Being one of the most significant and valuable coral reef systems in the South China Sea, the Xisha Islands has undergone rapid transformation due to increasing stressors from human impacts and climate change in recent years. However, as indispensable information for coral reef monitoring and management, the detailed reef extent, geomorphic zonation, or benthic composition of the Xisha Islands is not well documented. Considering limited access to the Xisha Islands, the rapid development of optical remote sensing technology provides us with a feasible mean for coral reef observation. This study adopted a water depth substitution index – probabilistic inundation (PI) – combined with depth-invariant index (DII) to achieve reef extent exploration, geomorphologic and benthic habitat types classification with unsupervised classification algorithms based on Landsat-8 time-series satellite data. Compared with two open-access datasets, the extent of each independent reef extracted from PI exhibited higher similarity with the actual boundary conditions displayed in RGB (Red-Green-Blue) composite images from Landsat-8. Based on PI and derived slope, we obtained geomorphic zonation classification results, and similarly benthic compositions were retrieved based on PI, DII, and reflectance. The overall accuracy of geomorphic zonation and benthic habitat classification results were 72% and 86%, respectively. We also interestingly discovered that corals of the Xisha Islands may be capable of an ability to resist chronic heat stress as a growth trend of reef area after two successive stress events in 2014–2015 were observed at most reefs. The proposed mapping framework of this study provides a repeatable and flexible scheme in depicting the comprehensive situation of coral reefs at Xisha Islands based only on publicly available remote sensing data without complicated pre-set parameters, which could be easily extended to coral reef research around the world. Simultaneously, the findings also provide requisite information supporting the sustainable management and conservation of coral reef ecosystems in the Xisha Islands.
{"title":"Coral reef applications of Landsat-8: geomorphic zonation and benthic habitat mapping of Xisha Islands, China","authors":"Mingjun He, Junyu He, Yajun Zhou, Liyuan Sun, Shuangyan He, Cong Liu, Yanzhen Gu, Peiliang Li","doi":"10.1080/15481603.2023.2261213","DOIUrl":"https://doi.org/10.1080/15481603.2023.2261213","url":null,"abstract":"Being one of the most significant and valuable coral reef systems in the South China Sea, the Xisha Islands has undergone rapid transformation due to increasing stressors from human impacts and climate change in recent years. However, as indispensable information for coral reef monitoring and management, the detailed reef extent, geomorphic zonation, or benthic composition of the Xisha Islands is not well documented. Considering limited access to the Xisha Islands, the rapid development of optical remote sensing technology provides us with a feasible mean for coral reef observation. This study adopted a water depth substitution index – probabilistic inundation (PI) – combined with depth-invariant index (DII) to achieve reef extent exploration, geomorphologic and benthic habitat types classification with unsupervised classification algorithms based on Landsat-8 time-series satellite data. Compared with two open-access datasets, the extent of each independent reef extracted from PI exhibited higher similarity with the actual boundary conditions displayed in RGB (Red-Green-Blue) composite images from Landsat-8. Based on PI and derived slope, we obtained geomorphic zonation classification results, and similarly benthic compositions were retrieved based on PI, DII, and reflectance. The overall accuracy of geomorphic zonation and benthic habitat classification results were 72% and 86%, respectively. We also interestingly discovered that corals of the Xisha Islands may be capable of an ability to resist chronic heat stress as a growth trend of reef area after two successive stress events in 2014–2015 were observed at most reefs. The proposed mapping framework of this study provides a repeatable and flexible scheme in depicting the comprehensive situation of coral reefs at Xisha Islands based only on publicly available remote sensing data without complicated pre-set parameters, which could be easily extended to coral reef research around the world. Simultaneously, the findings also provide requisite information supporting the sustainable management and conservation of coral reef ecosystems in the Xisha Islands.","PeriodicalId":55091,"journal":{"name":"GIScience & Remote Sensing","volume":"83 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-09-27","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135535606","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-09-22DOI: 10.1080/15481603.2023.2260637
Yifang Shi, W. Daniel Kissling
