article

(This article belongs to the Special Issue Digital Innovations in Agriculture)

► Show Figures

Figure 1

31 pages, 6866 KiB

Open AccessArticle

Modeling the Agricultural Soil Landscape of Germany—A Data Science Approach Involving Spatially Allocated Functional Soil Process Units

by Mareike Ließ

Agriculture 2022, 12(11), 1784; https://0-doi-org.brum.beds.ac.uk/10.3390/agriculture12111784 - 27 Oct 2022

Cited by 1 | Viewed by 2076

Abstract

The national-scale evaluation and modeling of the impact of agricultural management and climate change on soils, crop growth, and the environment require soil information at a spatial resolution addressing individual agricultural fields. This manuscript presents a data science approach that agglomerates the soil [...] Read more.

The national-scale evaluation and modeling of the impact of agricultural management and climate change on soils, crop growth, and the environment require soil information at a spatial resolution addressing individual agricultural fields. This manuscript presents a data science approach that agglomerates the soil parameter space into a limited number of functional soil process units (SPUs) that may be used to run agricultural process models. In fact, two unsupervised classification methods were developed to generate a multivariate 3D data product consisting of SPUs, each being defined by a multivariate parameter distribution along the depth profile from 0 to 100 cm. The two methods account for differences in variable types and distributions and involve genetic algorithm optimization to identify those SPUs with the lowest internal variability and maximum inter-unit difference with regards to both their soil characteristics and landscape setting. The high potential of the methods was demonstrated by applying them to the agricultural German soil landscape. The resulting data product consists of 20 SPUs. It has a 100 m raster resolution in the 2D mapping space, and its resolution along the depth profile is 1 cm. It includes the soil properties texture, stone content, bulk density, hydromorphic properties, total organic carbon content, and pH. Full article

(This article belongs to the Special Issue Digital Innovations in Agriculture)

► Show Figures

Graphical abstract

25 pages, 5357 KiB

Open AccessArticle

Research on the Optimization of Fresh Agricultural Products Trade Distribution Path Based on Genetic Algorithm

by Jun Sun, Tianhang Jiang, Yufei Song, Hao Guo and Yushi Zhang

Agriculture 2022, 12(10), 1669; https://0-doi-org.brum.beds.ac.uk/10.3390/agriculture12101669 - 11 Oct 2022

Cited by 3 | Viewed by 2996

Abstract

This study, taking the R fresh agricultural products distribution center (R-FAPDC) as an example, constructs a multi-objective optimization model of a logistics distribution path with time window constraints, and uses a genetic algorithm to optimize the optimal trade distribution path of fresh agricultural [...] Read more.

This study, taking the R fresh agricultural products distribution center (R-FAPDC) as an example, constructs a multi-objective optimization model of a logistics distribution path with time window constraints, and uses a genetic algorithm to optimize the optimal trade distribution path of fresh agricultural products. By combining the genetic algorithm with the actual case to explore, this study aims to solve enterprises’ narrow distribution paths and promote the model’s application in similar enterprises with similar characteristics. The results reveal that: (1) The trade distribution path scheme optimized by the genetic algorithm can reduce the distribution cost of distribution centers and improve customer satisfaction. (2) The genetic algorithm can bring economic benefits and reduce transportation losses in trade for trade distribution centers with the same spatial and quality characteristics as R fresh agricultural products distribution centers. According to our study, fresh agricultural products distribution enterprises should emphasize the use of genetic algorithms in planning distribution paths, develop a highly adaptable planning system of trade distribution routes, strengthen organizational and operational management, and establish a standard system for high-quality logistics services to improve distribution efficiency and customer satisfaction. Full article

(This article belongs to the Special Issue Digital Innovations in Agriculture)

► Show Figures

Figure 1

13 pages, 616 KiB

Open AccessCommunication

Fuzzy Quality Certification of Wheat

by Cristian Silviu Simionescu, Ciprian Petrisor Plenovici, Constanta Laura Augustin, Maria Magdalena Turek Rahoveanu, Adrian Turek Rahoveanu and Gheorghe Adrian Zugravu

Agriculture 2022, 12(10), 1640; https://0-doi-org.brum.beds.ac.uk/10.3390/agriculture12101640 - 08 Oct 2022

Cited by 2 | Viewed by 1311

Abstract

This paper presents a fuzzy quality certification of wheat. This analysis is based on the fuzzy analysis model of wheat. We developed a Matlab application with the help of which we modeled the perceptions in relation to the main quality physical and chemical [...] Read more.

This paper presents a fuzzy quality certification of wheat. This analysis is based on the fuzzy analysis model of wheat. We developed a Matlab application with the help of which we modeled the perceptions in relation to the main quality physical and chemical characteristics of wheat obtaining a quality index of wheat lots. The algorithm presented in this article allows for obtaining and using the global quality index, generating applicability not only to the commercial sphere as a quality reference and price setting, but also a measure of appreciation of processing opportunities. Indices of fuzzy quality associated with wheat lots using a fuzzy model offer the opportunity to develop local markets through quality certification. Full article

(This article belongs to the Special Issue Digital Innovations in Agriculture)

► Show Figures

Figure 1

17 pages, 5139 KiB

Open AccessArticle

Compact Convolutional Transformer (CCT)-Based Approach for Whitefly Attack Detection in Cotton Crops

by Aqeel Iftikhar Jajja, Assad Abbas, Hasan Ali Khattak, Gniewko Niedbała, Abbas Khalid, Hafiz Tayyab Rauf and Sebastian Kujawa

Agriculture 2022, 12(10), 1529; https://0-doi-org.brum.beds.ac.uk/10.3390/agriculture12101529 - 23 Sep 2022

Cited by 14 | Viewed by 2768

Abstract

Cotton is one of the world’s most economically significant agricultural products; however, it is susceptible to numerous pest and virus attacks during the growing season. Pests (whitefly) can significantly affect a cotton crop, but timely disease detection can help pest control. Deep learning [...] Read more.

Cotton is one of the world’s most economically significant agricultural products; however, it is susceptible to numerous pest and virus attacks during the growing season. Pests (whitefly) can significantly affect a cotton crop, but timely disease detection can help pest control. Deep learning models are best suited for plant disease classification. However, data scarcity remains a critical bottleneck for rapidly growing computer vision applications. Several deep learning models have demonstrated remarkable results in disease classification. However, these models have been trained on small datasets that are not reliable due to model generalization issues. In this study, we first developed a dataset on whitefly attacked leaves containing 5135 images that are divided into two main classes, namely, (i) healthy and (ii) unhealthy. Subsequently, we proposed a Compact Convolutional Transformer (CCT)-based approach to classify the image dataset. Experimental results demonstrate the proposed CCT-based approach’s effectiveness compared to the state-of-the-art approaches. Our proposed model achieved an accuracy of 97.2%, whereas Mobile Net, ResNet152v2, and VGG-16 achieved accuracies of 95%, 92%, and 90%, respectively. Full article

(This article belongs to the Special Issue Digital Innovations in Agriculture)

► Show Figures

Figure 1

12 pages, 2580 KiB

Open AccessArticle

Hyperspectral Imaging-Based Multiple Predicting Models for Functional Component Contents in Brassica juncea

by Jae-Hyeong Choi, Soo Hyun Park, Dae-Hyun Jung, Yun Ji Park, Jung-Seok Yang, Jai-Eok Park, Hyein Lee and Sang Min Kim

Agriculture 2022, 12(10), 1515; https://0-doi-org.brum.beds.ac.uk/10.3390/agriculture12101515 - 21 Sep 2022

Cited by 4 | Viewed by 1632

Abstract

Partial least squares regression (PLSR) prediction models were developed using hyperspectral imaging for noninvasive detection of the five most representative functional components in Brassica juncea leaves: chlorophyll, carotenoid, phenolic, glucosinolate, and anthocyanin contents. The region of interest for functional component analysis was chosen [...] Read more.

Partial least squares regression (PLSR) prediction models were developed using hyperspectral imaging for noninvasive detection of the five most representative functional components in Brassica juncea leaves: chlorophyll, carotenoid, phenolic, glucosinolate, and anthocyanin contents. The region of interest for functional component analysis was chosen by polygon selection and the extracted average spectra were used for model development. For pre-processing, 10 combinations of Savitzky–Golay filter (S. G. filter), standard normal variate (SNV), multiplicative scatter correction (MSC), 1st-order derivative (1st-Der), 2nd-order derivative (2nd-Der), and normalization were applied. Root mean square errors of calibration (RMSEP) was used to assess the performance accuracy of the constructed prediction models. The prediction model for total anthocyanins exhibited the highest prediction level (R_V² = 0.8273; RMSEP = 2.4277). Pre-processing combination of SNV and 1st-Der with spectral data resulted in high-performance prediction models for total chlorophyll, carotenoid, and glucosinolate contents. Pre-processing combination of S. G. filter and SNV gave the highest prediction rate for total phenolics. SNV inclusion in the pre-processing conditions was essential for developing high-performance accurate prediction models for functional components. By enabling visualization of the distribution of functional components on the hyperspectral images, PLSR prediction models will prove valuable in determining the harvest time. Full article

(This article belongs to the Special Issue Digital Innovations in Agriculture)

► Show Figures

Figure 1

23 pages, 3065 KiB

Open AccessArticle

An Improved Multi-Objective Optimization Decision Method Using NSGA-III for a Bivariate Precision Fertilizer Applicator

by Yugong Dang, Hongen Ma, Jun Wang, Zhigang Zhou and Zhidong Xu

Agriculture 2022, 12(9), 1492; https://0-doi-org.brum.beds.ac.uk/10.3390/agriculture12091492 - 17 Sep 2022

Cited by 3 | Viewed by 1744

Abstract

In order to boost the performance of a bivariable granular fertilizer applicator and simplify the control methodology of fertilization rate regulation, this paper proposed a fertilization decision method to obtain the optimal combination of rotational speed and opening length by selecting the accuracy, [...] Read more.

