Submit to Applied Sciences Review for Applied Sciences Propose a Special Issue

Journal Browser

Deep Learning for Signal Processing Applications

Print Special Issue Flyer
Special Issue Editors
Special Issue Information
Published Papers

A special issue of Applied Sciences (ISSN 2076-3417). This special issue belongs to the section "Computing and Artificial Intelligence".

Deadline for manuscript submissions: closed (31 May 2023) | Viewed by 32630

Share This Special Issue

Special Issue Editor

Prof. Dr. Valentina E. Balas

E-Mail Website
Guest Editor

Automatics and Applied Software Department, Aurel Vlaicu University of Arad, 310130 Arad, Romania
Interests: intelligent systems; soft computing; fuzzy control; modeling and simulation; biometrics
Special Issues, Collections and Topics in MDPI journals

Special Issue Information

Dear Colleagues,

In recent times, deep learning has emerged as one of the most effective learning techniques in the broader area of artificial intelligence, especially for image and video analysis. Deep learning techniques have been used extensively for computer vision and recently for video analysis. In fact, in industry and academia scientists and research scholars have come up with effective solutions for various image and video related problems using different deep learning algorithms. The prime reason for the growing popularity of deep learning is that it can achieve high recognition accuracy than earlier methods that were existing before. With the applications of deep learning, excellent results have been achieved in image and video related classification and segmentation. While substantial progress has been achieved in medical image analysis with deep learning, many issues still remain and new problems emerge such as deep learning in medical imaging focusing on MRI with high accuracy, availability of limited datasets for classification task, major problems due to imbalance data sets, detecting diseases from medical imaging, image registration, and computer-aided diagnosis. Apart from medical images, deep learning can also be applied to solve other problems such as image inpainting, sound classification, voice assistants and augmented intelligence, high-resolution medical image reconstruction. Recently due to the introduction of deep learning, video analysis has become more interesting. Earlier it was a challenging task as the video was a data-intensive media with huge variations and complexities. Thanks to the deep learning technology, the multimedia people are now able develop better performance-intensive techniques to analyse the content of the video.

This special issue aims to provide comprehensive coverage on cutting edge research and state of the art methods on deep learning applications, especially with images and videos. Authors are requested to submit papers on the following topics (but not limited to):

Image classification and segmentation by deep learning techniques
Object Detection, image reconstruction, image super-resolution and image synthesis by deep learning techniques.
Cancer imaging using deep learning techniques.
Deep Learning in Gastrointestinal Endoscopy.
Tumour detection using deep learning.
Deep learning for image analysis using multimodality fusion.
Image quality recognition methods inspired by deep learning.
Advanced Deep Learning methods in computer vision with 3D data.
Deep Learning models to solve the task of MOT (Multiple Object Tracking).
Novel applications of Deep learning in a video classification framework
Deep learning techniques for video semantic segmentation.
Applications of deep learning video and image forensics.
Video summarization using deep learning.
Human action recognition using deep learning.
Application of deep learning in satellite imagery.
Aerospace, defense and communications.
Industrial automation
Automotive

Prof. Dr. Valentina E. Balas
Guest Editor

Manuscript Submission Information

Manuscripts should be submitted online at www.mdpi.com by registering and logging in to this website. Once you are registered, click here to go to the submission form. Manuscripts can be submitted until the deadline. All submissions that pass pre-check are peer-reviewed. Accepted papers will be published continuously in the journal (as soon as accepted) and will be listed together on the special issue website. Research articles, review articles as well as short communications are invited. For planned papers, a title and short abstract (about 100 words) can be sent to the Editorial Office for announcement on this website.

Submitted manuscripts should not have been published previously, nor be under consideration for publication elsewhere (except conference proceedings papers). All manuscripts are thoroughly refereed through a single-blind peer-review process. A guide for authors and other relevant information for submission of manuscripts is available on the Instructions for Authors page. Applied Sciences is an international peer-reviewed open access semimonthly journal published by MDPI.

Please visit the Instructions for Authors page before submitting a manuscript. The Article Processing Charge (APC) for publication in this open access journal is 2400 CHF (Swiss Francs). Submitted papers should be well formatted and use good English. Authors may use MDPI's English editing service prior to publication or during author revisions.

