Submit to Remote Sensing Review for Remote Sensing Propose a Special Issue

Journal Browser

Thematic Information Extraction and Application in Forests

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

A special issue of Remote Sensing (ISSN 2072-4292). This special issue belongs to the section "Remote Sensing in Agriculture and Vegetation".

Deadline for manuscript submissions: closed (30 June 2021) | Viewed by 34479

Share This Special Issue

Special Issue Editors

Prof. Dr. Xinlian Liang

E-Mail Website
Guest Editor

State Key Laboratory of Surveying and Mapping Remote Sensing Information Engineering, Wuhan University, Wuhan 430072, China
Interests: geodesy and surveying; geoinformatics (GIS); remote sensing; laser scanning; terrestrial laser scanning; mobile laser scanning; 3D modelling; forest inventory
Special Issues, Collections and Topics in MDPI journals

Prof. Dr. Hans Verbeeck

E-Mail Website
Guest Editor

CAVElab-Computational & Applied Vegetation Ecology, Department of Environment, Ghent University, 9000 Gent, Belgium
Interests: carbon cycling; forests; ecosystem modelling
Special Issues, Collections and Topics in MDPI journals

Dr. Markus Hollaus

E-Mail Website
Guest Editor

Department of Geodesy and Geoinformation, University of Technology, 1040 Vienna, Austria
Interests: lidar; forest; biomass; vegetation; change detection; environmental studies; forest inventories
Special Issues, Collections and Topics in MDPI journals

Dr. Ninni Saarinen

E-Mail
Guest Editor

Department of Forest Sciences, University of Helsinki, 00014 Helsinki, Finland

Dr. Yunsheng Wang

E-Mail Website
Guest Editor

Department of Remote Sensing and Photogrammetry, Finnish Geospatial Research Institute (National Land Survey of Finland), 02431 Masala, Finland
Interests: remote sensing（Lidar); tree and forest 3D modeling
Special Issues, Collections and Topics in MDPI journals

Prof. Dr. Juha Hyyppä

E-Mail Website
Guest Editor

Department of Remote Sensing and Photogrammetry, Finnish Geospatial Research Institute, 02430 Masala, Finland
Interests: laser scanning (airborne, mobile, and terrestrial); 3D remote sensing; individual tree detection; virtual forests
Special Issues, Collections and Topics in MDPI journals

Special Issue Information

Dear Colleagues,

Forests, being the largest terrestrial carbon sink and the main host of terrestrial biodiversity, take a significant position in terrestrial ecosystem functions and services. Effective management and sustainable use of forests and forest resources require accurate and efficient understanding of forest attributes in structure, biology, ecology, and phenology domains at various spatial and temporal scales. Those forest ecosystem attributes that are important to understand a phenomenon or are associated with a specific research question are regarded as thematic information (e.g., forest functional or structural traits), which is generally connected with a specific geographic area and in a span of time. Robust and efficient data collection as well as intelligent and reliable data interpretation are the key elements supporting the thematic information extraction. Novel applications of such information trigger a new understanding of forest functionalities and support management and policy decision makings.

This Special Issue calls for original papers that demonstrate new advances in thematic information extractions in various forest types (boreal, temperate, and tropical), and new advances in applications of such information in the forest resource management, climate protection and biodiversity conservation, etc.

Topics of this Special Issue include but are not limited to the following:

Novel sensors, platforms, and their integration for forest thematic information extractions;
New algorithms, methodologies, and procedures of data interpretation and information extraction;
Multiscale, multitemporal information analyses;
Applications of thematic information in silviculture, forest ecology, climate and biodiversity protections, etc.