In order to boost the performance of a bivariable granular fertilizer applicator and simplify the control methodology of fertilization rate regulation, this paper proposed a fertilization decision method to obtain the optimal combination of rotational speed and opening length by selecting the accuracy, uniformity, adjustment time, and breakage rate as the optimization objectives. We processed the outlier data collected using the indoor bench test, segmented the data with the fertilization growth rate as the index, and proved the rationality of the data segmentation by an independent sample t-test. SVM, BPNN, ELM, and RVM were used to train the two data sections to create the fertilization rate prediction model, and the models with the highest accuracy in the two data sections were selected for the assembly of the final prediction model used to describe the fertilization process of the bivariate fertilizer applicator. Moreover, the fertilization performance problem model was established with the objectives of accuracy, uniformity, adjustment time, and breakage rate and was solved using the NSGA-III algorithm to gain an optimal fertilization decision. Compared with GA and MOEA-D-DE methods, the results show that, using the new method, the average relative error declines from 8.64% and 6.05% to 3.09%, and the average coefficient of variation reduces from 6.67% and 6.81% to 6.41%, respectively. In addition, the adjustment time lowers from 2.01 s and 1.33 s to 0.78 s, and the average breakage rate drops from 1.084% and 0.845% to 0.803%, respectively. It is indicated that the presented method offers the most notable improvements in accuracy and adjustment time, while the advancements in regard to uniformity and breakage rate is slight, but both are within a reasonable range. Full article

(This article belongs to the Special Issue Digital Innovations in Agriculture)

► Show Figures

Figure 1

28 pages, 5448 KiB

Open AccessArticle

Towards a Real-Time Oil Palm Fruit Maturity System Using Supervised Classifiers Based on Feature Analysis

by Meftah Salem M. Alfatni, Siti Khairunniza-Bejo, Mohammad Hamiruce B. Marhaban, Osama M. Ben Saaed, Aouache Mustapha and Abdul Rashid Mohamed Shariff

Agriculture 2022, 12(9), 1461; https://0-doi-org.brum.beds.ac.uk/10.3390/agriculture12091461 - 14 Sep 2022

Cited by 8 | Viewed by 3109

Abstract

Remote sensing sensors-based image processing techniques have been widely applied in non-destructive quality inspection systems of agricultural crops. Image processing and analysis were performed with computer vision and external grading systems by general and standard steps, such as image acquisition, pre-processing and segmentation, [...] Read more.

Remote sensing sensors-based image processing techniques have been widely applied in non-destructive quality inspection systems of agricultural crops. Image processing and analysis were performed with computer vision and external grading systems by general and standard steps, such as image acquisition, pre-processing and segmentation, extraction and classification of image characteristics. This paper describes the design and implementation of a real-time fresh fruit bunch (FFB) maturity classification system for palm oil based on unrestricted remote sensing (CCD camera sensor) and image processing techniques using five multivariate techniques (statistics, histograms, Gabor wavelets, GLCM and BGLAM) to extract fruit image characteristics and incorporate information on palm oil species classification FFB and maturity testing. To optimize the proposed solution in terms of performance reporting and processing time, supervised classifiers, such as support vector machine (SVM), K-nearest neighbor (KNN) and artificial neural network (ANN), were performed and evaluated via ROC and AUC measurements. The experimental results showed that the FFB classification system of non-destructive palm oil maturation in real time provided a significant result. Although the SVM classifier is generally a robust classifier, ANN has better performance due to the natural noise of the data. The highest precision was obtained on the basis of the ANN and BGLAM algorithms applied to the texture of the fruit. In particular, the robust image processing algorithm based on BGLAM feature extraction technology and the ANN classifier largely provided a high AUC test accuracy of over 93% and an image-processing time of 0,44 (s) for the detection of FFB palm oil species. Full article

(This article belongs to the Special Issue Digital Innovations in Agriculture)

► Show Figures

Figure 1

20 pages, 5388 KiB

Open AccessArticle

Application of a Fractional Order Differential to the Hyperspectral Inversion of Soil Iron Oxide

by Hailong Zhao, Shu Gan, Xiping Yuan, Lin Hu, Junjie Wang and Shuai Liu

Agriculture 2022, 12(8), 1163; https://0-doi-org.brum.beds.ac.uk/10.3390/agriculture12081163 - 05 Aug 2022

Cited by 8 | Viewed by 1667

Abstract

Iron oxide is the main form of iron present in soils, and its accumulation and migration activities reflect the leaching process and the degree of weathering development of the soil. Therefore, it is important to have information on the iron oxide content of [...] Read more.

Iron oxide is the main form of iron present in soils, and its accumulation and migration activities reflect the leaching process and the degree of weathering development of the soil. Therefore, it is important to have information on the iron oxide content of soils. However, due to the overlapping characteristic spectra of iron oxide and organic matter in the visible-near infrared, appropriate spectral transformation methods are important. In this paper, we first used conventional spectral transformation (continuum removal, CR; standard normal variate, SNV; absorbance, log (1/R)), continuous wavelet transform (CWT), and fractional order differential (FOD) transform to process original spectra (OS). Secondly, competitive adaptive reweighted sampling (CARS) was used to extract characteristic wavelengths. Finally, two regression models (backpropagation neural network, BPNN; support vector regression (SVR) were used to predict the content of iron oxide. The results show that the FOD can significantly improve the correlation with iron oxide compared with the CR, SNV, log (1/R) and CWT; the baseline drift and overlapping peaks decrease with increasing the order of FOD; the CARS algorithm based on 50th averaging can select more stable characteristic wavelengths; the FOD achieves better results regardless of the modelling method, and the model based on 0.5-order differential has the best prediction performance (R² = 0.851, RMSE = 5.497, RPIQ = 3.686). Full article

(This article belongs to the Special Issue Digital Innovations in Agriculture)

► Show Figures

Figure 1

16 pages, 10486 KiB

Open AccessFeature PaperEditor’s ChoiceArticle

Tea Category Identification Using Wavelet Signal Reconstruction of Hyperspectral Imagery and Machine Learning

by Qiang Cui, Baohua Yang, Biyun Liu, Yunlong Li and Jingming Ning

Agriculture 2022, 12(8), 1085; https://0-doi-org.brum.beds.ac.uk/10.3390/agriculture12081085 - 23 Jul 2022

Cited by 16 | Viewed by 2354

Abstract

Accurately distinguishing the types of tea is of great significance to the pricing, production, and processing of tea. The similarity of the internal spectral characteristics and appearance characteristics of different types of tea greatly limits further research on tea identification. However, wavelet transform [...] Read more.

Accurately distinguishing the types of tea is of great significance to the pricing, production, and processing of tea. The similarity of the internal spectral characteristics and appearance characteristics of different types of tea greatly limits further research on tea identification. However, wavelet transform can simultaneously extract time domain and frequency domain features, which is a powerful tool in the field of image signal processing. To address this gap, a method for tea recognition based on a lightweight convolutional neural network and support vector machine (L-CNN-SVM) was proposed, aiming to realize tea recognition using wavelet feature figures generated by wavelet time-frequency signal decomposition and reconstruction. Firstly, the redundant discrete wavelet transform was used to decompose the wavelet components of the hyperspectral images of the three teas (black tea, green tea, and yellow tea), which were used to construct the datasets. Secondly, improve the lightweight CNN model to generate a tea recognition model. Finally, compare and evaluate the recognition results of different models. The results demonstrated that the results of tea recognition based on the L-CNN-SVM method outperformed MobileNet v2+RF, MobileNet v2+KNN, MobileNet v2+AdaBoost, AlexNet, and MobileNet v2. For the recognition results of the three teas using reconstruction of wavelet components LL + HL + LH, the overall accuracy rate reached 98.7%, which was 4.7%, 3.4%, 1.4%, and 2.0% higher than that of LH + HL + HH, LL + HH + HH, LL + LL + HH, and LL + LL + LL. This research can provide new inspiration and technical support for grade and quality assessment of cross-category tea. Full article

(This article belongs to the Special Issue Digital Innovations in Agriculture)

► Show Figures

Figure 1

14 pages, 4525 KiB

Open AccessEditor’s ChoiceArticle

Lightweight Detection Algorithm of Kiwifruit Based on Improved YOLOX-S

by Junchi Zhou, Wenwu Hu, Airu Zou, Shike Zhai, Tianyu Liu, Wenhan Yang and Ping Jiang

Agriculture 2022, 12(7), 993; https://0-doi-org.brum.beds.ac.uk/10.3390/agriculture12070993 - 09 Jul 2022

Cited by 18 | Viewed by 2787

Abstract

Considering the high requirements of current kiwifruit picking recognition systems for mobile devices, including the small number of available features for image targets and small-scale aggregation, an enhanced YOLOX-S target detection algorithm for kiwifruit picking robots is proposed in this study. This involved [...] Read more.

