Remote Sensing

24 pages, 6287 KiB

Open AccessArticle

Quantitative Genetic Aspects of Accuracy of Tree Biomass Measurement Using LiDAR

by Haruka Sano, Naoko Miura, Minoru Inamori, Yamato Unno, Wei Guo, Sachiko Isobe, Kazutaka Kusunoki and Hiroyoshi Iwata

Remote Sens. 2024, 16(24), 4790; https://doi.org/10.3390/rs16244790 (registering DOI) - 22 Dec 2024

Abstract

The growing focus on the role of forests in carbon sequestration highlights the importance of accurately and efficiently measuring biophysical traits, such as diameter at breast height (DBH) and tree height. Understanding genetic contributions to trait variation is crucial for enhancing carbon storage [...] Read more.

The growing focus on the role of forests in carbon sequestration highlights the importance of accurately and efficiently measuring biophysical traits, such as diameter at breast height (DBH) and tree height. Understanding genetic contributions to trait variation is crucial for enhancing carbon storage through the genetic improvement of forest trees. Light detection and ranging (LiDAR) has been used to estimate DBH and tree height; however, few studies have explored the heritability of these traits or assessed the accuracy of biomass increment selection based on them. Therefore, this study aimed to leverage LiDAR to measure DBH and tree height, estimate tree heritability, and evaluate the accuracy of timber volume selection based on these traits, using 60-year-old larch as the study material. Unmanned aerial vehicle laser scanning (ULS) and backpack laser scanning (BLS) were compared against hand-measured values. The accuracy of DBH estimations using BLS resulted in a root mean square error (RMSE) of 2.7 cm and a coefficient of determination of 0.67. Conversely, the accuracy achieved with ULS was 4.0 cm in RMSE and a 0.24 coefficient of determination. The heritability of DBH was higher with BLS than with ULS and even exceeded that of hand measurements. Comparisons of timber volume selection accuracy based on the measured traits demonstrated comparable performance between BLS and ULS. These findings underscore the po Full article

(This article belongs to the Topic Challenges, Development and Frontiers of Smart Agriculture and Forestry—2nd Volume)

18 pages, 7388 KiB

Open AccessArticle

A Full-Scale Shadow Detection Network Based on Multiple Attention Mechanisms for Remote-Sensing Images

by Lei Zhang, Qing Zhang, Yu Wu, Yanfeng Zhang, Shan Xiang, Donghai Xie and Zeyu Wang

Remote Sens. 2024, 16(24), 4789; https://doi.org/10.3390/rs16244789 (registering DOI) - 22 Dec 2024

Abstract

Shadows degrade image quality and complicate interpretation, underscoring the importance of accurate shadow detection for many image analysis tasks. However, due to the complex backgrounds and variable shadow characteristics of remote sensing images (RSIs), existing methods often struggle with accurately detecting shadows of [...] Read more.

Shadows degrade image quality and complicate interpretation, underscoring the importance of accurate shadow detection for many image analysis tasks. However, due to the complex backgrounds and variable shadow characteristics of remote sensing images (RSIs), existing methods often struggle with accurately detecting shadows of various scales and misclassifying dark, non-shaded areas as shadows. To address these issues, we proposed a comprehensive shadow detection network called MAMNet. Firstly, we proposed a multi-scale spatial channel attention fusion module, which extracted multi-scale features incorporating both spatial and channel information, allowing the model to flexibly adapt to shadows of different scales. Secondly, to address the issue of false detection in non-shadow areas, we introduced a criss-cross attention module, enabling non-shadow pixels to be compared with other shadow and non-shadow pixels in the same row and column, learning similar features of pixels in the same category, which improved the classification accuracy of non-shadow pixels. Finally, to address the issue of important information from the other two modules being lost due to continuous upsampling during the decoding phase, we proposed an auxiliary branch module to assist the main branch in decision-making, ensuring that the final output retained the key information from all stages. The experimental results demonstrated that the model outperformed the current state-of-the-art RSI shadow detection method on the aerial imagery dataset for shadow detection (AISD). The model achieved an overall accuracy (OA) of 97.50%, an

F_{1}

score of 94.07%, an intersection over union (IOU) of 88.87%, a precision of 95.06%, and a BER of 4.05%, respectively. Additionally, visualization results indicated that our model could effectively detect shadows of various scales while avoiding false detection in non-shadow areas. Therefore, this model offers an efficient solution for shadow detection in aerial imagery. Full article

(This article belongs to the Special Issue The Emerging Trends and Applications of Big Data and Machine Learning/Artificial Intelligence (AI) in Remote Sensing II)

23 pages, 10716 KiB

Open AccessArticle

Leveraging the Potential of PRISMA Hyperspectral Data for Forest Tree Species Classification: A Case Study in Southern Italy

by Gabriele Delogu, Miriam Perretta, Eros Caputi, Alessio Patriarca, Cassandra Carroll Funsten, Fabio Recanatesi, Maria Nicolina Ripa and Lorenzo Boccia

Remote Sens. 2024, 16(24), 4788; https://doi.org/10.3390/rs16244788 (registering DOI) - 22 Dec 2024

Abstract

Hyperspectral imagery and advanced classification techniques can significantly enhance remote sensing’s role in forest monitoring. Thanks to recent missions, such as the Italian Space Agency’s PRISMA (PRecursore IperSpettrale della Missione Applicativa—Hyperspectral PRecursor of the Application Mission), hyperspectral data in narrow bands spanning visible/near [...] Read more.