National and regional data products of the ecosystem structure derived from airborne laser scanning (ALS) surveys with Light Detection And Ranging (LiDAR) technology are essential for ecology, biodiversity, and ecosystem monitoring. However, noises like powerlines often remain, hindering the accurate measurement of 3D ecosystem structures from LiDAR. Currently, there is a lack of studies assessing powerline noise removal in the context of generating data products of ecosystem structures from ALS point clouds. Here, we assessed the (1) performance and accuracy, (2) effectiveness, and (3) time efficiency and execution time of three powerline extraction methods (i.e. two point-based methods based on deep learning and eigenvalue decomposition, respectively, and one hybrid method) for removing powerline noise when quantifying 3D ecosystem structures in landscapes with varying canopy heights and vegetation openness. Twenty-five LiDAR metrics representing three key dimensions of the ecosystem structure (i.e. vegetation height, cover, and vertical variability) across 10 study areas in the Netherlands were used for our assessment. The deep learning method had the best performance and showed the highest accuracy of powerline removal across various landscape types (average F1 score = 96%), closely followed by the hybrid method (average F1 score = 95%). In contrast, the accuracy of the eigenvalue decomposition method was lower (average F1 score = 82%) and depended on landscape context and vegetation composition (e.g. the F1 score decreased from 96% to 63% when the average canopy height increased across landscapes). Powerline noise removal had the highest effectiveness (i.e. generating LiDAR metrics closest to those derived from manually labeled ground truth data) for LiDAR metrics capturing height and cover of low- and high-vegetation layers. Time efficiency (processed points per second) was highest for the eigenvalue decomposition method, yet the hybrid method reduced the execution time by > 50% compared to the deep learning method (ranging from 20% to 89% in study areas with different landscape composition). Based on our findings, we recommend the hybrid method for upscaling powerline removal on multi-terabyte ALS datasets to a regional or national extent because of its high accuracy and computational efficiency. Remaining misclassifications in LiDAR metrics could be further minimized by improving the training dataset for deep learning models (e.g. including various shapes of transmission towers from different datasets). Our findings provide novel insights into the performance of different powerline extraction methods, how their effectiveness varies for improving vegetation metrics and mapping the 3D ecosystem structure from LiDAR, and their computational efficiency for upscaling powerline removal in multi-terabyte ALS datasets to a national extent.
{"title":"Performance, effectiveness and computational efficiency of powerline extraction methods for quantifying ecosystem structure from light detection and ranging","authors":"Yifang Shi, W. Daniel Kissling","doi":"10.1080/15481603.2023.2260637","DOIUrl":"https://doi.org/10.1080/15481603.2023.2260637","url":null,"abstract":"National and regional data products of the ecosystem structure derived from airborne laser scanning (ALS) surveys with Light Detection And Ranging (LiDAR) technology are essential for ecology, biodiversity, and ecosystem monitoring. However, noises like powerlines often remain, hindering the accurate measurement of 3D ecosystem structures from LiDAR. Currently, there is a lack of studies assessing powerline noise removal in the context of generating data products of ecosystem structures from ALS point clouds. Here, we assessed the (1) performance and accuracy, (2) effectiveness, and (3) time efficiency and execution time of three powerline extraction methods (i.e. two point-based methods based on deep learning and eigenvalue decomposition, respectively, and one hybrid method) for removing powerline noise when quantifying 3D ecosystem structures in landscapes with varying canopy heights and vegetation openness. Twenty-five LiDAR metrics representing three key dimensions of the ecosystem structure (i.e. vegetation height, cover, and vertical variability) across 10 study areas in the Netherlands were used for our assessment. The deep learning method had the best performance and showed the highest accuracy of powerline removal across various landscape types (average F1 score = 96%), closely followed by the hybrid method (average F1 score = 95%). In contrast, the accuracy of the eigenvalue decomposition method was lower (average F1 score = 82%) and depended on landscape context and vegetation composition (e.g. the F1 score decreased from 96% to 63% when the average canopy height increased across landscapes). Powerline noise removal had the highest effectiveness (i.e. generating LiDAR metrics closest to those derived from manually labeled ground truth data) for LiDAR metrics capturing height and cover of low- and high-vegetation layers. Time efficiency (processed points per second) was highest for the eigenvalue decomposition method, yet the hybrid method reduced the execution time by > 50% compared to the deep learning method (ranging from 20% to 89% in study areas with different landscape composition). Based on our findings, we recommend the hybrid method for upscaling powerline removal on multi-terabyte ALS datasets to a regional or national extent because of its high accuracy and computational efficiency. Remaining misclassifications in LiDAR metrics could be further minimized by improving the training dataset for deep learning models (e.g. including various shapes of transmission towers from different datasets). Our findings provide novel insights into the performance of different powerline extraction methods, how their effectiveness varies for improving vegetation metrics and mapping the 3D ecosystem structure from LiDAR, and their computational efficiency for upscaling powerline removal in multi-terabyte ALS datasets to a national extent.","PeriodicalId":55091,"journal":{"name":"GIScience & Remote Sensing","volume":"17 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-09-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"136060370","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-09-19DOI: 10.1080/15481603.2023.2257978
Babak Mohammadi, Petter Pilesjö, Zheng Duan