Published Papers (9 papers)

Download All Papers

Research

21 pages, 772 KiB

Open AccessArticle

Fractional Derivative Gradient-Based Optimizers for Neural Networks and Human Activity Recognition

by Oscar Herrera-Alcántara

Appl. Sci. 2022, 12(18), 9264; https://0-doi-org.brum.beds.ac.uk/10.3390/app12189264 - 15 Sep 2022

Cited by 3 | Viewed by 1554

Abstract

In this paper, fractional calculus principles are considered to implement fractional derivative gradient optimizers for the Tensorflow backend. The performance of these fractional derivative optimizers is compared with that of other well-known ones. Our experiments consider some human activity recognition (HAR) datasets, and the results show that there is a subtle difference between the performance of the proposed method and other existing ones. The main conclusion is that fractional derivative gradient descent optimizers could help to improve the performance of training and validation tasks and opens the possibility to include more fractional calculus concepts to neural networks applied to HAR. Full article

(This article belongs to the Special Issue Deep Learning for Signal Processing Applications)

► Show Figures

Figure 1

16 pages, 7066 KiB

Open AccessArticle

Study on Underwater Target Tracking Technology Based on an LSTM–Kalman Filtering Method

by Maofa Wang, Chuzhen Xu, Chuanping Zhou, Youping Gong and Baochun Qiu

Appl. Sci. 2022, 12(10), 5233; https://0-doi-org.brum.beds.ac.uk/10.3390/app12105233 - 22 May 2022

Cited by 5 | Viewed by 1974

Abstract

In the marine environment, underwater targets are often affected by interference from other targets and environmental fluctuations, so traditional target tracking methods are difficult to use for tracking underwater targets stably and accurately. Among the traditional methods, the Kalman filtering method is widely used; however, it only has advantages in solving linear problems and it is difficult to use to realize effective tracking problems when the trajectory of the moving target is nonlinear. Aiming to solve this limitation, an LSTM–Kalman filtering method was proposed, which can efficiently solve the problem of overly large deviations in underwater target tracking. Using this method, we first studied the features of typical underwater targets and, according to these rules, constructed the corresponding target dataset. Second, we built a convolutional neural network (CNN) model to detect the target and determine the tracking value of the moving target. We used a long-term and short-term memory artificial neural network (LSTM-NN) to modify the Kalman filter to predict the azimuth and distance of the target and to update it iteratively. Then, we verified the new method using simulation tests and the measured data from an acoustic sea trial. The results showed that compared to the traditional Kalman filtering method, the relative error of the LSTM–Kalman filtering method was reduced by 60% in the simulation tests and 72.25% in the sea trial and that the estimation variance was only 4.79. These results indicate that the method that is proposed in this paper achieves good prediction results and a high prediction efficiency for underwater target tracking. Full article

(This article belongs to the Special Issue Deep Learning for Signal Processing Applications)

► Show Figures

Figure 1

12 pages, 4536 KiB

Open AccessArticle

Vector Magnetic Anomaly Detection via an Attention Mechanism Deep-Learning Model

by Xueshan Wu, Song Huang, Min Li and Yufeng Deng

Appl. Sci. 2021, 11(23), 11533; https://0-doi-org.brum.beds.ac.uk/10.3390/app112311533 - 06 Dec 2021

Cited by 9 | Viewed by 2728

Abstract

Magnetic anomaly detection (MAD) is used for detecting moving ferromagnetic targets. In this study, we present an end-to-end deep-learning model for magnetic anomaly detection on data recorded by a single static three-axis magnetometer. We incorporate an attention mechanism into our network to improve [...] Read more.

1 / f^{α}

, which is similar to the field magnetic noise. Our method does not require another magnetometer to eliminate the effects of the Earth’s magnetic field or external interferences. We evaluate the network’s performance through computer simulations and real-world experiments. The high detection performance and the single magnetometer implementation show great potential for real-time detection and edge computing. Full article

(This article belongs to the Special Issue Deep Learning for Signal Processing Applications)

► Show Figures

Figure 1

21 pages, 14296 KiB

Open AccessArticle

Study on Active Tracking of Underwater Acoustic Target Based on Deep Convolution Neural Network

by Maofa Wang, Baochun Qiu, Zeifei Zhu, Huanhuan Xue and Chuanping Zhou

Appl. Sci. 2021, 11(16), 7530; https://0-doi-org.brum.beds.ac.uk/10.3390/app11167530 - 17 Aug 2021

Cited by 8 | Viewed by 2025

Abstract

The active tracking technology of underwater acoustic targets is an important research direction in the field of underwater acoustic signal processing and sonar, and it has always been issued that draws researchers’ attention. The commonly used Kalman filter active tracking (KFAT) method is an effective tracking method, however, it is difficult to detect weak SNR signals, and it is easy to lose the target after the azimuth of different targets overlaps. This paper proposes a KFAT based on deep convolutional neural network (DCNN) method, which can effectively solve the problem of target loss. First, we use Kalman filtering to predict the azimuth and distance of the target, and then use the trained model to identify the azimuth-weighted time-frequency image to obtain the azimuth and label of the target and obtain the target distance by the time the target appears in the time-frequency image. Finally, we associate the data according to the target category, and update the target azimuth and distance information for this cycle. In this paper, two methods, KFAT and DCNN-KFAT, are simulated and tested, and the results are obtained for two cases of tracking weak signal-to-noise signals and tracking different targets with overlapping azimuths. The simulation results show that the DCNN-KFAT method can solve the problem that the KFAT method is difficult to track the target under the weak SNR and the problem that the target is easily lost when two different targets overlap in azimuth. It reduces the deviation range of the active tracking to within 200 m, which is 500~700 m less than the KFAT method. Full article