Hyperspectral imagery and advanced classification techniques can significantly enhance remote sensing’s role in forest monitoring. Thanks to recent missions, such as the Italian Space Agency’s PRISMA (PRecursore IperSpettrale della Missione Applicativa—Hyperspectral PRecursor of the Application Mission), hyperspectral data in narrow bands spanning visible/near infrared to shortwave infrared are now available. In this study, hyperspectral data from PRISMA were used with the aim of testing the applicability of PRISMA with different band sizes to classify tree species in highly biodiverse forest environments. The Serre Regional Park in southern Italy was used as a case study. The classification focused on forest category classes based on the predominant tree species in sample plots. Ground truth data were collected using a global positioning system together with a smartphone application to test its contribution to facilitating field data collection. The final result, measured on a test dataset, showed an F1 greater than 0.75 for four classes: fir (0.81), pine (0.77), beech (0.90), and holm oak (0.82). Beech forests showed the highest accuracy (0.92), while chestnut forests (0.68) and a mixed class of hygrophilous species (0.69) showed lower accuracy. These results demonstrate the potential of hyperspectral spaceborne data for identifying trends in spectral signatures for forest tree classification. Full article

(This article belongs to the Special Issue Machine Learning in Global Change Ecology: Methods and Applications)

► Show Figures

Figure 1

13 pages, 4861 KiB

Open AccessTechnical Note

Research on 2-D Direction of Arrival (DOA) Estimation for an L-Shaped Array

by Kun Ye, Lang Zhou, Shaohua Hong, Xuebo Zhang and Haixin Sun

Remote Sens. 2024, 16(24), 4787; https://doi.org/10.3390/rs16244787 (registering DOI) - 22 Dec 2024

Abstract

Lately, there has been a significant increase in interest in coprime array configurations, as they offer the advantage of generating more extensive array apertures and enhanced degrees of freedom when contrasted with standard linear arrays. This document introduces an innovative two-dimensional direction-of-arrival (2-D [...] Read more.

Lately, there has been a significant increase in interest in coprime array configurations, as they offer the advantage of generating more extensive array apertures and enhanced degrees of freedom when contrasted with standard linear arrays. This document introduces an innovative two-dimensional direction-of-arrival (2-D DOA) estimation technique founded on the zero-completion principle. In particular, our initial step involves interpolating the synthetic co-array signals to achieve completion, followed by the regeneration of the covariance matrix utilizing the interpolated synthetic signals. Then, the 2-D angle estimation is realized based on the complemented matrix using the auto-pairing method. The computational modeling outcomes indicate that the suggested method demonstrates superior angular discriminability. Moreover, this method excels in estimation accuracy when contrasted with its algorithmic counterparts. Full article

(This article belongs to the Special Issue Ocean Remote Sensing Based on Radar, Sonar and Optical Techniques)

► Show Figures

Figure 1

21 pages, 1323 KiB

Open AccessArticle

Classification of Karst Rocky Desertification Levels in Jinsha County Using a Feature Space Method Based on SDGSAT-1 Multispectral Data

by Qi Chen, Han Fu, Xiaoming Li, Xiaochuan Qin and Lin Yan

Remote Sens. 2024, 16(24), 4786; https://doi.org/10.3390/rs16244786 (registering DOI) - 22 Dec 2024

Abstract

Karst rocky desertification (KRD) is a significant issue that affects the ecological and economic sustainability of southwest China. Obtaining the accurate distribution of different levels of KRD can provide decision-making support for the effective management of KRD. The Sustainable Development Goals Science Satellite [...] Read more.

Karst rocky desertification (KRD) is a significant issue that affects the ecological and economic sustainability of southwest China. Obtaining the accurate distribution of different levels of KRD can provide decision-making support for the effective management of KRD. The Sustainable Development Goals Science Satellite 1 (SDGSAT-1) is the world’s first scientific satellite serving the 2030 Agenda for Sustainable Development of the United Nations, and is dedicated to developing high-resolution, multi-scale, global public datasets to support policy and decision-making support systems for sustainable development. SDGSAT-1 multispectral data provide detailed ground information with a spatial resolution of 10 m and a rich spectral resolution. In this study, we combined the red-modified carbonate rock index (RCRI, an index that characterizes the degree of carbonate rock exposure) and the normalized difference red edge index (NDRE, an index that characterizes the degree of vegetation coverage) to propose a novel feature space method based on SDGSAT-1 multispectral data to classify the different levels of KRD in the Jinsha County of Guizhou Province, a representative region with significant KRD in southwest China. This method effectively identified different levels of KRD with an overall classification accuracy of 87%. This was 20% higher than that of the grading index method, indicating that SDGSAT-1 multispectral data have promising potential for KRD classification. In this study, we offer a new insight into the classification of KRD and a greater quantity of remote-sensing data to monitor KRD over a wider area and for a longer period of time, contributing to the economic development and environmental protection of KRD areas. Full article

31 pages, 9286 KiB

Open AccessArticle

Exploring Trends and Variability of Water Quality over Lake Titicaca Using Global Remote Sensing Products

by Vann Harvey Maligaya, Analy Baltodano, Afnan Agramont and Ann van Griensven

Remote Sens. 2024, 16(24), 4785; https://doi.org/10.3390/rs16244785 (registering DOI) - 22 Dec 2024

Abstract

Understanding the current water quality dynamics is necessary to ensure that ecological and sociocultural services are provided to the population and the natural environment. Water quality monitoring of lakes is usually performed with in situ measurements; however, these are costly, time consuming, laborious, [...] Read more.