Considering the high requirements of current kiwifruit picking recognition systems for mobile devices, including the small number of available features for image targets and small-scale aggregation, an enhanced YOLOX-S target detection algorithm for kiwifruit picking robots is proposed in this study. This involved designing a new multi-scale feature integration structure in which, with the aim of providing a small and lightweight model, the feature maps used for detecting large targets in the YOLOX model are eliminated, the feature map of small targets is sampled through the nearest neighbor values, the superficial features are spliced with the final features, the gradient of the SiLU activation function is perturbed, and the loss function at the output is optimized. The experimental results show that, compared with the original YOLOX-S, the enhanced model improved the detection average precision (AP) of kiwifruit images by 6.52%, reduced the number of model parameters by 44.8%, and improved the model detection speed by 63.9%. Hence, with its outstanding effectiveness and relatively light weight, the proposed model is capable of effectively providing data support for the 3D positioning and automated picking of kiwifruit. It may also successfully provide solutions in similar fields related to small target detection. Full article

(This article belongs to the Special Issue Digital Innovations in Agriculture)

► Show Figures

Figure 1

22 pages, 1405 KiB

Open AccessArticle

Performance Evaluation of Communication Systems Used for Internet of Things in Agriculture

by Geovanny Yascaribay, Mónica Huerta, Miguel Silva and Roger Clotet

Agriculture 2022, 12(6), 786; https://0-doi-org.brum.beds.ac.uk/10.3390/agriculture12060786 - 30 May 2022

Cited by 7 | Viewed by 3075

Abstract

The rapid development of Internet of Things (IoT) technology has provided ample opportunity for the implementation of intelligent agricultural production. Such technology can be used to connect various types of agricultural devices, which can collect and send data to servers for analysis. These [...] Read more.

The rapid development of Internet of Things (IoT) technology has provided ample opportunity for the implementation of intelligent agricultural production. Such technology can be used to connect various types of agricultural devices, which can collect and send data to servers for analysis. These tools can help farmers optimize the production of their crops. However, one of the main problems that arises in agricultural areas is a lack of connectivity or poor connection quality. For these reasons, in this paper, we present a method that can be used for the performance evaluation of communication systems used in IoT for agriculture, considering metrics such as the packet delivery ratio, energy consumption, and packet collisions. To achieve this aim, we carry out an analysis of the main Low-Power Wide-Area Networks (LPWAN) protocols and their applicability, from which we conclude that those most suited to this context are Long Range (LoRa) and Long Range Wide Area Network (LoRaWAN). After that, we analyze various simulation tools and select Omnet++ together with the Framework for LoRa (FLoRa) library as the best option. In the first stage of the simulations, the performances of LoRa and LoRaWAN are evaluated by comparing the average propagation under ideal conditions against moderate propagation losses, emulating a rural environment in the coastal region of Ecuador. In the second phase, metrics such as the package delivery ratio and energy consumption are evaluated by simulating communication between an increasing number of nodes and one or two gateways. The results show that using two gateways with the Adaptive Data Rate technique can actively increase the delivery ratio of the network while consuming the same amount of energy per node. Finally, a comparison is made between the results of the simulation scenario considered in this project and those of other research works, allowing for the validation of our analytical and simulation results. Full article

(This article belongs to the Special Issue Digital Innovations in Agriculture)

► Show Figures

Figure 1

17 pages, 4448 KiB

Open AccessArticle

Application of Artificial Neural Network Sensitivity Analysis to Identify Key Determinants of Harvesting Date and Yield of Soybean (Glycine max [L.] Merrill) Cultivar Augusta

by Gniewko Niedbała, Danuta Kurasiak-Popowska, Magdalena Piekutowska, Tomasz Wojciechowski, Michał Kwiatek and Jerzy Nawracała

Agriculture 2022, 12(6), 754; https://0-doi-org.brum.beds.ac.uk/10.3390/agriculture12060754 - 25 May 2022

Cited by 9 | Viewed by 2009

Abstract

Genotype and weather conditions play crucial roles in determining the volume and stability of a soybean yield. The aim of this study was to identify the key meteorological factors affecting the harvest date (model M_HARV) and yield of the soybean variety Augusta (model [...] Read more.

Genotype and weather conditions play crucial roles in determining the volume and stability of a soybean yield. The aim of this study was to identify the key meteorological factors affecting the harvest date (model M_HARV) and yield of the soybean variety Augusta (model M_YIELD) using a neural network sensitivity analysis. The dates of the start of flowering and maturity, the yield data, the average daily temperatures and precipitation were collected, and the Selyaninov hydrothermal coefficients were calculated during a fifteen-year study (2005–2020 growing seasons). During the experiment, highly variable weather conditions occurred, strongly modifying the course of phenological phases in soybean and the achieved seed yield of Augusta cultivar. The harvesting of mature soybean seeds took place between 131 and 156 days after sowing, while the harvested yield ranged from 0.6 t·ha⁻¹ to 2.6 t·ha⁻¹. The sensitivity analysis of the MLP neural network made it possible to identify the factors which had the greatest impact on the tested dependent variables among all the analyzed factors. It was revealed that the variables assigned ranks 1 and 2 in the sensitivity analysis of the neural network forming the M_HARV model were total rainfall in the first decade of June and the first decade of August. The variables with the highest impact on the Augusta soybean seed yield (model M_YIELD) were the mean daily air temperature in the second decade of May and the Seljaninov coefficient values calculated for the sowing–flowering date period. Full article

(This article belongs to the Special Issue Digital Innovations in Agriculture)

► Show Figures

Figure 1

16 pages, 6165 KiB

Open AccessArticle

LA-DeepLab V3+: A Novel Counting Network for Pigs

by Chengqi Liu, Jie Su, Longhe Wang, Shuhan Lu and Lin Li

Agriculture 2022, 12(2), 284; https://0-doi-org.brum.beds.ac.uk/10.3390/agriculture12020284 - 17 Feb 2022

Cited by 11 | Viewed by 2405

Abstract

Accurate identification and intelligent counting of pig herds can effectively improve the level of fine management of pig farms. A semantic segmentation and counting network was proposed in this study to improve the segmentation accuracy and counting efficiency of pigs in complex image [...] Read more.

Accurate identification and intelligent counting of pig herds can effectively improve the level of fine management of pig farms. A semantic segmentation and counting network was proposed in this study to improve the segmentation accuracy and counting efficiency of pigs in complex image segmentation. In this study, we built our own datasets of pigs under different scenarios, and set three levels of number detection difficulty—namely, lightweight, middleweight, and heavyweight. First, an image segmentation model of a small sample of pigs was established based on the DeepLab V3+ deep learning method to reduce the training cost and obtain initial features. Second, a lightweight attention mechanism was introduced, and attention modules based on rows and columns can accelerate the efficiency of feature calculation and reduce the problem of excessive parameters and feature redundancy caused by network depth. Third, a recursive cascade method was used to optimize the fusion of high- and low-frequency features for mining potential semantic information. Finally, the improved model was integrated to build a graphical platform for the accurate counting of pigs. Compared with FCNNs, U-Net, SegNet, and DenseNet methods, the DeepLab V3+ experimental results show that the values of the comprehensive evaluation indices P, R, AP, F₁-score, and MIoU of LA-DeepLab V3+ (single tag) are higher than those of other semantic segmentation models, at 86.04%, 75.06%, 78.67%, 0.8, and 76.31%, respectively. The P, AP, and MIoU values of LA-DeepLab V3+ (multiple tags) are also higher than those of other models, at 88.36%, 76.75%, and 74.62%, respectively. The segmentation accuracy of pig images with simple backgrounds reaches 99%. The pressure test of the counting network can calculate the number of pigs with a maximum of 50, which meets the requirements of free-range breeding in standard piggeries. The model has strong generalization ability in pig herd detection under different scenarios, which can serve as a reference for intelligent pig farm management and animal life research. Full article

(This article belongs to the Special Issue Digital Innovations in Agriculture)

► Show Figures

Figure 1

13 pages, 990 KiB

Open AccessEditor’s ChoiceArticle

Modelling of Mechanical Properties of Fresh and Stored Fruit of Large Cranberry Using Multiple Linear Regression and Machine Learning

by Józef Gorzelany, Justyna Belcar, Piotr Kuźniar, Gniewko Niedbała and Katarzyna Pentoś

Agriculture 2022, 12(2), 200; https://0-doi-org.brum.beds.ac.uk/10.3390/agriculture12020200 - 31 Jan 2022

Cited by 18 | Viewed by 2921

Abstract

The study investigated the selected mechanical properties of fresh and stored large cranberries. The analyses focused on changes in the energy requirement up to the breaking point and aimed to identify the apparent elasticity index of the fruit of the investigated large cranberry [...] Read more.

The study investigated the selected mechanical properties of fresh and stored large cranberries. The analyses focused on changes in the energy requirement up to the breaking point and aimed to identify the apparent elasticity index of the fruit of the investigated large cranberry fruit varieties relating to harvest time, water content, as well as storage duration and conditions. After 25 days in storage, the fruit of the investigated varieties were found with a decrease in mean acidity, from 1.56 g⋅100 g⁻¹ to 1.42 g⋅100 g⁻¹, and mean water content, from 89.71% to 87.95%. The findings showed a decrease in breaking energy; there was also a change in the apparent modulus of elasticity, its mean value in the fresh fruit was 0.431 ± 0.07 MPa, and after 25 days of storage it decreased to 0.271 ± 0.08 MPa. The relationships between the cranberry varieties, storage temperature, duration of storage, x, y, and z dimensions of the fruits, and their selected mechanical parameters were modeled with the use of multiple linear regression, artificial neural networks, and support vector machines. Machine learning techniques outperformed multiple linear regression. Full article

(This article belongs to the Special Issue Digital Innovations in Agriculture)

► Show Figures

Figure 1

23 pages, 8079 KiB

Open AccessArticle

Multi-Temporal Data Fusion in MS and SAR Images Using the Dynamic Time Warping Method for Paddy Rice Classification

by Tsu Chiang Lei, Shiuan Wan, You Cheng Wu, Hsin-Ping Wang and Chia-Wen Hsieh

Agriculture 2022, 12(1), 77; https://0-doi-org.brum.beds.ac.uk/10.3390/agriculture12010077 - 07 Jan 2022

Cited by 6 | Viewed by 1928

Abstract

This study employed a data fusion method to extract the high-similarity time series feature index of a dataset through the integration of MS (Multi-Spectrum) and SAR (Synthetic Aperture Radar) images. The farmlands are divided into small pieces that consider the different behaviors of [...] Read more.