(This article belongs to the Special Issue Deep Learning for Signal Processing Applications)

► Show Figures

Figure 1

16 pages, 65872 KiB

Open AccessArticle

Implementation of Pavement Defect Detection System on Edge Computing Platform

by Yu-Chen Lin, Wen-Hui Chen and Cheng-Hsuan Kuo

Appl. Sci. 2021, 11(8), 3725; https://doi.org/10.3390/app11083725 - 20 Apr 2021

Cited by 11 | Viewed by 3275

Abstract

Road surfaces in Taiwan, as well as other developed countries, often experience structural failures, such as patches, bumps, longitudinal and lateral cracking, and potholes, which cause discomfort and pose direct safety risks to motorists. To minimize damage to vehicles from pavement defects or provide the corresponding comfortable ride promotion strategy later, in this study, we developed a pavement defect detection system using a deep learning perception scheme for implementation on Xilinx Edge AI platforms. To increase the detection distance and accuracy of pavement defects, two cameras with different fields of view, at

70^{\circ}

and

30^{\circ}

, respectively, were used to capture the front views of a car, and then the YOLOv3 (you only look once, version 3) model was employed to recognize the pavement defects, such as potholes, cracks, manhole covers, patches, and bumps. In addition, to promote continuous pavement defect recognition rate, a tracking-via-detection strategy was employed, which first detects pavement defects in each frame and then associates them to different frames using the Kalman filter method. Thus, the average detection accuracy of the pothole category could reach 71%, and the miss rate was about 29%. To confirm the effectiveness of the proposed detection strategy, experiments were conducted on an established Taiwan pavement defect image dataset (TPDID), which is the first dataset for Taiwan pavement defects. Moreover, different AI methods were used to detect the pavement defects for quantitative comparative analysis. Finally, a field-programmable gate-array-based edge computing platform was used as an embedded system to implement the proposed YOLOv3-based pavement defect detection system; the execution speed reached 27.8 FPS while maintaining the accuracy of the original system model. Full article

(This article belongs to the Special Issue Deep Learning for Signal Processing Applications)

► Show Figures

Figure 1

18 pages, 10771 KiB

Open AccessArticle

A Deep Neural Network Model for Speaker Identification

by Feng Ye and Jun Yang

Appl. Sci. 2021, 11(8), 3603; https://0-doi-org.brum.beds.ac.uk/10.3390/app11083603 - 16 Apr 2021

Cited by 58 | Viewed by 8172

Abstract

Speaker identification is a classification task which aims to identify a subject from a given time-series sequential data. Since the speech signal is a continuous one-dimensional time series, most of the current research methods are based on convolutional neural network (CNN) or recurrent neural network (RNN). Indeed, these methods perform well in many tasks, but there is no attempt to combine these two network models to study the speaker identification task. Due to the spectrogram that a speech signal contains, the spatial features of voiceprint (which corresponds to the voice spectrum) and CNN are effective for spatial feature extraction (which corresponds to modeling spectral correlations in acoustic features). At the same time, the speech signal is in a time series, and deep RNN can better represent long utterances than shallow networks. Considering the advantage of gated recurrent unit (GRU) (compared with traditional RNN) in the segmentation of sequence data, we decide to use stacked GRU layers in our model for frame-level feature extraction. In this paper, we propose a deep neural network (DNN) model based on a two-dimensional convolutional neural network (2-D CNN) and gated recurrent unit (GRU) for speaker identification. In the network model design, the convolutional layer is used for voiceprint feature extraction and reduces dimensionality in both the time and frequency domains, allowing for faster GRU layer computation. In addition, the stacked GRU recurrent network layers can learn a speaker’s acoustic features. During this research, we tried to use various neural network structures, including 2-D CNN, deep RNN, and deep LSTM. The above network models were evaluated on the Aishell-1 speech dataset. The experimental results showed that our proposed DNN model, which we call deep GRU, achieved a high recognition accuracy of 98.96%. At the same time, the results also demonstrate the effectiveness of the proposed deep GRU network model versus other models for speaker identification. Through further optimization, this method could be applied to other research similar to the study of speaker identification. Full article