Accurate monitoring of glaciers’ extents and their dynamics is essential for improving our understanding of the impacts of climate and environmental changes in cold regions. The satellite-based Normalized Difference Snow Index (NDSI) has been widely used for mapping snow cover and glaciers around the globe. However, mapping glaciers in snow-covered areas using existing indices remains a challenging task due to their incapabilities in separating snow, glaciers, and water. This study aimed to evaluate a new satellite-based index and apply machine learning algorithms to improve the accuracy of mapping glaciers. A new index based on satellite data from Sentinel-2 was tested, which we call the Adjusted Normalized Difference Snow Index (ANDSI). ANDSI (besides NDSI) was used with five different machine learning algorithms, namely Artificial Neural Network, C5.0 Decision Tree Algorithm, Naive Bayes classifier, Support Vector Machine, and Extreme Gradient Boosting, to map glaciers, and their performance was evaluated against ground reference data. Four glacierized regions in different countries (Canada, China, Sweden, and Switzerland-Italy) were selected as study sites to evaluate the performance of the proposed ANDSI. Results showed that the proposed ANDSI outperformed the original NDSI, and the C5.0 classifier showed the best overall accuracy and Kappa among the selected five machine learning classifiers in the majority of cases. The original NDSI yielded results with an average overall accuracy of (around) 91% and the proposed ANDSI with (around) 95% for glacier mapping across all models and study regions. This study demonstrates that the proposed ANDSI serves as a superior and improved method for accurately mapping glaciers in cold regions.
{"title":"The superiority of the Adjusted Normalized Difference Snow Index (ANDSI) for mapping glaciers using Sentinel-2 multispectral satellite imagery","authors":"Babak Mohammadi, Petter Pilesjö, Zheng Duan","doi":"10.1080/15481603.2023.2257978","DOIUrl":"https://doi.org/10.1080/15481603.2023.2257978","url":null,"abstract":"Accurate monitoring of glaciers’ extents and their dynamics is essential for improving our understanding of the impacts of climate and environmental changes in cold regions. The satellite-based Normalized Difference Snow Index (NDSI) has been widely used for mapping snow cover and glaciers around the globe. However, mapping glaciers in snow-covered areas using existing indices remains a challenging task due to their incapabilities in separating snow, glaciers, and water. This study aimed to evaluate a new satellite-based index and apply machine learning algorithms to improve the accuracy of mapping glaciers. A new index based on satellite data from Sentinel-2 was tested, which we call the Adjusted Normalized Difference Snow Index (ANDSI). ANDSI (besides NDSI) was used with five different machine learning algorithms, namely Artificial Neural Network, C5.0 Decision Tree Algorithm, Naive Bayes classifier, Support Vector Machine, and Extreme Gradient Boosting, to map glaciers, and their performance was evaluated against ground reference data. Four glacierized regions in different countries (Canada, China, Sweden, and Switzerland-Italy) were selected as study sites to evaluate the performance of the proposed ANDSI. Results showed that the proposed ANDSI outperformed the original NDSI, and the C5.0 classifier showed the best overall accuracy and Kappa among the selected five machine learning classifiers in the majority of cases. The original NDSI yielded results with an average overall accuracy of (around) 91% and the proposed ANDSI with (around) 95% for glacier mapping across all models and study regions. This study demonstrates that the proposed ANDSI serves as a superior and improved method for accurately mapping glaciers in cold regions.","PeriodicalId":55091,"journal":{"name":"GIScience & Remote Sensing","volume":"5 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-09-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135060883","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Agricultural drought (AD) is a serious threat to food security for many regions worldwide. Understanding the dynamics of AD contributes to preventing or mitigating its adverse impacts. Soil moisture (SM) anomaly is a relatively straightforward indicator of AD. However, most of the previous studies on AD dynamics of China were conducted with non-remotely sensed SM indicators due to the lack of long-term and spatial-continuous SM datasets. Here, such an SM dataset was created by enhancing a satellite remote sensing SM dataset with a machine learning method XGBoost, various remote sensing datasets, and some surface or meteorological parameters from reanalysis data. The new SM dataset has a period of 1982–2020, a spatial resolution of 0.25°, and a temporal resolution of 1 month. Furthermore, Standardized SM Index at one-month scale (SSMI1) was calculated, and AD events were identified using the SSMI1 and a 3-dimensional clustering method. Results demonstrated that 1) the new SM presented comparable or even better performances with the original SM as evaluated with spatial distributions, in-situ SM observations, and manufactured data gaps. 2) The AD was most frequent in North China, followed by the western parts of East China, Northeast, and Southwest China. The centroids of identified AD events were found chiefly in the Northeast, North, Southwest, and western parts of East China. 3) The severity of AD events presented a decreasing trend from 1982 to 2020, while significant drying trends were found mostly in the southern parts of North China, western parts of East China, and Southwest China. 4) The AD dynamics revealed in this study are basically consistent with other studies but also have unique features such as more space details and less drought frequency and count than that of meteorological drought. Further studies are expected to create a long-term satellite SM with faster timeliness, higher resolution, and greater depth.