This study employed a data fusion method to extract the high-similarity time series feature index of a dataset through the integration of MS (Multi-Spectrum) and SAR (Synthetic Aperture Radar) images. The farmlands are divided into small pieces that consider the different behaviors of farmers for their planting contents in Taiwan. Hence, the conventional image classification process cannot produce good outcomes. The crop phenological information will be a core factor to multi-period image data. Accordingly, the study intends to resolve the previous problem by using three different SPOT6 satellite images and nine Sentinel-1A synthetic aperture radar images, which were used to calculate features such as texture and indicator information, in 2019. Considering that a Dynamic Time Warping (DTW) index (i) can integrate different image data sources, (ii) can integrate data of different lengths, and (iii) can generate information with time characteristics, this type of index can resolve certain classification problems with long-term crop classification and monitoring. More specifically, this study used the time series data analysis of DTW to produce “multi-scale time series feature similarity indicators”. We used three approaches (Support Vector Machine, Neural Network, and Decision Tree) to classify paddy patches into two groups: (a) the first group did not apply a DTW index, and (b) the second group extracted conflict predicted data from (a) to apply a DTW index. The outcomes from the second group performed better than the first group in regard to overall accuracy (OA) and kappa. Among those classifiers, the Neural Network approach had the largest improvement of OA and kappa from 89.51, 0.66 to 92.63, 0.74, respectively. The rest of the two classifiers also showed progress. The best performance of classification results was obtained from the Decision Tree of 94.71, 0.81. Observing the outcomes, the interference effects of the image were resolved successfully by various image problems using the spectral image and radar image for paddy rice classification. The overall accuracy and kappa showed improvement, and the maximum kappa was enhanced by about 8%. The classification performance was improved by considering the DTW index. Full article

(This article belongs to the Special Issue Digital Innovations in Agriculture)

► Show Figures

Figure 1

19 pages, 46219 KiB

Open AccessArticle

Detection and Analysis of Sow Targets Based on Image Vision

by Kaidong Lei, Chao Zong, Ting Yang, Shanshan Peng, Pengfei Zhu, Hao Wang, Guanghui Teng and Xiaodong Du

Agriculture 2022, 12(1), 73; https://0-doi-org.brum.beds.ac.uk/10.3390/agriculture12010073 - 06 Jan 2022

Cited by 8 | Viewed by 2504

Abstract

In large-scale sow production, real-time detection and recognition of sows is a key step towards the application of precision livestock farming techniques. In the pig house, the overlap of railings, floors, and sows usually challenge the accuracy of sow target detection. In this [...] Read more.

In large-scale sow production, real-time detection and recognition of sows is a key step towards the application of precision livestock farming techniques. In the pig house, the overlap of railings, floors, and sows usually challenge the accuracy of sow target detection. In this paper, a non-contact machine vision method was used for sow targets perception in complex scenarios, and the number position of sows in the pen could be detected. Two multi-target sow detection and recognition models based on the deep learning algorithms of Mask-RCNN and UNet-Attention were developed, and the model parameters were tuned. A field experiment was carried out. The data-set obtained from the experiment was used for algorithm training and validation. It was found that the Mask-RCNN model showed a higher recognition rate than that of the UNet-Attention model, with a final recognition rate of 96.8% and complete object detection outlines. In the process of image segmentation, the area distribution of sows in the pens was analyzed. The position of the sow’s head in the pen and the pixel area value of the sow segmentation were analyzed. The feeding, drinking, and lying behaviors of the sow have been identified on the basis of image recognition. The results showed that the average daily lying time, standing time, feeding and drinking time of sows were 12.67 h(MSE 1.08), 11.33 h(MSE 1.08), 3.25 h(MSE 0.27) and 0.391 h(MSE 0.10), respectively. The proposed method in this paper could solve the problem of target perception of sows in complex scenes and would be a powerful tool for the recognition of sows. Full article

(This article belongs to the Special Issue Digital Innovations in Agriculture)

► Show Figures

Figure 1

18 pages, 7515 KiB

Open AccessArticle

A Four Stage Image Processing Algorithm for Detecting and Counting of Bagworm, Metisa plana Walker (Lepidoptera: Psychidae)

by Mohd Najib Ahmad, Abdul Rashid Mohamed Shariff, Ishak Aris and Izhal Abdul Halin

Agriculture 2021, 11(12), 1265; https://0-doi-org.brum.beds.ac.uk/10.3390/agriculture11121265 - 14 Dec 2021

Cited by 2 | Viewed by 2669

Abstract

The bagworm is a vicious leaf eating insect pest that threatens the oil palm plantations in Malaysia. The economic impact from defoliation of approximately 10% to 13% due to bagworm attack might cause about 33% to 40% yield loss over 2 years. Due [...] Read more.

The bagworm is a vicious leaf eating insect pest that threatens the oil palm plantations in Malaysia. The economic impact from defoliation of approximately 10% to 13% due to bagworm attack might cause about 33% to 40% yield loss over 2 years. Due to this, monitoring and detecting of bagworm populations in oil palm plantations is required as the preliminary steps to ensure proper planning of control actions in these areas. Hence, the development of an image processing algorithm for detection and counting of Metisa plana Walker, a species of Malaysia’s local bagworm, using image segmentation has been researched and completed. The color and shape features from the segmented images for real time object detection showed an average detection accuracy of 40% and 34%, at 30 cm and 50 cm camera distance, respectively. After some improvements on training dataset and marking detected bagworm with bounding box, a deep learning algorithm with Faster Regional Convolutional Neural Network (Faster R-CNN) algorithm was applied leading to the percentage of the detection accuracy increased up to 100% at a camera distance of 30 cm in close conditions. The proposed solution is also designed to distinguish between the living and dead larvae of the bagworms using motion detection which resulted in approximately 73–100% accuracy at a camera distance of 30 cm in the close conditions. Through false color analysis, distinct differences in the pixel count based on the slope was observed for dead and live pupae at 630 nm and 940 nm, with the slopes recorded at 0.38 and 0.28, respectively. The higher pixel count and slope correlated with the dead pupae while the lower pixel count and slope, represented the living pupae. Full article

(This article belongs to the Special Issue Digital Innovations in Agriculture)

► Show Figures

Figure 1

17 pages, 2366 KiB

Open AccessEditor’s ChoiceArticle

Modeling the Essential Oil and Trans-Anethole Yield of Fennel (Foeniculum vulgare Mill. var. vulgare) by Application Artificial Neural Network and Multiple Linear Regression Methods

by Mohsen Sabzi-Nojadeh, Gniewko Niedbała, Mehdi Younessi-Hamzekhanlu, Saeid Aharizad, Mohammad Esmaeilpour, Moslem Abdipour, Sebastian Kujawa and Mohsen Niazian

Agriculture 2021, 11(12), 1191; https://0-doi-org.brum.beds.ac.uk/10.3390/agriculture11121191 - 26 Nov 2021

Cited by 12 | Viewed by 2433

Abstract

Foeniculum vulgare Mill. (commonly known as fennel) is used in the pharmaceutical, cosmetic, and food industries. Fennel widely used as a digestive, carminative, galactagogue and diuretic and in treating gastrointestinal and respiratory disorders. Improving low heritability traits such as essential oil yield (EOY%) [...] Read more.

Foeniculum vulgare Mill. (commonly known as fennel) is used in the pharmaceutical, cosmetic, and food industries. Fennel widely used as a digestive, carminative, galactagogue and diuretic and in treating gastrointestinal and respiratory disorders. Improving low heritability traits such as essential oil yield (EOY%) and trans-anethole yield (TAY%) of fennel by direct selection does not result in rapid gains of EOY% and TAY%. Identification of high-heritable traits and using efficient modeling methods can be a beneficial approach to overcome this limitation and help breeders select the most advantageous traits in medicinal plant breeding programs. The present study aims to compare the performance of the artificial neural network (ANN) and multilinear regression (MLR) to predict the EOY% and TAY% of fennel populations. Stepwise regression (SWR) was used to assess the effect of various input variables. Based on SWR, nine traits—number of days to 50% flowering (NDF50%), number of days to maturity (NDM), final plant height (FPH), number of internodes (NI), number of umbels (NU), seed yield per square meter (SY/m²), number of seeds per plant (NS/P), number of seeds per umbel (NS/U) and 1000-seed weight (TSW)—were chosen as input variables. The network with Sigmoid Axon transfer function and two hidden layers was selected as the final ANN model for the prediction of EOY%, and the TanhAxon function with one hidden layer was used for the prediction of TAY%. The results revealed that the ANN method could predict the EOY% and TAY% with more accuracy and efficiency (R² of EOY% = 0.929, R² of TAY% = 0.777, RMSE of EOY% = 0.544, RMSE of TAY% = 0.264, MAE of EOY% = 0.385 and MAE of TAY% = 0.352) compared with the MLR model (R² of EOY% = 0.553, R² of TAY% = 0.467, RMSE of EOY% = 0.819, RMSE of TAY% = 0.448, MAE of EOY% = 0.624 and MAE of TAY% = 0.452). Based on the sensitivity analysis, SY/m², NDF50% and NS/P were the most important traits to predict EOY% as well as SY/m², NS/U and NDM to predict of TAY%. The results demonstrate the potential of ANNs as a promising tool to predict the EOY% and TAY% of fennel, and they can be used in future fennel breeding programs. Full article

(This article belongs to the Special Issue Digital Innovations in Agriculture)

► Show Figures

Figure 1

17 pages, 5488 KiB

Open AccessEditor’s ChoiceArticle

Estimation of Soil Nutrient Content Using Hyperspectral Data

by Yiping Peng, Lu Wang, Li Zhao, Zhenhua Liu, Chenjie Lin, Yueming Hu and Luo Liu

Agriculture 2021, 11(11), 1129; https://0-doi-org.brum.beds.ac.uk/10.3390/agriculture11111129 - 11 Nov 2021

Cited by 17 | Viewed by 3798

Abstract

Soil nutrients play a vital role in plant growth and thus the rapid acquisition of soil nutrient content is of great significance for agricultural sustainable development. Hyperspectral remote-sensing techniques allow for the quick monitoring of soil nutrients. However, at present, obtaining accurate estimates [...] Read more.