Understanding the current water quality dynamics is necessary to ensure that ecological and sociocultural services are provided to the population and the natural environment. Water quality monitoring of lakes is usually performed with in situ measurements; however, these are costly, time consuming, laborious, and can have limited spatial coverage. Nowadays, remote sensing offers an alternative source of data to be used in water quality monitoring; by applying appropriate algorithms to satellite imagery, it is possible to retrieve water quality parameters. The use of global remote sensing water quality products increased in the last decade, and there are a multitude of products available from various databases. However, in Latin America, studies on the inter-comparison of the applicability of these products for water quality monitoring is rather scarce. Therefore, in this study, global remote sensing products estimating various water quality parameters were explored on Lake Titicaca and compared with each other and sources of data. Two products, the Copernicus Global Land Service (CGLS) and the European Space Agency Lakes Climate Change Initiative (ESA-CCI), were evaluated through a comparison with in situ measurements and with each other for analysis of the spatiotemporal variability of lake surface water temperature (LSWT), turbidity, and chlorophyll-a. The results of this study showed that the two products had limited accuracy when compared to in situ data; however, remarkable performance was observed in terms of exhibiting spatiotemporal variability of the WQ parameters. The ESA-CCI LSWT product performed better than the CGLS product in estimating LSWT, while the two products were on par with each other in terms of demonstrating the spatiotemporal patterns of the WQ parameters. Overall, these two global remote sensing water quality products can be used to monitor Lake Titicaca, currently with limited accuracy, but they can be improved with precise pixel identification, accurate optical water type definition, and better algorithms for atmospheric correction and retrieval. This highlights the need for the improvement of global WQ products to fit local conditions and make the products more useful for decision-making at the appropriate scale. Full article

(This article belongs to the Special Issue Environmental Monitoring Using Satellite Remote Sensing (Second Edition))

18 pages, 4247 KiB

Open AccessArticle

Temporal Vine Water Status Modeling Through Machine Learning Ensemble Technique and Sentinel-2 Multispectral Images Under Semi-Arid Conditions

by Vincenzo Giannico, Simone Pietro Garofalo, Luca Brillante, Pietro Sciusco, Mario Elia, Giuseppe Lopriore, Salvatore Camposeo, Raffaele Lafortezza, Giovanni Sanesi and Gaetano Alessandro Vivaldi

Remote Sens. 2024, 16(24), 4784; https://doi.org/10.3390/rs16244784 (registering DOI) - 22 Dec 2024

Abstract

New challenges will be experienced by the agriculture sector in the near future, especially due to the effects of climate change. For example, rising temperatures could result in increased evapotranspiration demand, causing difficulties in the management of irrigation practices. Generally, an important predictor [...] Read more.

New challenges will be experienced by the agriculture sector in the near future, especially due to the effects of climate change. For example, rising temperatures could result in increased evapotranspiration demand, causing difficulties in the management of irrigation practices. Generally, an important predictor of plant water status to be taken into account for irrigation monitoring and management is the stem water potential. However, it requires a huge amount of time-consuming fieldwork, particularly when an adequate data amount is necessary to fully investigate the spatial and temporal variability of large areas under monitoring. In this study, the integration of machine learning and satellite remote sensing (Sentinel-2) was investigated to obtain a model able to predict the stem water potential in viticulture using multispectral imagery. Vine water status data were acquired within a Montepulciano vineyard in the south of Italy (Puglia region), under semi-arid conditions; data were acquired over two years during the irrigation seasons. Different machine learning algorithms (lasso, ridge, elastic net, and random forest) were compared using vegetation indices and spectral bands as predictors in two independent analyses. The results show that it is possible to remotely estimate vine water status with random forest from vegetation indices (R² = 0.72). Integrating machine learning techniques and satellite remote sensing could help farmers and technicians manage and plan irrigation, avoiding or reducing fieldwork. Full article

(This article belongs to the Special Issue Advanced Pattern Recognition and Image Processing Technology for Agricultural Engineering)

► Show Figures

Figure 1

21 pages, 12247 KiB

Open AccessArticle

The Impact of Autumn Snowfall on Vegetation Indices and Autumn Phenology Estimation

by Yao Tang, Jin Chen, Jingyi Xu, Jiahui Xu, Jingwen Ni, Zhaojun Zheng, Bailang Yu, Jianping Wu and Yan Huang

Remote Sens. 2024, 16(24), 4783; https://doi.org/10.3390/rs16244783 (registering DOI) - 22 Dec 2024

Abstract

Monitoring autumn vegetation dynamics in alpine regions is crucial for managing local livestock, understanding regional productivity, and assessing the responses of alpine regions to climate change. However, remote sensing-based vegetation monitoring is significantly affected by snowfall. The impact of autumn snowfall, particularly when [...] Read more.