{"title":"Agricultural drought dynamics in China during 1982–2020: a depiction with satellite remotely sensed soil moisture","authors":"Hao Sun, Qian Xu, Yunjia Wang, Zhiyu Zhao, Xiaohan Zhang, Hao Liu, Jinhua Gao","doi":"10.1080/15481603.2023.2257469","DOIUrl":"https://doi.org/10.1080/15481603.2023.2257469","url":null,"abstract":"Agricultural drought (AD) is a serious threat to food security for many regions worldwide. Understanding the dynamics of AD contributes to preventing or mitigating its adverse impacts. Soil moisture (SM) anomaly is a relatively straightforward indicator of AD. However, most of the previous studies on AD dynamics of China were conducted with non-remotely sensed SM indicators due to the lack of long-term and spatial-continuous SM datasets. Here, such an SM dataset was created by enhancing a satellite remote sensing SM dataset with a machine learning method XGBoost, various remote sensing datasets, and some surface or meteorological parameters from reanalysis data. The new SM dataset has a period of 1982–2020, a spatial resolution of 0.25°, and a temporal resolution of 1 month. Furthermore, Standardized SM Index at one-month scale (SSMI1) was calculated, and AD events were identified using the SSMI1 and a 3-dimensional clustering method. Results demonstrated that 1) the new SM presented comparable or even better performances with the original SM as evaluated with spatial distributions, in-situ SM observations, and manufactured data gaps. 2) The AD was most frequent in North China, followed by the western parts of East China, Northeast, and Southwest China. The centroids of identified AD events were found chiefly in the Northeast, North, Southwest, and western parts of East China. 3) The severity of AD events presented a decreasing trend from 1982 to 2020, while significant drying trends were found mostly in the southern parts of North China, western parts of East China, and Southwest China. 4) The AD dynamics revealed in this study are basically consistent with other studies but also have unique features such as more space details and less drought frequency and count than that of meteorological drought. Further studies are expected to create a long-term satellite SM with faster timeliness, higher resolution, and greater depth.","PeriodicalId":55091,"journal":{"name":"GIScience & Remote Sensing","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-09-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135014172","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Optimizing color consistency across multiple images is a crucial step in creating accurate digital orthophoto maps (DOMs). However, current color balance methods that rely on a reference image are susceptible to cloud and cloud shadow interference, making it challenging to ensure color fidelity and a uniform color transition between images. To address these issues, an improved method for color consistency optimization has been proposed to enhance image quality using optimized low-resolution reference images. Initially, the original image is utilized to reconstruct areas affected by clouds or cloud shadows on the reference image. For seamless cloning, a Poisson blending algorithm is employed to minimize color differences between reconstructed and other regions. Subsequently, based on a weighting approach, the high-frequency information obtained through Gaussian and bilateral filtering is superimposed to smooth the image boundary and ensure color continuity between images. Finally, local linear models are constructed to correct image color based on the optimized reference and down-sampled images. To validate the robustness of this approach, we tested it on two challenging datasets covering a wide area. Compared to state-of-the-art methods, our approach offers significant advantages in both quantitative indicators and visual quality.