Soil nutrients play a vital role in plant growth and thus the rapid acquisition of soil nutrient content is of great significance for agricultural sustainable development. Hyperspectral remote-sensing techniques allow for the quick monitoring of soil nutrients. However, at present, obtaining accurate estimates proves to be difficult due to the weak spectral features of soil nutrients and the low accuracy of soil nutrient estimation models. This study proposed a new method to improve soil nutrient estimation. Firstly, for obtaining characteristic variables, we employed partial least squares regression (PLSR) fit degree to select an optimal screening algorithm from three algorithms (Pearson correlation coefficient, PCC; least absolute shrinkage and selection operator, LASSO; and gradient boosting decision tree, GBDT). Secondly, linear (multi-linear regression, MLR; ridge regression, RR) and nonlinear (support vector machine, SVM; and back propagation neural network with genetic algorithm optimization, GABP) algorithms with 10-fold cross-validation were implemented to determine the most accurate model for estimating soil total nitrogen (TN), total phosphorus (TP), and total potassium (TK) contents. Finally, the new method was used to map the soil TK content at a regional scale using the soil component spectral variables retrieved by the fully constrained least squares (FCLS) method based on an image from the HuanJing-1A Hyperspectral Imager (HJ-1A HSI) of the Conghua District of Guangzhou, China. The results identified the GBDT-GABP was observed as the most accurate estimation method of soil TN ( of 0.69, the root mean square error of cross-validation (RMSECV) of 0.35 g kg⁻¹ and ratio of performance to interquartile range (RPIQ) of 2.03) and TP ( of 0.73, RMSECV of 0.30 g kg⁻¹ and RPIQ = 2.10), and the LASSO-GABP proved to be optimal for soil TK estimations ( of 0.82, RMSECV of 3.39 g kg⁻¹ and RPIQ = 3.57). Additionally, the highly accurate LASSO-GABP-estimated soil TK (R² = 0.79) reveals the feasibility of the LASSO-GABP method to retrieve soil TK content at the regional scale. Full article

(This article belongs to the Special Issue Digital Innovations in Agriculture)

► Show Figures

Figure 1

24 pages, 6336 KiB

Open AccessEditor’s ChoiceArticle

Simplified and Hybrid Remote Sensing-Based Delineation of Management Zones for Nitrogen Variable Rate Application in Wheat

by Mohammad Rokhafrouz, Hooman Latifi, Ali A. Abkar, Tomasz Wojciechowski, Mirosław Czechlowski, Ali Sadeghi Naieni, Yasser Maghsoudi and Gniewko Niedbała

Agriculture 2021, 11(11), 1104; https://0-doi-org.brum.beds.ac.uk/10.3390/agriculture11111104 - 05 Nov 2021

Cited by 15 | Viewed by 2819

Abstract

Enhancing digital and precision agriculture is currently inevitable to overcome the economic and environmental challenges of the agriculture in the 21st century. The purpose of this study was to generate and compare management zones (MZ) based on the Sentinel-2 satellite data for variable [...] Read more.

Enhancing digital and precision agriculture is currently inevitable to overcome the economic and environmental challenges of the agriculture in the 21st century. The purpose of this study was to generate and compare management zones (MZ) based on the Sentinel-2 satellite data for variable rate application of mineral nitrogen in wheat production, calculated using different remote sensing (RS)-based models under varied soil, yield and crop data availability. Three models were applied, including (1) a modified “RS- and threshold-based clustering”, (2) a “hybrid-based, unsupervised clustering”, in which data from different sources were combined for MZ delineation, and (3) a “RS-based, unsupervised clustering”. Various data processing methods including machine learning were used in the model development. Statistical tests such as the Paired Sample T-test, Kruskal–Wallis H-test and Wilcoxon signed-rank test were applied to evaluate the final delineated MZ maps. Additionally, a procedure for improving models based on information about phenological phases and the occurrence of agricultural drought was implemented. The results showed that information on agronomy and climate enables improving and optimizing MZ delineation. The integration of prior knowledge on new climate conditions (drought) in image selection was tested for effective use of the models. Lack of this information led to the infeasibility of obtaining optimal results. Models that solely rely on remote sensing information are comparatively less expensive than hybrid models. Additionally, remote sensing-based models enable delineating MZ for fertilizer recommendations that are temporally closer to fertilization times. Full article

(This article belongs to the Special Issue Digital Innovations in Agriculture)

► Show Figures

Figure 1

15 pages, 7031 KiB

Open AccessEditor’s ChoiceArticle

Evaluation of Deep Learning for Automatic Multi-View Face Detection in Cattle

by Beibei Xu, Wensheng Wang, Leifeng Guo, Guipeng Chen, Yaowu Wang, Wenju Zhang and Yongfeng Li

Agriculture 2021, 11(11), 1062; https://0-doi-org.brum.beds.ac.uk/10.3390/agriculture11111062 - 28 Oct 2021

Cited by 26 | Viewed by 3987

Abstract

Individual identification plays an important part in disease prevention and control, traceability of meat products, and improvement of agricultural false insurance claims. Automatic and accurate detection of cattle face is prior to individual identification and facial expression recognition based on image analysis technology. [...] Read more.

Individual identification plays an important part in disease prevention and control, traceability of meat products, and improvement of agricultural false insurance claims. Automatic and accurate detection of cattle face is prior to individual identification and facial expression recognition based on image analysis technology. This paper evaluated the possibility of the cutting-edge object detection algorithm, RetinaNet, performing multi-view cattle face detection in housing farms with fluctuating illumination, overlapping, and occlusion. Seven different pretrained CNN models (ResNet 50, ResNet 101, ResNet 152, VGG 16, VGG 19, Densenet 121 and Densenet 169) were fine-tuned by transfer learning and re-trained on the dataset in the paper. Experimental results showed that RetinaNet incorporating the ResNet 50 was superior in accuracy and speed through performance evaluation, which yielded an average precision score of 99.8% and an average processing time of 0.0438 s per image. Compared with the typical competing algorithms, the proposed method was preferable for cattle face detection, especially in particularly challenging scenarios. This research work demonstrated the potential of artificial intelligence towards the incorporation of computer vision systems for individual identification and other animal welfare improvements. Full article

(This article belongs to the Special Issue Digital Innovations in Agriculture)

► Show Figures

Figure 1

14 pages, 5241 KiB

Open AccessArticle

Estimation and Forecasting of Rice Yield Using Phenology-Based Algorithm and Linear Regression Model on Sentinel-II Satellite Data

by Abid Nazir, Saleem Ullah, Zulfiqar Ahmad Saqib, Azhar Abbas, Asad Ali, Muhammad Shahid Iqbal, Khalid Hussain, Muhammad Shakir, Munawar Shah and Muhammad Usman Butt

Agriculture 2021, 11(10), 1026; https://0-doi-org.brum.beds.ac.uk/10.3390/agriculture11101026 - 19 Oct 2021

Cited by 24 | Viewed by 6561

Abstract

Rice is a primary food for more than three billion people worldwide and cultivated on about 12% of the world’s arable land. However, more than 88% production is observed in Asian countries, including Pakistan. Due to higher population growth and recent climate change [...] Read more.

Rice is a primary food for more than three billion people worldwide and cultivated on about 12% of the world’s arable land. However, more than 88% production is observed in Asian countries, including Pakistan. Due to higher population growth and recent climate change scenarios, it is crucial to get timely and accurate rice yield estimates and production forecast of the growing season for governments, planners, and decision makers in formulating policies regarding import/export in the event of shortfall and/or surplus. This study aims to quantify the rice yield at various phenological stages from hyper-temporal satellite-derived-vegetation indices computed from time series Sentinel-II images. Different vegetation indices (viz. NDVI, EVI, SAVI, and REP) were used to predict paddy yield. The predicted yield was validated through RMSE and ME statistical techniques. The integration of PLSR and sequential time-stamped vegetation indices accurately predicted rice yield (i.e., maximum R² = 0.84 and minimum RMSE = 0.12 ton ha⁻¹ equal to 3% of the mean rice yield). Moreover, our results also established that optimal time spans for predicting rice yield are late vegetative and reproductive (flowering) stages. The output would be useful for the farmer and decision makers in addressing food security. Full article

(This article belongs to the Special Issue Digital Innovations in Agriculture)

► Show Figures

Figure 1

8 pages, 2518 KiB

Open AccessFeature PaperArticle

Lettuce Growth Pattern Analysis Using U-Net Pre-Trained with Arabidopsis

by Sungyul Chang, Unseok Lee, Min Jeong Hong, Yeong Deuk Jo and Jin-Baek Kim

Agriculture 2021, 11(9), 890; https://0-doi-org.brum.beds.ac.uk/10.3390/agriculture11090890 - 16 Sep 2021

Cited by 5 | Viewed by 3049

Abstract

To overcome the challenges related to food security, digital farming has been proposed, wherein the status of a plant using various sensors could be determined in real time. The high-throughput phenotyping platform (HTPP) and analysis with deep learning (DL) are increasingly being used [...] Read more.