Monitoring autumn vegetation dynamics in alpine regions is crucial for managing local livestock, understanding regional productivity, and assessing the responses of alpine regions to climate change. However, remote sensing-based vegetation monitoring is significantly affected by snowfall. The impact of autumn snowfall, particularly when vegetation has not fully entered dormancy, has been largely overlooked. To demonstrate the uncertainties caused by autumn snowfall in remote sensing-based vegetation monitoring, we analyzed 16 short-term snowfall events in the Qinghai–Tibet Plateau. We employed a synthetic difference-in-differences estimation framework and conducted simulated experiments to isolate the impact of snowfall from other factors, revealing its effects on vegetation indices (VIs) and autumn phenology estimation. Our findings indicate that autumn snowfall notably affects commonly used VIs and their associated phenology estimates. Modified VIs (i.e., Normalized Difference Infrared Index (NDII), Phenology Index (PI), Normalized Difference Phenology Index (NDPI), and Normalized Difference Greenness Index (NDGI)) revealed greater resilience to snowfall compared to conventional VIs (i.e., Normalized Difference Vegetation Index (NDVI) and Enhanced Vegetation Index (EVI)) in phenology estimation. Areas with remaining green vegetation in autumn showed more pronounced numerical changes in VIs due to snowfall. Furthermore, the impact of autumn snowfall closely correlated with underlying vegetation types. Forested areas experienced less impact from snowfall compared to grass- and shrub-dominated regions. Earlier snowfall onset and increased snowfall frequency further exacerbated deviations in estimated phenology caused by snowfall. This study highlights the significant impact of autumn snowfall on remote sensing-based vegetation monitoring and provides a scientific basis for accurate vegetation studies in high-altitude regions. Full article

(This article belongs to the Section Environmental Remote Sensing)

► Show Figures

Figure 1

21 pages, 55404 KiB

Open AccessArticle

Significant Wave Height Retrieval in Tropical Cyclone Conditions Using CYGNSS Data

by Xiangyang Han, Xianwei Wang, Zhi He and Jinhua Wu

Remote Sens. 2024, 16(24), 4782; https://doi.org/10.3390/rs16244782 (registering DOI) - 22 Dec 2024

Abstract

The retrieval of global significant wave height (SWH) data is crucial for maritime navigation, aquaculture safety, and oceanographic research. Leveraging the high temporal resolution and spatial coverage of Cyclone Global Navigation Satellite System (CYGNSS) data, machine learning models have shown promise in SWH [...] Read more.

The retrieval of global significant wave height (SWH) data is crucial for maritime navigation, aquaculture safety, and oceanographic research. Leveraging the high temporal resolution and spatial coverage of Cyclone Global Navigation Satellite System (CYGNSS) data, machine learning models have shown promise in SWH retrieval. However, existing models struggle with accuracy under high-SWH conditions and discard a significant number of such observations due to low quality, which limits their effectiveness in global SWH retrieval, particularly for monitoring tropical cyclone (TC) events. To address this, this study proposes a daily global SWH retrieval framework through the enhanced eXtreme Gradient Boosting model (XGBoost-SC), which incorporates Cumulative Distribution Function (CDF) matching to introduce prior distribution information and reduce errors for SWH values exceeding 3 m. An enhanced loss function is employed to improve accuracy and mitigate the distribution bias in low-SWH retrieval induced by CDF matching. The results were tested over one million sample points and validated against the European Centre for Medium-Range Weather Forecasts (ECMWF) SWH product. With the help of CDF matching, XGBoost-SC outperformed all models, significantly reducing RMSE and bias while improving the retrieval capability for high SWHs. For SWH values between 3–6 m, the RMSE and bias were 0.94 m and −0.44 m, and for values above 6 m, they were 2.79 m and −2.0 m. The enhanced performance of XGBoost-SC for large SWHs was further confirmed in TC conditions over the Western North Pacific and in the Western Atlantic Ocean. This study provides a reference for large-scale SWH retrieval, particularly under TC conditions. Full article

(This article belongs to the Special Issue Latest Advances and Application in the GNSS-R Field)

26 pages, 15194 KiB

Open AccessArticle

Cross-Attention-Based High Spatial-Temporal Resolution Fusion of Sentinel-2 and Sentinel-3 Data for Ocean Water Quality Assessment

by Yanfeng Wen, Peng Chen, Zhenhua Zhang and Yunzhou Li

Remote Sens. 2024, 16(24), 4781; https://doi.org/10.3390/rs16244781 (registering DOI) - 22 Dec 2024

Abstract

Current marine research that leverages remote sensing data urgently requires gridded data of high spatial and temporal resolution. However, such high-quality data is often lacking due to the inherent physical and technical constraints of sensors. A necessary trade-off therefore exists between spatial, temporal, [...] Read more.