{"title":"An improved color consistency optimization method based on the reference image contaminated by clouds","authors":"Zhonghua Hong, Changyou Xu, Xiaohua Tong, Shijie Liu, Ruyan Zhou, Haiyan Pan, Yun Zhang, Yanling Han, Jing Wang, Shuhu Yang","doi":"10.1080/15481603.2023.2259559","DOIUrl":"https://doi.org/10.1080/15481603.2023.2259559","url":null,"abstract":"Optimizing color consistency across multiple images is a crucial step in creating accurate digital orthophoto maps (DOMs). However, current color balance methods that rely on a reference image are susceptible to cloud and cloud shadow interference, making it challenging to ensure color fidelity and a uniform color transition between images. To address these issues, an improved method for color consistency optimization has been proposed to enhance image quality using optimized low-resolution reference images. Initially, the original image is utilized to reconstruct areas affected by clouds or cloud shadows on the reference image. For seamless cloning, a Poisson blending algorithm is employed to minimize color differences between reconstructed and other regions. Subsequently, based on a weighting approach, the high-frequency information obtained through Gaussian and bilateral filtering is superimposed to smooth the image boundary and ensure color continuity between images. Finally, local linear models are constructed to correct image color based on the optimized reference and down-sampled images. To validate the robustness of this approach, we tested it on two challenging datasets covering a wide area. Compared to state-of-the-art methods, our approach offers significant advantages in both quantitative indicators and visual quality.","PeriodicalId":55091,"journal":{"name":"GIScience & Remote Sensing","volume":"1 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-09-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135014454","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-09-19DOI: 10.1080/15481603.2023.2257470
Zhengdong Wang, Bingfang Wu, Miao Zhang, Zonghan Ma
Monitoring the distribution and area change of biological soil crusts (BSCs) can enhance our understanding of the interactions between nonvascular plants and the environment in drylands. However, using only pixel-based binary classification methods results in large-area estimation errors at large scales. The lack of available calculation methods for directly measuring BSC coverage using multispectral satellite images makes it challenging to obtain BSC area data for further studies at large scales. To address these issues, this study developed feature space conceptual models for desert and sandy land based on the characteristics of BSC in drylands. The desert feature space comprised the normalized difference vegetation index (NDVI) combined with the brightness index (BI), encompassing moss, lichen, and non-BSC. The sandy land feature space relied on the biological soil crust index (BSCI) and the NDVI, including vegetation, mixed BSCs and sandy soil. Using Sentinel-2 satellite imagery and a spectral unmixing model, the abundance of BSCs was quantified in four BSC growth areas located in the Gurbantunggut Desert and Mu Us Sandy Land of China. Validation of the method indicated that the root mean square error (RMSE) of the BSC coverage estimation results was 10% and 8% in desert and sandy land, respectively (estimation accuracies of 79% and 81%, respectively). This demonstrated that the proposed method can effectively estimate BSC coverage at a subpixel scale. The resulting BSC coverage data can provide the possibility to evaluate the functions of regional ecosystems.
{"title":"An advanced coverage estimation method to quantify biological soil crust coverage using Sentinel-2 imagery in desert and sandy land of China","authors":"Zhengdong Wang, Bingfang Wu, Miao Zhang, Zonghan Ma","doi":"10.1080/15481603.2023.2257470","DOIUrl":"https://doi.org/10.1080/15481603.2023.2257470","url":null,"abstract":"Monitoring the distribution and area change of biological soil crusts (BSCs) can enhance our understanding of the interactions between nonvascular plants and the environment in drylands. However, using only pixel-based binary classification methods results in large-area estimation errors at large scales. The lack of available calculation methods for directly measuring BSC coverage using multispectral satellite images makes it challenging to obtain BSC area data for further studies at large scales. To address these issues, this study developed feature space conceptual models for desert and sandy land based on the characteristics of BSC in drylands. The desert feature space comprised the normalized difference vegetation index (NDVI) combined with the brightness index (BI), encompassing moss, lichen, and non-BSC. The sandy land feature space relied on the biological soil crust index (BSCI) and the NDVI, including vegetation, mixed BSCs and sandy soil. Using Sentinel-2 satellite imagery and a spectral unmixing model, the abundance of BSCs was quantified in four BSC growth areas located in the Gurbantunggut Desert and Mu Us Sandy Land of China. Validation of the method indicated that the root mean square error (RMSE) of the BSC coverage estimation results was 10% and 8% in desert and sandy land, respectively (estimation accuracies of 79% and 81%, respectively). This demonstrated that the proposed method can effectively estimate BSC coverage at a subpixel scale. The resulting BSC coverage data can provide the possibility to evaluate the functions of regional ecosystems.","PeriodicalId":55091,"journal":{"name":"GIScience & Remote Sensing","volume":"52 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-09-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135015040","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-09-19DOI: 10.1080/15481603.2023.2257980
Qing Ding, Zhenfeng Shao, Xiao Huang, Xiaoxiao Feng, Orhan Altan, Bin Hu
Precise identification of binary building changes through remote sensing observations plays a crucial role in sustainable urban development. However, many supervised change detection (CD) methods overly rely on labeled samples, thus limiting their generalizability. In addition, existing semi-supervised CD methods suffer from instability, complexity, and limited applicability. To overcome these challenges and fully utilize unlabeled samples, we proposed a consistency-guided lightweight semi-supervised binary change detection method (Semi-LCD). We designed a lightweight dual-branch CD network to extract image features while reducing model size and complexity. Semi-LCD fully exploits unlabeled samples by data augmentation, consistency regularization, and pseudo-labeling, thereby enhancing its detection performance and generalization capability. To validate the effectiveness and superior performance of Semi-LCD, we conducted experiments on three building CD datasets. Detection results indicate that Semi-LCD outperforms competing methods, quantitatively and qualitatively, achieving the optimal balance between performance and model size. Furthermore, ablation experiments validate the robustness and advantages of the Semi-LCD in effectively utilizing unlabeled samples.