To overcome the challenges related to food security, digital farming has been proposed, wherein the status of a plant using various sensors could be determined in real time. The high-throughput phenotyping platform (HTPP) and analysis with deep learning (DL) are increasingly being used but require a lot of resources. For botanists who have no prior knowledge of DL, the image analysis method is relatively easy to use. Hence, we aimed to explore a pre-trained Arabidopsis DL model to extract the projected area (PA) for lettuce growth pattern analysis. The accuracies of the extract PA of the lettuce cultivar “Nul-chung” with a pre-trained model was measured using the Jaccard Index, and the median value was 0.88 and 0.87 in two environments. Moreover, the growth pattern of green lettuce showed reproducible results in the same environment (p < 0.05). The pre-trained model successfully extracted the time-series PA of lettuce under two lighting conditions (p < 0.05), showing the potential application of a pre-trained DL model of target species in the study of traits in non-target species under various environmental conditions. Botanists and farmers would benefit from fewer challenges when applying up-to-date DL in crop analysis when few resources are available for image analysis of a target crop. Full article

(This article belongs to the Special Issue Digital Innovations in Agriculture)

► Show Figures

Figure 1

13 pages, 1651 KiB

Open AccessArticle

Impact of Lameness on Attributes of Feeding Registered with Noseband Sensor in Fresh Dairy Cows

by Ramūnas Antanaitis, Vida Juozaitienė, Gediminas Urbonavičius, Dovilė Malašauskienė, Mindaugas Televičius, Mingaudas Urbutis and Walter Baumgartner

Agriculture 2021, 11(9), 851; https://0-doi-org.brum.beds.ac.uk/10.3390/agriculture11090851 - 06 Sep 2021

Cited by 5 | Viewed by 2283

Abstract

We hypothesized that lameness in fresh dairy cows (1–30 days after calving) has an impact on attributes of feeding registered with a noseband sensor. The aim of this study was to investigate the impact of lameness in fresh dairy cows on attributes of [...] Read more.

We hypothesized that lameness in fresh dairy cows (1–30 days after calving) has an impact on attributes of feeding registered with a noseband sensor. The aim of this study was to investigate the impact of lameness in fresh dairy cows on attributes of feeding (registered with the RumiWatch noseband sensor): rumination time (RT), drinking time (DT), eating time (ET), rumination chews (RC), eating chews (EC), chews per minute (CM), drinking gulps (DG), bolus count (B), and chews per bolus (CB). The measurement registration was started at the first day after calving and continued until 30 days after calving. There were 20 Lithuanian black and white breed cows selected. Lameness diagnosis was performed by trained staff based on a locomotion score system and it was diagnosed on average on the 15th day after calving. The causes of lameness were categorized as sole ulcer, abscess and foot rot. Special attention was paid to attributes of feeding registered 14 days before and 13 days after diagnosis. The 10 lame cows (LG) used in this experiment had a lameness score of 3–4 presented with severe lameness: they were reluctant to move and unwilling to complete weight transfer off the affected limb. The 10 healthy cows (HG) were given a lameness score of 1. We found that lameness of fresh dairy cows has an impact on inline registered ingestive behaviors biomarkers—the mean RT of HG cows was as much as 2.19 times higher than that of LG cows on the day of diagnosis of lameness, later this difference between the groups decreased to the sixth day of treatment, then increased again and decreased at the end of the experiment. The lowest eating time was found on diagnosis day and the highest on the ninth day before determination of lameness. Drinking time was higher in the HG group, with the exception of 10 and 9 days prior to clinical signs of disease in LG cows. A downward trend in rumination chews was observed in LG cows from day 7 until the onset of clinical symptoms. The bolus count decreased from day 3 before diagnosis to day 1 after diagnosis in LG cows. The largest difference in this indicator between groups was found on day of diagnosis. Analysing the pattern of CM values in the LG group, we found a decrease from 10 days before to 2 days after diagnosis. The CB value was almost the same in both groups of cows at the end of the experiment, but largest difference between the groups was found on day 7 after clinical sings of lameness. Full article

(This article belongs to the Special Issue Digital Innovations in Agriculture)

► Show Figures

Figure 1

15 pages, 1414 KiB

Open AccessArticle

Identification of Risk Factors for Lameness Detection with Help of Biosensors

by Ramūnas Antanaitis, Vida Juozaitienė, Gediminas Urbonavičius, Dovilė Malašauskienė, Mindaugas Televičius, Mingaudas Urbutis, Karina Džermeikaitė and Walter Baumgartner

Agriculture 2021, 11(7), 610; https://0-doi-org.brum.beds.ac.uk/10.3390/agriculture11070610 - 29 Jun 2021

Cited by 7 | Viewed by 2206

Abstract

In this study we hypothesized that the lameness of early lactation dairy cows would have an impact on inline biomarkers, such as rumination time (RT), milk fat (%), milk protein (%), milk fat/protein ratio (F/P), milk lactose (L, %), milk electrical conductivity of [...] Read more.

In this study we hypothesized that the lameness of early lactation dairy cows would have an impact on inline biomarkers, such as rumination time (RT), milk fat (%), milk protein (%), milk fat/protein ratio (F/P), milk lactose (L, %), milk electrical conductivity of all udder quarters, body weight (BW), temperature of reticulorumen content (TRR), pH of reticulorumen content (pH), and walking activity (activity). All 30 lame cows (LCs) used in this experiment had a score of 3–4, identified according to the standard procedure of Sprecher et al. The 30 healthy cows (HC) showed a lameness score of one. RT, milk fat, MY, milk protein, F/P, L, milk electrical conductivity of all udder quarters, and BW were registered using Lely Astronaut^® A3 milking robots each time the cow was being milked. The TRR, cow activity, and pH of the contents of each cow’s reticulorumen were registered using specific smaXtec boluses. The study lasted a total of 28 days. Days “−14” to “−1” denote the days of the experimental period before the onset of clinical signs of lameness (day “0”), and days “1” to “13” indicate the period after the start of treatment. We found that from the ninth day before the diagnosis of laminitis until the end of our study, LCs had higher milk electrical conductivity in all udder quarters, and higher milk fat to protein ratios. On the 3rd day before the onset of clinical signs of the disease until the day of diagnosis, the milk fat of the LC group was reduced. The activity of the LCs decreased sharply from the second day to the first day after treatment. RT in the HC group tended to decrease during the experiment. pH in LCs also increased on the day of the appearance of clinical signs. Full article

(This article belongs to the Special Issue Digital Innovations in Agriculture)

► Show Figures

Figure 1

15 pages, 3840 KiB

Open AccessEditor’s ChoiceArticle

Automated Chicken Counting in Surveillance Camera Environments Based on the Point Supervision Algorithm: LC-DenseFCN

by Liangben Cao, Zihan Xiao, Xianghui Liao, Yuanzhou Yao, Kangjie Wu, Jiong Mu, Jun Li and Haibo Pu

Agriculture 2021, 11(6), 493; https://0-doi-org.brum.beds.ac.uk/10.3390/agriculture11060493 - 26 May 2021

Cited by 19 | Viewed by 6197

Abstract

The density of a chicken population has a great influence on the health and growth of the chickens. For free-range chicken producers, an appropriate population density can increase their economic benefit and be utilized for estimating the economic value of the flock. However, [...] Read more.

The density of a chicken population has a great influence on the health and growth of the chickens. For free-range chicken producers, an appropriate population density can increase their economic benefit and be utilized for estimating the economic value of the flock. However, it is very difficult to calculate the density of chickens quickly and accurately because of the complicated environmental background and the dynamic number of chickens. Therefore, we propose an automated method for quickly and accurately counting the number of chickens on a chicken farm, rather than doing so manually. The contributions of this paper are twofold: (1) we innovatively designed a full convolutional network—DenseFCN—and counted the chickens in an image using the method of point supervision, which achieved an accuracy of 93.84% and 9.27 frames per second (FPS); (2) the point supervision method was used to detect the density of chickens. Compared with the current mainstream object detection method, the higher effectiveness of this method was proven. From the performance evaluation of the algorithm, the proposed method is practical for measuring the density statistics of chickens in a farm environment and provides a new feasible tool for the density estimation of farm poultry breeding. Full article

(This article belongs to the Special Issue Digital Innovations in Agriculture)

► Show Figures

Graphical abstract

14 pages, 4253 KiB

Open AccessArticle

Extraction of Arecanut Planting Distribution Based on the Feature Space Optimization of PlanetScope Imagery

by Yu Jin, Jiawei Guo, Huichun Ye, Jinling Zhao, Wenjiang Huang and Bei Cui

Agriculture 2021, 11(4), 371; https://0-doi-org.brum.beds.ac.uk/10.3390/agriculture11040371 - 19 Apr 2021

Cited by 8 | Viewed by 3677

Abstract

The remote sensing extraction of large areas of arecanut (Areca catechu L.) planting plays an important role in investigating the distribution of arecanut planting area and the subsequent adjustment and optimization of regional planting structures. Satellite imagery has previously been used to [...] Read more.