Current marine research that leverages remote sensing data urgently requires gridded data of high spatial and temporal resolution. However, such high-quality data is often lacking due to the inherent physical and technical constraints of sensors. A necessary trade-off therefore exists between spatial, temporal, and spectral resolution in satellite remote sensing technology: increasing spatial resolution often reduces the coverage area, thereby diminishing temporal resolution. This manuscript introduces an innovative remote sensing image fusion algorithm that combines Sentinel-2 (high spatial resolution) and Sentinel-3 (relatively high spectral and temporal resolution) satellite data. The algorithm, based on a cross-attention mechanism and referred to as the Cross-Attention Spatio-Temporal Spectral Fusion (CASTSF) model, accounts for variations in spectral channels, spatial resolution, and temporal phase among different sensor images. The proposed method enables the fusion of atmospherically corrected ocean remote sensing reflectance products (Level 2 OSR), yielding high-resolution spatial data at 10 m resolution with a temporal frequency of 1–2 days. Subsequently, the algorithm generates chlorophyll-a concentration remote sensing products characterized by enhanced spatial and temporal fidelity. A comparative analysis against existing chlorophyll-a concentration products demonstrates the robustness and effectiveness of the proposed approach, highlighting its potential for advancing remote sensing applications. Full article

(This article belongs to the Section Ocean Remote Sensing)

► Show Figures

Figure 1

20 pages, 13020 KiB

Open AccessArticle

Multi-Dimensional and Multi-Scale Physical Dehazing Network for Remote Sensing Images

by Hao Zhou, Le Wang, Qiao Li, Xin Guan and Tao Tao

Remote Sens. 2024, 16(24), 4780; https://doi.org/10.3390/rs16244780 (registering DOI) - 22 Dec 2024

Abstract

Haze obscures remote sensing images, making it difficult to extract valuable information. To address this problem, we propose a fine detail extraction network that aims to restore image details and improve image quality. Specifically, to capture fine details, we design multi-scale and multi-dimensional [...] Read more.

Haze obscures remote sensing images, making it difficult to extract valuable information. To address this problem, we propose a fine detail extraction network that aims to restore image details and improve image quality. Specifically, to capture fine details, we design multi-scale and multi-dimensional extraction blocks and then fuse them to optimize feature extraction. The multi-scale extraction block adopts multi-scale pixel attention and channel attention to extract and combine global and local information from the image. Meanwhile, the multi-dimensional extraction block uses depthwise separable convolutional layers to capture additional dimensional information. Additionally, we integrate an atmospheric scattering model unit into the network to enhance both the dehazing effectiveness and stability. Our experiments on the SateHaze1k and HRSD datasets demonstrate that the proposed method efficiently handles remote sensing images with varying levels of haze, successfully recovers fine details, and achieves superior results compared to existing state-of-the-art dehazing techniques. Full article

(This article belongs to the Special Issue Deep Learning for Remote Sensing Image Enhancement)

► Show Figures

Figure 1

23 pages, 11445 KiB

Open AccessArticle

Distributed Target Detection with Coherent Fusion in Tracking Based on Phase Prediction

by Aoya Wang, Jing Lu, Shenghua Zhou and Linhai Wang

Remote Sens. 2024, 16(24), 4779; https://doi.org/10.3390/rs16244779 (registering DOI) - 21 Dec 2024

Abstract

In distributed radar, a coherent system often gains attention for its higher detection potential in contrast to its non-coherent counterpart. However, even for a distributed coherent radar, it is difficult to coherently accumulate local observations in the searching mode if target returns in [...] Read more.

In distributed radar, a coherent system often gains attention for its higher detection potential in contrast to its non-coherent counterpart. However, even for a distributed coherent radar, it is difficult to coherently accumulate local observations in the searching mode if target returns in local channels are decorrelated. In order to obtain the superiority of coherent processing while overcoming the real implementation difficulties of a coherent framework, this paper studies a distributed coherent detection algorithm for fusion detection. It is utilized in detecting a target during tracking while a target is searched for in a non-coherent manner. From historic observations on target tracking, relative phase delays in different channels are predicted by a phase lock loop and then used to compensate phases for observations in the current frame. Moreover, to enhance the detection performance of distributed radar during tracking, a switching rule between phase prediction-based coherent and non-coherent processing is proposed based on their detection performance. Numerical results indicate that the switching operation can improve the detection probability during tracking, and the non-coherent operation can still provide a moderate detection performance if the phase prediction is unreliable. Full article

(This article belongs to the Special Issue Advances in Multiple Sensor Fusion and Classification for Object Detection and Tracking)

► Show Figures

Figure 1

21 pages, 28159 KiB

Open AccessArticle

Spatiotemporal Variation and Driving Factors of Ecological Environment Quality on the Loess Plateau in China from 2000 to 2020

by Shuaizhi Kang, Xia Jia, Yonghua Zhao, Lei Han, Chaoqun Ma and Yu Bai

Remote Sens. 2024, 16(24), 4778; https://doi.org/10.3390/rs16244778 (registering DOI) - 21 Dec 2024

Abstract

The Loess Plateau (LP) in China is an ecologically fragile region that has long faced challenges such as soil erosion, water shortages, and land degradation. The spatial and temporal variations in ecological environment quality on the LP from 2000 to 2020 were analyzed [...] Read more.