{"title":"Consistency-guided lightweight network for semi-supervised binary change detection of buildings in remote sensing images","authors":"Qing Ding, Zhenfeng Shao, Xiao Huang, Xiaoxiao Feng, Orhan Altan, Bin Hu","doi":"10.1080/15481603.2023.2257980","DOIUrl":"https://doi.org/10.1080/15481603.2023.2257980","url":null,"abstract":"Precise identification of binary building changes through remote sensing observations plays a crucial role in sustainable urban development. However, many supervised change detection (CD) methods overly rely on labeled samples, thus limiting their generalizability. In addition, existing semi-supervised CD methods suffer from instability, complexity, and limited applicability. To overcome these challenges and fully utilize unlabeled samples, we proposed a consistency-guided lightweight semi-supervised binary change detection method (Semi-LCD). We designed a lightweight dual-branch CD network to extract image features while reducing model size and complexity. Semi-LCD fully exploits unlabeled samples by data augmentation, consistency regularization, and pseudo-labeling, thereby enhancing its detection performance and generalization capability. To validate the effectiveness and superior performance of Semi-LCD, we conducted experiments on three building CD datasets. Detection results indicate that Semi-LCD outperforms competing methods, quantitatively and qualitatively, achieving the optimal balance between performance and model size. Furthermore, ablation experiments validate the robustness and advantages of the Semi-LCD in effectively utilizing unlabeled samples.","PeriodicalId":55091,"journal":{"name":"GIScience & Remote Sensing","volume":"26 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-09-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135014576","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2023-03-01DOI: 10.1080/15481603.2023.2181143
James Wickham, Stephen V. Stehman, Daniel G. Sorenson, Leila Gass, Jon A. Dewitz
The National Land Cover Database (NLCD), a product suite produced through the MultiResolution Land Characteristics (MRLC) consortium, is an operational land cover monitoring program. Starting from a base year of 2001, NLCD releases a land cover database every 2–3-years. The recent release of NLCD2019 extends the database to 18 years. We implemented a stratified random sample to collect land cover reference data for the 2016 and 2019 components of the NLCD2019 database at Level II and Level I of the classification hierarchy. For both dates, Level II land cover overall accuracies (OA) were 77.5% ± 1% (± value is the standard error) when agreement was defined as a match between the map label and primary reference label only, and increased to 87.1% ± 0.7% when agreement was defined as a match between the map label and either the primary or alternate reference label. At Level I of the classification hierarchy, land cover OA was 83.1% ± 0.9% for both 2016 and 2019 when agreement was defined as a match between the map label and primary reference label only, and increased to 90.3% ± 0.7% when agreement also included the alternate reference label. The Level II and Level I OA for the 2016 land cover in the NLCD2019 database were 5% higher compared to the 2016 land cover component of the NLCD2016 database when agreement was defined as a match between the map label and primary reference label only. No improvement was realized by the NLCD2019 database when agreement also included the alternate reference label. User’s accuracies (UA) for forest loss and grass gain were>70% when agreement included either the primary or alternate label, and UA was generally<50% for all other change themes. Producer’s accuracies (PA) were>70% for grass loss and gain and water gain and generally<50% for the other change themes. We conducted a post-analysis review for map-reference agreement to identify patterns of disagreement, and these findings are discussed in the context of potential adjustments to mapping and reference data collection procedures that may lead to improved map accuracy going forward.