(This article belongs to the Special Issue Deep Learning for Signal Processing Applications)

► Show Figures

Figure 1

25 pages, 1749 KiB

Open AccessArticle

Real-Time Physical Activity Recognition on Smart Mobile Devices Using Convolutional Neural Networks

by Konstantinos Peppas, Apostolos C. Tsolakis, Stelios Krinidis and Dimitrios Tzovaras

Appl. Sci. 2020, 10(23), 8482; https://0-doi-org.brum.beds.ac.uk/10.3390/app10238482 - 27 Nov 2020

Cited by 27 | Viewed by 3311

Abstract

Given the ubiquity of mobile devices, understanding the context of human activity with non-intrusive solutions is of great value. A novel deep neural network model is proposed, which combines feature extraction and convolutional layers, able to recognize human physical activity in real-time from tri-axial accelerometer data when run on a mobile device. It uses a two-layer convolutional neural network to extract local features, which are combined with 40 statistical features and are fed to a fully-connected layer. It improves the classification performance, while it takes up 5–8 times less storage space and outputs more than double the throughput of the current state-of-the-art user-independent implementation on the Wireless Sensor Data Mining (WISDM) dataset. It achieves 94.18% classification accuracy on a 10-fold user-independent cross-validation of the WISDM dataset. The model is further tested on the Actitracker dataset, achieving 79.12% accuracy, while the size and throughput of the model are evaluated on a mobile device. Full article

(This article belongs to the Special Issue Deep Learning for Signal Processing Applications)

► Show Figures

Figure 1

19 pages, 2281 KiB

Open AccessArticle

Deep Learning Models Compression for Agricultural Plants

by Arnauld Nzegha Fountsop, Jean Louis Ebongue Kedieng Fendji and Marcellin Atemkeng

Appl. Sci. 2020, 10(19), 6866; https://0-doi-org.brum.beds.ac.uk/10.3390/app10196866 - 30 Sep 2020

Cited by 20 | Viewed by 3579

Abstract

Deep learning has been successfully showing promising results in plant disease detection, fruit counting, yield estimation, and gaining an increasing interest in agriculture. Deep learning models are generally based on several millions of parameters that generate exceptionally large weight matrices. The latter requires large memory and computational power for training, testing, and deploying. Unfortunately, these requirements make it difficult to deploy on low-cost devices with limited resources that are present at the fieldwork. In addition, the lack or the bad quality of connectivity in farms does not allow remote computation. An approach that has been used to save memory and speed up the processing is to compress the models. In this work, we tackle the challenges related to the resource limitation by compressing some state-of-the-art models very often used in image classification. For this we apply model pruning and quantization to LeNet5, VGG16, and AlexNet. Original and compressed models were applied to the benchmark of plant seedling classification (V2 Plant Seedlings Dataset) and Flavia database. Results reveal that it is possible to compress the size of these models by a factor of 38 and to reduce the FLOPs of VGG16 by a factor of 99 without considerable loss of accuracy. Full article

(This article belongs to the Special Issue Deep Learning for Signal Processing Applications)

► Show Figures

Figure 1

14 pages, 2893 KiB

Open AccessArticle

Incremental Dilations Using CNN for Brain Tumor Classification

by Sanjiban Sekhar Roy, Nishant Rodrigues and Y-h. Taguchi

Appl. Sci. 2020, 10(14), 4915; https://0-doi-org.brum.beds.ac.uk/10.3390/app10144915 - 17 Jul 2020

Cited by 31 | Viewed by 3714

Abstract

Brain tumor classification is a challenging task in the field of medical image processing. Technology has now enabled medical doctors to have additional aid for diagnosis. We aim to classify brain tumors using MRI images, which were collected from anonymous patients and artificial brain simulators. In this article, we carry out a comparative study between Simple Artificial Neural Networks with dropout, Basic Convolutional Neural Networks (CNN), and Dilated Convolutional Neural Networks. The experimental results shed light on the high classification performance (accuracy 97%) of Dilated CNN. On the other hand, Dilated CNN suffers from the gridding phenomenon. An incremental, even number dilation rate takes advantage of the reduced computational overhead and also overcomes the adverse effects of gridding. Comparative analysis between different combinations of dilation rates for the different convolution layers, help validate the results. The computational overhead in terms of efficiency for training the model to reach an acceptable threshold accuracy of 90% is another parameter to compare the model performance. Full article

(This article belongs to the Special Issue Deep Learning for Signal Processing Applications)

► Show Figures

Figure 1

Journal Menu

Journal Browser

Deep Learning for Signal Processing Applications

Share This Special Issue

Special Issue Editor

Special Issue Information

Published Papers (9 papers)

Research

Further Information

Guidelines

MDPI Initiatives

Follow MDPI