The Loess Plateau (LP) in China is an ecologically fragile region that has long faced challenges such as soil erosion, water shortages, and land degradation. The spatial and temporal variations in ecological environment quality on the LP from 2000 to 2020 were analyzed using the Remote Sensing Ecological Index (RSEI) on the Google Earth Engine (GEE) platform. The Sen, Mann–Kendall, and Hurst exponent analyses were used to examine the spatial variation trends over the past 20 years, while Geodetector identified key factors influencing RSEI changes and their interactions. The results indicate that (1) RSEI effectively represents the ecological and environmental quality of the LP, with 47% of the study area’s annual mean RSEI values over the 20-year period classified as moderate, ranging from 0.017 to 0.815. (2) Ecological quality trends showed improvement in 72% of the area, with a 90% overall increase, but 84% of these trends are not likely to continue. (3) Key factors influencing RSEI changes during abrupt change years included precipitation, land use/land cover, and soil sediment content, with precipitation and topography emerging as primary influences on ecological quality. Although natural factors largely drive ecological changes, human activities also exert both positive and negative effects. This study underscores the importance of sustainable ecological management and provides policy insights for advancing ecological civilization on the LP, contributing to the achievement of the Sustainable Development Goals (SDGs). Full article

(This article belongs to the Special Issue Integration of Remote Sensing and GIS to Forest and Grassland Ecosystem Monitoring)

► Show Figures

Figure 1

19 pages, 14386 KiB

Open AccessArticle

Deep Learning Method for Wetland Segmentation in Unmanned Aerial Vehicle Multispectral Imagery

by Pakezhamu Nuradili, Ji Zhou, Guiyun Zhou and Farid Melgani

Remote Sens. 2024, 16(24), 4777; https://doi.org/10.3390/rs16244777 (registering DOI) - 21 Dec 2024

Abstract

This study highlights the importance of unmanned aerial vehicle (UAV) multispectral (MS) imagery for the accurate delineation and analysis of wetland ecosystems, which is crucial for their conservation and management. We present an enhanced semantic segmentation algorithm designed for UAV MS imagery, which [...] Read more.

This study highlights the importance of unmanned aerial vehicle (UAV) multispectral (MS) imagery for the accurate delineation and analysis of wetland ecosystems, which is crucial for their conservation and management. We present an enhanced semantic segmentation algorithm designed for UAV MS imagery, which incorporates thermal infrared (TIR) data to improve segmentation outcomes. Our approach, involving meticulous image preprocessing, customized network architecture, and iterative training procedures, aims to refine wetland boundary delineation. The algorithm demonstrates strong segmentation results, including a mean pixel accuracy (MPA) of 90.35% and a mean intersection over union (MIOU) of 73.87% across different classes, with a pixel accuracy (PA) of 95.42% and an intersection over union (IOU) of 90.46% for the wetland class. The integration of TIR data with MS imagery not only enriches the feature set for segmentation but also, to some extent, helps address data imbalance issues, contributing to a more refined ecological analysis. This approach, along with the development of a comprehensive dataset that reflects the diversity of wetland environments and advances the utility of remote sensing technologies in ecological monitoring. This research lays the groundwork for more detailed and informative UAV-based evaluations of wetland health and integrity. Full article

(This article belongs to the Special Issue Deep Learning for the Analysis of Multi-/Hyperspectral Images II)

► Show Figures

Figure 1

27 pages, 23239 KiB

Open AccessArticle

Integration of Structural Characteristics from GEDI Waveforms for Improved Forest Type Classification

by Mary M. McClure, Satoshi Tsuyuki and Takuya Hiroshima

Remote Sens. 2024, 16(24), 4776; https://doi.org/10.3390/rs16244776 (registering DOI) - 21 Dec 2024

Abstract

Abstract: Forest types correspond to differences in structural characteristics and species composition that influence biomass and biodiversity values, which are essential measurements for ecological monitoring and management. However, differentiating forest types in tropical regions remains a challenge. This study aimed to improve forest [...] Read more.

Abstract: Forest types correspond to differences in structural characteristics and species composition that influence biomass and biodiversity values, which are essential measurements for ecological monitoring and management. However, differentiating forest types in tropical regions remains a challenge. This study aimed to improve forest type extent mapping by combining structural information from discrete full-waveform LiDAR returns with multitemporal images. This study was conducted in a tropical forest region over complex terrain in north-eastern Tanzania. First, structural classes were generated by applying time-series clustering algorithms. The results showed four different structural clusters corresponding to forest types, montane–humid forest, montane–dry forest, submontane forest, and non-forest, when using the Kshape algorithm. Kshape considers the shape of the full-sequence LiDAR waveform, requiring little preprocessing. Despite the overlap amongst the original clusters, the averages of structural characteristics were significantly different across all but five metrics. The labeled clusters were then further refined and used as training data to generate a wall-to-wall forest cover type map by classifying biannual images. The highest-performing model was a KNN model with 13 spectral and 3 terrain features achieving 81.7% accuracy. The patterns in the distributions of forest types provide better information from which to adapt forest management, particularly in forest–non-forest transitional zones. Full article

(This article belongs to the Section Environmental Remote Sensing)

28 pages, 14547 KiB

Open AccessArticle

A Contrastive-Augmented Memory Network for Anti-UAV Tracking in TIR Videos

by Ziming Wang, Yuxin Hu, Jianwei Yang, Guangyao Zhou, Fangjian Liu and Yuhan Liu

Remote Sens. 2024, 16(24), 4775; https://doi.org/10.3390/rs16244775 (registering DOI) - 21 Dec 2024

Abstract

With the development of unmanned aerial vehicle (UAV) technology, the threat of UAV intrusion is no longer negligible. Therefore, drone perception, especially anti-UAV tracking technology, has gathered considerable attention. However, both traditional Siamese and transformer-based trackers struggle in anti-UAV tasks due to the [...] Read more.