{"title":"Thematic accuracy assessment of the NLCD 2019 land cover for the conterminous United States","authors":"James Wickham, Stephen V. Stehman, Daniel G. Sorenson, Leila Gass, Jon A. Dewitz","doi":"10.1080/15481603.2023.2181143","DOIUrl":"https://doi.org/10.1080/15481603.2023.2181143","url":null,"abstract":"The National Land Cover Database (NLCD), a product suite produced through the MultiResolution Land Characteristics (MRLC) consortium, is an operational land cover monitoring program. Starting from a base year of 2001, NLCD releases a land cover database every 2–3-years. The recent release of NLCD2019 extends the database to 18 years. We implemented a stratified random sample to collect land cover reference data for the 2016 and 2019 components of the NLCD2019 database at Level II and Level I of the classification hierarchy. For both dates, Level II land cover overall accuracies (OA) were 77.5% ± 1% (± value is the standard error) when agreement was defined as a match between the map label and primary reference label only, and increased to 87.1% ± 0.7% when agreement was defined as a match between the map label and either the primary or alternate reference label. At Level I of the classification hierarchy, land cover OA was 83.1% ± 0.9% for both 2016 and 2019 when agreement was defined as a match between the map label and primary reference label only, and increased to 90.3% ± 0.7% when agreement also included the alternate reference label. The Level II and Level I OA for the 2016 land cover in the NLCD2019 database were 5% higher compared to the 2016 land cover component of the NLCD2016 database when agreement was defined as a match between the map label and primary reference label only. No improvement was realized by the NLCD2019 database when agreement also included the alternate reference label. User’s accuracies (UA) for forest loss and grass gain were>70% when agreement included either the primary or alternate label, and UA was generally<50% for all other change themes. Producer’s accuracies (PA) were>70% for grass loss and gain and water gain and generally<50% for the other change themes. We conducted a post-analysis review for map-reference agreement to identify patterns of disagreement, and these findings are discussed in the context of potential adjustments to mapping and reference data collection procedures that may lead to improved map accuracy going forward.","PeriodicalId":55091,"journal":{"name":"GIScience & Remote Sensing","volume":"84 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2023-03-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"135479570","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-12-31DOI: 10.1080/15481603.2022.2112391
Yichen Lu, T. James, C. Schillaci, Aldo Lipani
ABSTRACT Accurately monitoring the variation of snow cover from remote sensing is vital since it assists in various fields including prediction of floods, control of runoff values, and the ice regime of rivers. Spectral indices methods are traditional ways to realize snow segmentation, including the most common one – the Normalized Difference Snow Index (NDSI), which utilizes the combination of green and short-wave infrared (SWIR) bands. In addition, spectral indices methods heavily depend on the optimal threshold to determine the accuracy, making it time-consuming to find optimal values for different places. Convolutional neural networks ensemble model with DeepLabV3+ was employed as sub-models for snow segmentation using (Sentinel-2), which aims to distinguish clouds and water body from snow. The imagery dataset generated in this article contains sites in global alpine regions such as Tibetan Plateau in China, the Alps in Switzerland, Alaska in the United States, Southern Patagonian Icefield in Chile, Tsylos Provincial Park, Tatsamenie Peak, and Dalton Peak in Canada. To overcome the limitation of DeepLabV3+, which only accepts three channels as input features, and the need to use six features: green, red, blue, near-infraRed, SWIR, and NDSI, 20 three-channel DeepLabV3+ sub-models, were constructed with different combinations of three features and then ensembled together. The proposed ensemble model showed superior performance than benchmark spectral indices method, with mIoU values ranging from 0.8075 to 0.9538 in different test sites. The results of this project contribute to the development of automated snow segmentation tools to assist earth observation applications.