Dr. Xinlian Liang
Prof. Hans Verbeeck
Dr. Markus Hollaus
Dr. Ninni Saarinen
Dr. Yunsheng Wang
Prof. Juha Hyyppä
Guest Editors

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. Remote Sensing 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 2700 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 (6 papers)

Download All Papers

Research

16 pages, 5908 KiB

Open AccessArticle

Evaluation of Deep Learning Segmentation Models for Detection of Pine Wilt Disease in Unmanned Aerial Vehicle Images

by Lang Xia, Ruirui Zhang, Liping Chen, Longlong Li, Tongchuan Yi, Yao Wen, Chenchen Ding and Chunchun Xie

Remote Sens. 2021, 13(18), 3594; https://0-doi-org.brum.beds.ac.uk/10.3390/rs13183594 - 09 Sep 2021

Cited by 38 | Viewed by 2755

Abstract

Pine wilt disease (PWD) is a serious threat to pine forests. Combining unmanned aerial vehicle (UAV) images and deep learning (DL) techniques to identify infected pines is the most efficient method to determine the potential spread of PWD over a large area. In particular, image segmentation using DL obtains the detailed shape and size of infected pines to assess the disease’s degree of damage. However, the performance of such segmentation models has not been thoroughly studied. We used a fixed-wing UAV to collect images from a pine forest in Laoshan, Qingdao, China, and conducted a ground survey to collect samples of infected pines and construct prior knowledge to interpret the images. Then, training and test sets were annotated on selected images, and we obtained 2352 samples of infected pines annotated over different backgrounds. Finally, high-performance DL models (e.g., fully convolutional networks for semantic segmentation, DeepLabv3+, and PSPNet) were trained and evaluated. The results demonstrated that focal loss provided a higher accuracy and a finer boundary than Dice loss, with the average intersection over union (IoU) for all models increasing from 0.656 to 0.701. From the evaluated models, DeepLLabv3+ achieved the highest IoU and an F1 score of 0.720 and 0.832, respectively. Also, an atrous spatial pyramid pooling module encoded multiscale context information, and the encoder–decoder architecture recovered location/spatial information, being the best architecture for segmenting trees infected by the PWD. Furthermore, segmentation accuracy did not improve as the depth of the backbone network increased, and neither ResNet34 nor ResNet50 was the appropriate backbone for most segmentation models. Full article

(This article belongs to the Special Issue Thematic Information Extraction and Application in Forests)

► Show Figures

Figure 1

14 pages, 2435 KiB

Open AccessArticle

Assessment of Close-Range Remote Sensing Methods for DTM Estimation in a Lowland Deciduous Forest

by Luka Jurjević, Mateo Gašparović, Xinlian Liang and Ivan Balenović

Remote Sens. 2021, 13(11), 2063; https://0-doi-org.brum.beds.ac.uk/10.3390/rs13112063 - 24 May 2021

Cited by 12 | Viewed by 2171

Abstract

Digital terrain models (DTMs) are important for a variety of applications in geosciences as a valuable information source in forest management planning, forest inventory, hydrology, etc. Despite their value, a DTM in a forest area is typically lower quality due to inaccessibility and limited data sources that can be used in the forest environment. In this paper, we assessed the accuracy of close-range remote sensing techniques for DTM data collection. In total, four data sources were examined, i.e., handheld personal laser scanning (PLS_hh, GeoSLAM Horizon), terrestrial laser scanning (TLS, FARO S70), unmanned aerial vehicle (UAV) photogrammetry (UAV_image), and UAV laser scanning (ULS, LS Nano M8). Data were collected within six sample plots located in a lowland pedunculate oak forest. The reference data were of the highest quality available, i.e., total station measurements. After normality and outliers testing, both robust and non-robust statistics were calculated for all close-range remote sensing data sources. The results indicate that close-range remote sensing techniques are capable of achieving higher accuracy (root mean square error < 15 cm; normalized median absolute deviation < 10 cm) than airborne laser scanning (ALS) and digital aerial photogrammetry (DAP) data that are generally understood to be the best data sources for DTM on a large scale. Full article

(This article belongs to the Special Issue Thematic Information Extraction and Application in Forests)