The remote sensing extraction of large areas of arecanut (Areca catechu L.) planting plays an important role in investigating the distribution of arecanut planting area and the subsequent adjustment and optimization of regional planting structures. Satellite imagery has previously been used to investigate and monitor the agricultural and forestry vegetation in Hainan. However, the monitoring accuracy is affected by the cloudy and rainy climate of this region, as well as the high level of land fragmentation. In this paper, we used PlanetScope imagery at a 3 m spatial resolution over the Hainan arecanut planting area to investigate the high-precision extraction of the arecanut planting distribution based on feature space optimization. First, spectral and textural feature variables were selected to form the initial feature space, followed by the implementation of the random forest algorithm to optimize the feature space. Arecanut planting area extraction models based on the support vector machine (SVM), BP neural network (BPNN), and random forest (RF) classification algorithms were then constructed. The overall classification accuracies of the SVM, BPNN, and RF models optimized by the RF features were determined as 74.82%, 83.67%, and 88.30%, with Kappa coefficients of 0.680, 0.795, and 0.853, respectively. The RF model with optimized features exhibited the highest overall classification accuracy and kappa coefficient. The overall accuracy of the SVM, BPNN, and RF models following feature optimization was improved by 3.90%, 7.77%, and 7.45%, respectively, compared with the corresponding unoptimized classification model. The kappa coefficient also improved. The results demonstrate the ability of PlanetScope satellite imagery to extract the planting distribution of arecanut. Furthermore, the RF is proven to effectively optimize the initial feature space, composed of spectral and textural feature variables, further improving the extraction accuracy of the arecanut planting distribution. This work can act as a theoretical and technical reference for the agricultural and forestry industries. Full article

(This article belongs to the Special Issue Digital Innovations in Agriculture)

► Show Figures

Figure 1

11 pages, 1025 KiB

Open AccessArticle

Prediction of Reproductive Success in Multiparous First Service Dairy Cows by Parameters from In-Line Sensors

by Ramūnas Antanaitis, Vida Juozaitienė, Dovilė Malašauskienė, Mindaugas Televičius, Mingaudas Urbutis, Gintaras Zamokas and Walter Baumgartner

Agriculture 2021, 11(4), 334; https://0-doi-org.brum.beds.ac.uk/10.3390/agriculture11040334 - 08 Apr 2021

Cited by 1 | Viewed by 1987

Abstract

The aim of the current study was to evaluate the relationship of different parameters from an automatic milking system (AMS) with the pregnancy status of multiparous cows at first service and to assess the accuracy of such a follow-up with regard to blood [...] Read more.

The aim of the current study was to evaluate the relationship of different parameters from an automatic milking system (AMS) with the pregnancy status of multiparous cows at first service and to assess the accuracy of such a follow-up with regard to blood parameters. Before the insemination of cows, blood samples for measuring biochemical indices were taken from the coccygeal vessels and the concentrations of blood serum albumin (ALB), cortisol, non-esterified fatty acids (NEFA) and the activities of aspartate aminotransferase (AST) and gamma glutamyltransferase (GGT) were determined. From oestrus day to seven days after oestrus, the following parameters were registered: milk yield (MY), electric milk conductivity, lactate dehydrogenase (LDH) and β-hydroxybutyric acid (BHB). The pregnancy status was evaluated using ultrasound “Easy scan” 30–35 days after insemination. Cows were grouped by reproductive status: PG− (non-pregnant; n = 48) and PG+ (pregnant; n = 44). The BHB level in PG− cows was 1.2 times higher (p < 0.005). The electrical conductivity of milk was statistically significantly higher in all quarters of PG− cows (1.07 times) than of PG+ cows (p < 0.05). The arithmetic mean of blood GGT was 1.61 times higher in PG− cows and the NEFA value 1.23 times higher (p < 0.05) compared with the PG+ group. The liver function was affected, the average ALB of PG− cows was 1.19 times lower (p < 0.05) and the AST activity was 1.16 times lower (p < 0.05) compared with PG+ cows. The non-pregnant group had a negative energy balance demonstrated by high in-line milk BHB and high blood NEFA concentrations. We found a greater number of cows with cortisol >0.0.75 mg/dL in the non-pregnant group. A higher milk electrical conductivity in the non-pregnant cows pointed towards a greater risk of mastitis while higher GGT activities together with lower albumin concentrations indicated that the cows were more affected by oxidative stress. Full article

(This article belongs to the Special Issue Digital Innovations in Agriculture)

► Show Figures

Figure 1

17 pages, 22622 KiB

Open AccessArticle

Simulation of Fuel Consumption Based on Engine Load Level of a 95 kW Partial Power-Shift Transmission Tractor

by Md. Abu Ayub Siddique, Seung-Min Baek, Seung-Yun Baek, Wan-Soo Kim, Yeon-Soo Kim, Yong-Joo Kim, Dae-Hyun Lee, Kwan-Ho Lee and Joon-Yeal Hwang

Agriculture 2021, 11(3), 276; https://0-doi-org.brum.beds.ac.uk/10.3390/agriculture11030276 - 23 Mar 2021

Cited by 10 | Viewed by 6120

Abstract

This study is focused on the estimation of fuel consumption of the power-shift transmission (PST) tractor based on PTO (power take-off) dynamometer test. The simulation model of PST tractor was developed using the configurations and powertrain of the real PST tractor. The PTO [...] Read more.

This study is focused on the estimation of fuel consumption of the power-shift transmission (PST) tractor based on PTO (power take-off) dynamometer test. The simulation model of PST tractor was developed using the configurations and powertrain of the real PST tractor. The PTO dynamometer was installed to measure the engine load and fuel consumption at various engine load levels (40, 50, 60, 70, 80, and 90%), and verify the simulation model. The axle load was also predicted using tractor’s specifications as an input parameter of the simulation model. The simulation and measured results were analyzed and compared statistically. It was observed that the engine load, as well as fuel consumption, were directly proportional to the engine load levels. However, it was statistically proved that there was no significant difference between the simulation and measured engine torque and fuel consumption at each load level. The regression equations show that there was an exponential relationship between the fuel consumption and engine load levels. However, the specific fuel consumptions (SFC) for both simulation and measured were linear relationships and had no significant difference between them at each engine load level. The results were statistically proved that the simulation and measured SFCs were similar trends. The plow tillage operation could be performed at the gear stage of 7.65 km/h with higher working efficiency at low fuel consumption. The drawback of this study is to use a constant axle load instead of dynamic load. This study can provide useful information for both researchers and manufacturers related to the automated transmission of an agricultural tractor, especially PST tractor for digital farming solutions. Finally, it could contribute to the manufacturers developing a new agricultural tractor with higher fuel efficiency. Full article

(This article belongs to the Special Issue Digital Innovations in Agriculture)

► Show Figures

Figure 1

13 pages, 2338 KiB

Open AccessArticle

Can Milk Flow Traits Act as Biomarkers of Lameness in Dairy Cows?

by Vida Juozaitienė, Ramūnas Antanaitis, Gediminas Urbonavičius, Mingaudas Urbutis, Saulius Tušas and Walter Baumgartner

Agriculture 2021, 11(3), 227; https://0-doi-org.brum.beds.ac.uk/10.3390/agriculture11030227 - 09 Mar 2021

Cited by 3 | Viewed by 2097

Abstract

We hypothesized that lameness has an impact on milk flow traits. The aim of the current study was therefore to investigate the relation between lameness and milk flow traits in dairy cows. For this study 73 healthy and 55 cows with lameness were [...] Read more.

We hypothesized that lameness has an impact on milk flow traits. The aim of the current study was therefore to investigate the relation between lameness and milk flow traits in dairy cows. For this study 73 healthy and 55 cows with lameness were selected. Lameness was diagnosed by a local specialized veterinarian, according to the standard procedure. The blood samples were collected during clinical examination. The milking properties of cows were evaluated twice in a row—during evening and morning milking. The selected cows in the current lactation did not receive veterinary treatment, and correct hoof trimming was not performed at least four weeks before the experiment. The measurements were taken by two electronic mobile milk flow meters (Lactocorder^®®, WMB AG, Balgache, Switzerland). Milk flow data were processed using LactoPro 5.2.0 software (Biomelktechnik Swiss). Cortisol concentration was measured with the automated analyzer TOSOH^®® AIA-360 (South San Francisco, CA, USA). We found out that milk flow traits can act as biomarkers of lameness in dairy cows. We determined that the milk yield in the first minute of healthy dairy cows was 1.77 kg higher than that of lame cows. The electrical conductivity during the initial time of milking of healthy cows was 0.24 mS/cm lower than that of the lame group. The milking duration of LA cows was 1.07 min shorter and the time of incline in milk flow from 0.5 kg/min till the reach of the plateau phase was longer. The risk of lameness was most clearly indicated by an increase in blood cortisol concentration; if its blood level in cows exceeds 1 µg/dL, the risk of identifying lameness increases 4.9 times. Full article

(This article belongs to the Special Issue Digital Innovations in Agriculture)

► Show Figures

Figure 1

Review

Jump to: Editorial, Research, Other

26 pages, 3156 KiB

Open AccessEditor’s ChoiceReview

Smart Farming: Internet of Things (IoT)-Based Sustainable Agriculture

by Muthumanickam Dhanaraju, Poongodi Chenniappan, Kumaraperumal Ramalingam, Sellaperumal Pazhanivelan and Ragunath Kaliaperumal

Agriculture 2022, 12(10), 1745; https://0-doi-org.brum.beds.ac.uk/10.3390/agriculture12101745 - 21 Oct 2022

Cited by 88 | Viewed by 91418

Abstract

Smart farming is a development that has emphasized information and communication technology used in machinery, equipment, and sensors in network-based hi-tech farm supervision cycles. Innovative technologies, the Internet of Things (IoT), and cloud computing are anticipated to inspire growth and initiate the use [...] Read more.