With the development of unmanned aerial vehicle (UAV) technology, the threat of UAV intrusion is no longer negligible. Therefore, drone perception, especially anti-UAV tracking technology, has gathered considerable attention. However, both traditional Siamese and transformer-based trackers struggle in anti-UAV tasks due to the small target size, clutter backgrounds and model degradation. To alleviate these challenges, a novel contrastive-augmented memory network (CAMTracker) is proposed for anti-UAV tracking tasks in thermal infrared (TIR) videos. The proposed CAMTracker conducts tracking through a two-stage scheme, searching for possible candidates in the first stage and matching the candidates with the template for final prediction. In the first stage, an instance-guided region proposal network (IG-RPN) is employed to calculate the correlation features between the templates and the searching images and further generate candidate proposals. In the second stage, a contrastive-augmented matching module (CAM), along with a refined contrastive loss function, is designed to enhance the discrimination ability of the tracker under the instruction of contrastive learning strategy. Moreover, to avoid model degradation, an adaptive dynamic memory module (ADM) is proposed to maintain a dynamic template to cope with the feature variation of the target in long sequences. Comprehensive experiments have been conducted on the Anti-UAV410 dataset, where the proposed CAMTracker achieves the best performance compared to advanced tracking algorithms, with significant advantages on all the evaluation metrics, including at least 2.40%, 4.12%, 5.43% and 5.48% on precision, success rate, success AUC and state accuracy, respectively. Full article

(This article belongs to the Special Issue Artificial Intelligence-Driven Methods for Remote Sensing Target and Object Detection II)

► Show Figures

Figure 1

18 pages, 23648 KiB

Open AccessArticle

Land Subsidence Velocity and High-Speed Railway Risks in the Coastal Cities of Beijing–Tianjin–Hebei, China, with 2015–2021 ALOS PALSAR-2 Multi-Temporal InSAR Analysis

by Qingli Luo, Mengli Li, Zhiyuan Yin, Peifeng Ma, Daniele Perissin and Yuanzhi Zhang

Remote Sens. 2024, 16(24), 4774; https://doi.org/10.3390/rs16244774 (registering DOI) - 21 Dec 2024

Abstract

Sea-level rise has important implications for the economic and infrastructure security of coastal cities. Land subsidence further exacerbates relative sea-level rise. The Beijing–Tianjin–Hebei region (BTHR) along the Bohai Bay is one of the areas most severely affected by ground subsidence in the world. [...] Read more.

Sea-level rise has important implications for the economic and infrastructure security of coastal cities. Land subsidence further exacerbates relative sea-level rise. The Beijing–Tianjin–Hebei region (BTHR) along the Bohai Bay is one of the areas most severely affected by ground subsidence in the world. This study applies the Small Baseline Subset Interferometric Synthetic Aperture Radar (SBAS InSAR) method to analyze 47 ALOS PALSAR-2 images with five frames, mapping subsidence across 21,677.7 km² and revealing spatial patterns and trends over time from 2015 to 2021. This is one of the few published research studies for large-scale and long-term analysis of its kind using ALOS-2 data in this region. The results reveal the existence of six major areas affected by severe subsidence in the study area, with the most pronounced in Jinzhan Town, Beijing, with the maximum subsiding velocity of −94.42 mm/y. Except for the two subsidence areas located in Chaoyang District of Beijing and Guangyang District of Langfang City, the other areas with serious subsidence detected are all located in suburban areas; this means that the strict regulations of controlling urban subsidence for downtown areas in the BTHR have worked. The accumulated subsidence is highly correlated with the time in the time series. Moreover, the subsidence of 161.4 km of the Beijing–Tianjin Inter-City High-Speed Railway (HSR) and 194.5 km of the Beijing–Shanghai HSR (out of a total length of 1318 km) were analyzed. It is the first time that PALSAR-2 data have been used to simultaneously investigate the subsidence along two important HSR lines in China and to analyze relatively long sections of the routes. The above two railways intersect five and seven subsiding areas, respectively. Within the range of the monitored railway line, the percentage of the section with subsidence velocity below −10 mm/y in the monitoring length range is 11.2% and 27.9%; this indicates that the Beijing–Shanghai HSR has suffered more serious subsidence than the Beijing–Tianjin Inter-City HSR within the monitoring period. This research is also beneficial for assessing the subsidence risk associated with different railways. In addition, this study further analyzed the potential reasons for the serious land subsidence of the identified areas. The results of the geological interpretation still indicate that the main cause of subsidence in the area is due to hydrogeological characteristics and underground water withdrawal. Full article