► Show Figures

Figure 1

21 pages, 6003 KiB

Open AccessArticle

Voxel-Based Automatic Tree Detection and Parameter Retrieval from Terrestrial Laser Scans for Plot-Wise Forest Inventory

by Gábor Brolly, Géza Király, Matti Lehtomäki and Xinlian Liang

Remote Sens. 2021, 13(4), 542; https://doi.org/10.3390/rs13040542 - 03 Feb 2021

Cited by 13 | Viewed by 3400

Abstract

This paper presents a fully automatic method addressing tree mapping and parameter extraction (tree position, stem diameter at breast height, stem curve, and tree height) from terrestrial laser scans in forest inventories. The algorithm is designed to detect trees of various sizes and architectures, produce smooth yet accurate stem curves, and achieve tree height estimates in multi-layered stands, all without employing constraints on the shape of the crown. The algorithm also aims to balance estimation accuracy and computational complexity. The method’s tree detection combines voxel operations and stem surface filtering based on scanning point density. Stem diameters are obtained by creating individual taper models, while tree heights are estimated from the segmentation of tree crowns in the voxel-space. Twenty-four sample plots representing diverse forest structures in the south boreal region of Finland have been assessed from single- and multiple terrestrial laser scans. The mean percentages of completeness in stem detection over all stand complexity categories are 50.9% and 68.5% from single and multiple scans, respectively, while the mean root mean square error (RMSE) of the stem curve estimates ranges from ±1.7 to ±2.3 cm, all of which demonstrates the robustness of the algorithm. Efforts were made to accurately locate tree tops by segmenting individual crowns. Nevertheless, with a mean bias of −2.9 m from single scans and −1.3 m from multiple scans, the algorithm proved conservative in tree height estimates. Full article

(This article belongs to the Special Issue Thematic Information Extraction and Application in Forests)

► Show Figures

Graphical abstract

18 pages, 4287 KiB

Open AccessArticle

A Novel Deep Learning Method to Identify Single Tree Species in UAV-Based Hyperspectral Images

by Gabriela Takahashi Miyoshi, Mauro dos Santos Arruda, Lucas Prado Osco, José Marcato Junior, Diogo Nunes Gonçalves, Nilton Nobuhiro Imai, Antonio Maria Garcia Tommaselli, Eija Honkavaara and Wesley Nunes Gonçalves

Remote Sens. 2020, 12(8), 1294; https://0-doi-org.brum.beds.ac.uk/10.3390/rs12081294 - 19 Apr 2020

Cited by 63 | Viewed by 7638

Abstract

Deep neural networks are currently the focus of many remote sensing approaches related to forest management. Although they return satisfactory results in most tasks, some challenges related to hyperspectral data remain, like the curse of data dimensionality. In forested areas, another common problem is the highly-dense distribution of trees. In this paper, we propose a novel deep learning approach for hyperspectral imagery to identify single-tree species in highly-dense areas. We evaluated images with 25 spectral bands ranging from 506 to 820 nm taken over a semideciduous forest of the Brazilian Atlantic biome. We included in our network’s architecture a band combination selection phase. This phase learns from multiple combinations between bands which contributed the most for the tree identification task. This is followed by a feature map extraction and a multi-stage model refinement of the confidence map to produce accurate results of a highly-dense target. Our method returned an f-measure, precision and recall values of 0.959, 0.973, and 0.945, respectively. The results were superior when compared with a principal component analysis (PCA) approach. Compared to other learning methods, ours estimate a combination of hyperspectral bands that most contribute to the mentioned task within the network’s architecture. With this, the proposed method achieved state-of-the-art performance for detecting and geolocating individual tree-species in UAV-based hyperspectral images in a complex forest. Full article

(This article belongs to the Special Issue Thematic Information Extraction and Application in Forests)

► Show Figures

Graphical abstract

20 pages, 7725 KiB

Open AccessEditor’s ChoiceArticle

Tree Species Classification of Drone Hyperspectral and RGB Imagery with Deep Learning Convolutional Neural Networks

by Somayeh Nezami, Ehsan Khoramshahi, Olli Nevalainen, Ilkka Pölönen and Eija Honkavaara

Remote Sens. 2020, 12(7), 1070; https://0-doi-org.brum.beds.ac.uk/10.3390/rs12071070 - 26 Mar 2020