{"title":"Snow detection in alpine regions with Convolutional Neural Networks: discriminating snow from cold clouds and water body","authors":"Yichen Lu, T. James, C. Schillaci, Aldo Lipani","doi":"10.1080/15481603.2022.2112391","DOIUrl":"https://doi.org/10.1080/15481603.2022.2112391","url":null,"abstract":"ABSTRACT Accurately monitoring the variation of snow cover from remote sensing is vital since it assists in various fields including prediction of floods, control of runoff values, and the ice regime of rivers. Spectral indices methods are traditional ways to realize snow segmentation, including the most common one – the Normalized Difference Snow Index (NDSI), which utilizes the combination of green and short-wave infrared (SWIR) bands. In addition, spectral indices methods heavily depend on the optimal threshold to determine the accuracy, making it time-consuming to find optimal values for different places. Convolutional neural networks ensemble model with DeepLabV3+ was employed as sub-models for snow segmentation using (Sentinel-2), which aims to distinguish clouds and water body from snow. The imagery dataset generated in this article contains sites in global alpine regions such as Tibetan Plateau in China, the Alps in Switzerland, Alaska in the United States, Southern Patagonian Icefield in Chile, Tsylos Provincial Park, Tatsamenie Peak, and Dalton Peak in Canada. To overcome the limitation of DeepLabV3+, which only accepts three channels as input features, and the need to use six features: green, red, blue, near-infraRed, SWIR, and NDSI, 20 three-channel DeepLabV3+ sub-models, were constructed with different combinations of three features and then ensembled together. The proposed ensemble model showed superior performance than benchmark spectral indices method, with mIoU values ranging from 0.8075 to 0.9538 in different test sites. The results of this project contribute to the development of automated snow segmentation tools to assist earth observation applications.","PeriodicalId":55091,"journal":{"name":"GIScience & Remote Sensing","volume":"59 1","pages":"1321 - 1343"},"PeriodicalIF":6.7,"publicationDate":"2022-12-31","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"44868738","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-12-13DOI: 10.1080/15481603.2022.2158521
Wenting Wu, Chao Zhi, C. Chen, B. Tian, Zuoqi Chen, Hua Su
ABSTRACT Intertidal vegetation plays an essential role in habitat provision for waterbirds but suffers great losses due to human activities. However, it is challenging in tracking the human-driven loss and degradation of intertidal vegetation due to rapid urbanization in a high temporal resolution. In this study, a methodological framework based on full Landsat time-series (FLTS) is proposed to detect the year of change (YOC) of intertidal vegetation converted to impervious surfaces (ISs) and artificial ponds (APs), and the condition of the remaining intertidal vegetation was also assessed by FLTS, in the Fujian province, a subtropical coastal area lying in southeast China. The accuracies of the YOC detection of intertidal vegetation converted to IS and AP were 91.84% and 72.73%, with mean absolute errors of 0.26 and 1.06, respectively. The total areas of intertidal vegetation encroached by IS and AP were 31.68 km2 and 23.85 km2, respectively. Most ISs were developed later than 2010, and most APs were developed earlier than 2005, which are highly related to the implementation of local policies for economic development. The remaining intertidal vegetation in growing, stable, and degraded conditions were 43.05%, 56.38%, and 0.57%, respectively. The results indicated that areas of intertidal vegetation were reclaimed for anthropogenic uses at a considerable rate, although the intertidal vegetation still increased owing to natural development after the establishment of natural reserves. The study demonstrates that the FLTS has capacities in monitoring the dynamics in coastal zones solely for its dense earth observations.
{"title":"Detecting annual anthropogenic encroachment on intertidal vegetation using full Landsat time-series in Fujian, China","authors":"Wenting Wu, Chao Zhi, C. Chen, B. Tian, Zuoqi Chen, Hua Su","doi":"10.1080/15481603.2022.2158521","DOIUrl":"https://doi.org/10.1080/15481603.2022.2158521","url":null,"abstract":"ABSTRACT Intertidal vegetation plays an essential role in habitat provision for waterbirds but suffers great losses due to human activities. However, it is challenging in tracking the human-driven loss and degradation of intertidal vegetation due to rapid urbanization in a high temporal resolution. In this study, a methodological framework based on full Landsat time-series (FLTS) is proposed to detect the year of change (YOC) of intertidal vegetation converted to impervious surfaces (ISs) and artificial ponds (APs), and the condition of the remaining intertidal vegetation was also assessed by FLTS, in the Fujian province, a subtropical coastal area lying in southeast China. The accuracies of the YOC detection of intertidal vegetation converted to IS and AP were 91.84% and 72.73%, with mean absolute errors of 0.26 and 1.06, respectively. The total areas of intertidal vegetation encroached by IS and AP were 31.68 km2 and 23.85 km2, respectively. Most ISs were developed later than 2010, and most APs were developed earlier than 2005, which are highly related to the implementation of local policies for economic development. The remaining intertidal vegetation in growing, stable, and degraded conditions were 43.05%, 56.38%, and 0.57%, respectively. The results indicated that areas of intertidal vegetation were reclaimed for anthropogenic uses at a considerable rate, although the intertidal vegetation still increased owing to natural development after the establishment of natural reserves. The study demonstrates that the FLTS has capacities in monitoring the dynamics in coastal zones solely for its dense earth observations.","PeriodicalId":55091,"journal":{"name":"GIScience & Remote Sensing","volume":"59 1","pages":"2266 - 2282"},"PeriodicalIF":6.7,"publicationDate":"2022-12-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"46916293","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":2,"RegionCategory":"地球科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：