(This article belongs to the Special Issue Synthetic Aperture Radar Interferometry Symposium 2024)

27 pages, 24916 KiB

Open AccessArticle

Multipath and Deep Learning-Based Detection of Ultra-Low Moving Targets Above the Sea

by Zhaolong Wang, Xiaokuan Zhang, Weike Feng, Binfeng Zong, Tong Wang, Cheng Qi and Xixi Chen

Remote Sens. 2024, 16(24), 4773; https://doi.org/10.3390/rs16244773 (registering DOI) - 21 Dec 2024

Abstract

An intelligent approach is proposed and investigated in this paper for the detection of ultra-low-altitude sea-skimming moving targets for airborne pulse Doppler radar. Without suppressing interferences, the proposed method uses both target and multipath information for detection based on their distinguishable image features [...] Read more.

An intelligent approach is proposed and investigated in this paper for the detection of ultra-low-altitude sea-skimming moving targets for airborne pulse Doppler radar. Without suppressing interferences, the proposed method uses both target and multipath information for detection based on their distinguishable image features and deep learning (DL) techniques. First, the image features of the target, multipath, and sea clutter in the real-measured range-Doppler (RD) map are analyzed, based on which the target and multipath are defined together as the generalized target. Then, based on the composite electromagnetic scattering mechanism of the target and the ocean surface, a scattering-based echo generation model is established and validated to generate sufficient data for DL network training. Finally, the RD features of the generalized target are learned by training the DL-based target detector, such as you-only-look-once version 7 (YOLOv7) and Faster R-CNN. The detection results show the high performance of the proposed method on both simulated and real-measured data without suppressing interferences (e.g., clutter, jamming, and noise). In particular, even if the target is submerged in clutter, the target can still be detected by the proposed method based on the multipath feature. Full article

(This article belongs to the Special Issue Array and Signal Processing for Radar)

25 pages, 8692 KiB

Open AccessArticle

Monitoring Spatial-Temporal Variability of Vegetation Coverage and Its Influencing Factors in the Yellow River Source Region from 2000 to 2020

by Boyang Wang, Jianhua Si, Bing Jia, Xiaohui He, Dongmeng Zhou, Xinglin Zhu, Zijin Liu, Boniface Ndayambaza and Xue Bai

Remote Sens. 2024, 16(24), 4772; https://doi.org/10.3390/rs16244772 (registering DOI) - 21 Dec 2024

Abstract

As a vital conservation area for water sources in the Yellow River Basin, understanding the spatial-temporal dynamics of vegetation coverage is crucial, along with the factors that affect it, to ensure ecological preservation and sustainable development of the Yellow River Source Region (YRSR). [...] Read more.

As a vital conservation area for water sources in the Yellow River Basin, understanding the spatial-temporal dynamics of vegetation coverage is crucial, along with the factors that affect it, to ensure ecological preservation and sustainable development of the Yellow River Source Region (YRSR). In this paper, we utilized Landsat surface reflectance data from 2000 to 2020 using de-clouding and masking methods implementing the Google Earth Engine (GEE) cloud platform. We investigated spatial-temporal changes in vegetation coverage by combining the maximum value composite (MVC), the dimidiate pixel model (DPM), the Theil–Sen median slope, and the Mann–Kendall test. The influencing factors on vegetation coverage were quantitatively analyzed using a geographic detector, and future tendencies in vegetation coverage were predicted utilizing the Future Land Use Simulation (FLUS) model. The outcomes suggested the following: (1) On the temporal scale, vegetation coverage exhibited a general upward trend between 2000 and 2020, with the YRSR showing a yearly growth rate of 0.23% (p < 0.001). In comparison to 2000, the area designated as having extremely high vegetation coverage increased by 19.3% in 2020. (2) Spatially, the central and southeast regions have higher values of vegetation coverage, whereas the northwest has lower values. In the study area, 75.5% of the region demonstrated a significant improvement trend, primarily in Xinghai County, Zeku County, and Dari County in the south and the northern portion of the YRSR; conversely, a notable tendency of degradation was identified in 11.8% of the area, mostly in the southeastern areas of Qumalai County, Chenduo County, Shiqu County, and scattered areas in the southeastern region. (3) With an explanatory power of exceeding 45%, the three influencing factors that had the biggest effects on vegetation coverage were mean annual temperature, elevation, and mean annual precipitation. Mean annual precipitation has been shown to have a major impact on vegetation covering; the interconnections involving these factors have increased the explanatory power of vegetation coverage’s regional distribution. (4) Predictions for 2030 show that the vegetation coverage is trending upward in the YRSR, with a notable recovery trend in the northwestern region. This study supplies a theoretical foundation to formulate strategies to promote sustainable development and ecological environmental preservation in the YRSR. Full article

Journal Description

Remote Sensing

Latest Articles

Journal Menu

Journal Browser

Highly Accessed Articles

Latest Books

E-Mail Alert

News

Topics

Conferences

Special Issues

Topical Collections

Further Information

Guidelines

MDPI Initiatives

Follow MDPI

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.