Cited by 116 | Viewed by 11839

Abstract

Interest in drone solutions in forestry applications is growing. Using drones, datasets can be captured flexibly and at high spatial and temporal resolutions when needed. In forestry applications, fundamental tasks include the detection of individual trees, tree species classification, biomass estimation, etc. Deep neural networks (DNN) have shown superior results when comparing with conventional machine learning methods such as multi-layer perceptron (MLP) in cases of huge input data. The objective of this research is to investigate 3D convolutional neural networks (3D-CNN) to classify three major tree species in a boreal forest: pine, spruce, and birch. The proposed 3D-CNN models were employed to classify tree species in a test site in Finland. The classifiers were trained with a dataset of 3039 manually labelled trees. Then the accuracies were assessed by employing independent datasets of 803 records. To find the most efficient set of feature combination, we compare the performances of 3D-CNN models trained with hyperspectral (HS) channels, Red-Green-Blue (RGB) channels, and canopy height model (CHM), separately and combined. It is demonstrated that the proposed 3D-CNN model with RGB and HS layers produces the highest classification accuracy. The producer accuracy of the best 3D-CNN classifier on the test dataset were 99.6%, 94.8%, and 97.4% for pines, spruces, and birches, respectively. The best 3D-CNN classifier produced ~5% better classification accuracy than the MLP with all layers. Our results suggest that the proposed method provides excellent classification results with acceptable performance metrics for HS datasets. Our results show that pine class was detectable in most layers. Spruce was most detectable in RGB data, while birch was most detectable in the HS layers. Furthermore, the RGB datasets provide acceptable results for many low-accuracy applications. Full article

(This article belongs to the Special Issue Thematic Information Extraction and Application in Forests)

► Show Figures

Figure 1

18 pages, 14920 KiB

Open AccessArticle

Quantifying Understory and Overstory Vegetation Cover Using UAV-Based RGB Imagery in Forest Plantation

by Linyuan Li, Jun Chen, Xihan Mu, Weihua Li, Guangjian Yan, Donghui Xie and Wuming Zhang

Remote Sens. 2020, 12(2), 298; https://0-doi-org.brum.beds.ac.uk/10.3390/rs12020298 - 16 Jan 2020

Cited by 38 | Viewed by 5586

Abstract

Vegetation cover estimation for overstory and understory layers provides valuable information for modeling forest carbon and water cycles and refining forest ecosystem function assessment. Although previous studies demonstrated the capability of light detection and ranging (LiDAR) in the three-dimensional (3D) characterization of forest overstory and understory communities, the high cost inhibits its application in frequent and successive survey tasks. Low-cost commercial red–green–blue (RGB) cameras mounted on unmanned aerial vehicles (UAVs), as LiDAR alternatives, provide operational systems for simultaneously quantifying overstory crown cover (OCC) and understory vegetation cover (UVC). We developed an effective method named back-projection of 3D point cloud onto superpixel-segmented image (BAPS) to extract overstory and forest floor pixels using 3D structure-from-motion (SfM) point clouds and two-dimensional (2D) superpixel segmentation. The OCC was estimated from the extracted overstory crown pixels. A reported method, called half-Gaussian fitting (HAGFVC), was used to segement green vegetation and non-vegetation pixels from the extracted forest floor pixels and derive UVC. The UAV-based RGB imagery and field validation data were collected from eight forest plots in Saihanba National Forest Park (SNFP) plantation in northern China. The consistency of the OCC estimates between BAPS and canopy height model (CHM)-based methods (coefficient of determination: 0.7171) demonstrated the capability of the BAPS method in the estimation of OCC. The segmentation of understory vegetation was verified by the supervised classification (SC) method. The validation results showed that the OCC and UVC estimates were in good agreement with reference values, where the root-mean-square error (RMSE) of OCC (unitless) and UVC (unitless) reached 0.0704 and 0.1144, respectively. The low-cost UAV-based observation system and the newly developed method are expected to improve the understanding of ecosystem functioning and facilitate ecological process modeling. Full article

(This article belongs to the Special Issue Thematic Information Extraction and Application in Forests)

► Show Figures

Graphical abstract

Journal Menu

Journal Browser

Thematic Information Extraction and Application in Forests

Share This Special Issue

Special Issue Editors

Special Issue Information

Published Papers (6 papers)

Research

Further Information

Guidelines

MDPI Initiatives

Follow MDPI