Smart farming is a development that has emphasized information and communication technology used in machinery, equipment, and sensors in network-based hi-tech farm supervision cycles. Innovative technologies, the Internet of Things (IoT), and cloud computing are anticipated to inspire growth and initiate the use of robots and artificial intelligence in farming. Such ground-breaking deviations are unsettling current agriculture approaches, while also presenting a range of challenges. This paper investigates the tools and equipment used in applications of wireless sensors in IoT agriculture, and the anticipated challenges faced when merging technology with conventional farming activities. Furthermore, this technical knowledge is helpful to growers during crop periods from sowing to harvest; and applications in both packing and transport are also investigated. Full article

(This article belongs to the Special Issue Digital Innovations in Agriculture)

► Show Figures

Figure 1

17 pages, 2518 KiB

Open AccessEditor’s ChoiceReview

The Role of FAIR Data towards Sustainable Agricultural Performance: A Systematic Literature Review

by Basharat Ali and Peter Dahlhaus

Agriculture 2022, 12(2), 309; https://0-doi-org.brum.beds.ac.uk/10.3390/agriculture12020309 - 21 Feb 2022

Cited by 20 | Viewed by 4135

Abstract

Feeding a growing global population requires improving agricultural production in the face of multidimensional challenges; and digital agriculture is increasingly seen as a strategy for better decision making. Agriculture and agricultural supply chains are increasingly reliant on data, including its access and provision [...] Read more.

Feeding a growing global population requires improving agricultural production in the face of multidimensional challenges; and digital agriculture is increasingly seen as a strategy for better decision making. Agriculture and agricultural supply chains are increasingly reliant on data, including its access and provision from the farm to the consumer. Far-reaching data provision inevitably needs the adoption of FAIR (Findable, Accessible, Interoperable, and Reusable) that offer data originators and depository custodians with a set of guidelines to safeguard a progressive data availability and reusability. Through a systematic literature review it is apparent that although FAIR data principles can play a key role in achieving sustainable agricultural operational and business performance, there are few published studies on how they have been adopted and used. The investigation examines: (1) how FAIR data assimilate with the sustainability framework; and (2) whether the use of FAIR data by the agriculture industry, has an impact on agricultural performance. The work identifies a social science research gap and suggests a method to guide agriculture practitioners in identifying the specific barriers in making their data FAIR. By troubleshooting the barriers, the value propositions of adopting FAIR data in agriculture can be better understood and addressed. Full article

(This article belongs to the Special Issue Digital Innovations in Agriculture)

► Show Figures

Figure 1

17 pages, 1259 KiB

Open AccessReview

Evaluating the Effectiveness and Efficiency of Climate Information Communication in the African Agricultural Sector: A Systematic Analysis of Climate Services

by Chidiebere Ofoegbu and Mark New

Agriculture 2022, 12(2), 160; https://0-doi-org.brum.beds.ac.uk/10.3390/agriculture12020160 - 24 Jan 2022

Cited by 4 | Viewed by 3298

Abstract

The use of climate services (CS) for the provisioning of climate information for informed decision-making on adaptation action has gained momentum. However, a comprehensive review of the literature to evaluate the lessons and experiences of CS implementation in the African agriculture sector is [...] Read more.

The use of climate services (CS) for the provisioning of climate information for informed decision-making on adaptation action has gained momentum. However, a comprehensive review of the literature to evaluate the lessons and experiences of CS implementation in the African agriculture sector is still lacking. Here, we present a systematic review (mapping) of 50 pieces of literature documenting lessons and experiences of CS adoption in the agriculture sector of 20 African countries. The qualitative analysis of the reviewed literature revealed: (1) CS implementation overwhelmingly relied on a participatory process through workshops and participatory scenario planning meetings to connect users with actors along the CS value chain of forecast production, translation, integration, and application. Additionally, innovations such as mobile phones and internet service are increasingly being integrated with CS to strengthen the relationship between CS providers and users. They are, however, mostly at the trial stage and tend to have a varying impact depending on available facilities and infrastructure in the community. (2) Although there is a growing recognition of the need for the integration of indigenous and scientific knowledge systems in the production of climate information, such integration is currently not happening. Rather, indigenous knowledge holders are engaged in a participatory process for insight on modalities of making scientific climate information locally relevant and acceptable. Given the aforementioned findings, we recommend further research on modalities for facilitating indigenous knowledge mainstreaming in climate information production, and investigation of options for using innovations (e.g., mobile) to enhance the interactions between CS users and CS providers. Such research will play a great role in scaling up the adoption of CS in the African agricultural sector. Full article

(This article belongs to the Special Issue Digital Innovations in Agriculture)

► Show Figures

Figure 1

29 pages, 2248 KiB

Open AccessFeature PaperEditor’s ChoiceReview

On Using Artificial Intelligence and the Internet of Things for Crop Disease Detection: A Contemporary Survey

by Houda Orchi, Mohamed Sadik and Mohammed Khaldoun

Agriculture 2022, 12(1), 9; https://0-doi-org.brum.beds.ac.uk/10.3390/agriculture12010009 - 22 Dec 2021

Cited by 51 | Viewed by 15251

Abstract

The agricultural sector remains a key contributor to the Moroccan economy, representing about 15% of gross domestic product (GDP). Disease attacks are constant threats to agriculture and cause heavy losses in the country’s economy. Therefore, early detection can mitigate the severity of diseases [...] Read more.

The agricultural sector remains a key contributor to the Moroccan economy, representing about 15% of gross domestic product (GDP). Disease attacks are constant threats to agriculture and cause heavy losses in the country’s economy. Therefore, early detection can mitigate the severity of diseases and protect crops. However, manual disease identification is both time-consuming and error prone, and requires a thorough knowledge of plant pathogens. Instead, automated methods save both time and effort. This paper presents a contemporary overview of research undertaken over the past decade in the field of disease identification of different crops using machine learning, deep learning, image processing techniques, the Internet of Things, and hyperspectral image analysis. Additionally, a comparative study of several techniques applied to crop disease detection was carried out. Furthermore, this paper discusses the different challenges to be overcome and possible solutions. Then, several suggestions to address these challenges are provided. Finally, this research provides a future perspective that promises to be a highly useful and valuable resource for researchers working in the field of crop disease detection. Full article

(This article belongs to the Special Issue Digital Innovations in Agriculture)

► Show Figures

Graphical abstract

Other

Jump to: Editorial, Research, Review

19 pages, 1387 KiB

Open AccessPerspective

Neural Modelling from the Perspective of Selected Statistical Methods on Examples of Agricultural Applications

by Piotr Boniecki, Agnieszka Sujak, Gniewko Niedbała, Hanna Piekarska-Boniecka, Agnieszka Wawrzyniak and Andrzej Przybylak

Agriculture 2023, 13(4), 762; https://0-doi-org.brum.beds.ac.uk/10.3390/agriculture13040762 - 25 Mar 2023

Cited by 5 | Viewed by 1178

Abstract

Modelling plays an important role in identifying and solving problems that arise in a number of scientific issues including agriculture. Research in the natural environment is often costly, labour demanding, and, in some cases, impossible to carry out. Hence, there is a need [...] Read more.

Modelling plays an important role in identifying and solving problems that arise in a number of scientific issues including agriculture. Research in the natural environment is often costly, labour demanding, and, in some cases, impossible to carry out. Hence, there is a need to create and use specific “substitutes” for originals, known in a broad sense as models. Owing to the dynamic development of computer techniques, simulation models, in the form of information technology (IT) systems that support cognitive processes (of various types), are acquiring significant importance. Models primarily serve to provide a better understanding of studied empirical systems, and for efficient design of new systems as well as their rapid (and also inexpensive) improvement. Empirical mathematical models that are based on artificial neural networks and mathematical statistical methods have many similarities. In practice, scientific methodologies all use different terminology, which is mainly due to historical factors. Unfortunately, this distorts an overview of their mutual correlations, and therefore, fundamentally hinders an adequate comparative analysis of the methods. Using neural modelling terminology, statisticians are primarily concerned with the process of generalisation that involves analysing previously acquired noisy empirical data. Indeed, the objects of analyses, whether statistical or neural, are generally the results of experiments that, by their nature, are subject to various types of errors, including measurement errors. In this overview, we identify and highlight areas of correlation and interfacing between several selected neural network models and relevant, commonly used statistical methods that are frequently applied in agriculture. Examples are provided on the assessment of the quality of plant and animal production, pest risks, and the quality of agricultural environments. Full article

(This article belongs to the Special Issue Digital Innovations in Agriculture)

► Show Figures

Figure 1

Journal Menu

Journal Browser

Digital Innovations in Agriculture

Share This Special Issue

Special Issue Editors

Special Issue Information

Keywords

Published Papers (51 papers)

Editorial

Research

Review

Other

Further Information

Guidelines

MDPI Initiatives

Follow MDPI