最新刊期

    28 11 2024
    封面故事

      Frontier Progress

    • 在遥感图像解译领域,专家总结了知识与数据双驱动的新方向,为智能解译提供新思路。
      MENG Yu,CHEN Jingbo,ZHANG Zheng,LIU Zhiqiang,ZHAO Zhitao,HUO Lianzhi,SHI Keli,LIU Diyou,DENG Yupeng,TANG Ping
      Vol. 28, Issue 11, Pages: 2698-2718(2024) DOI: 10.11834/jrs.20243547
      Knowledge and data driven remote sensing image interpretation: Recent developments and prospects
      摘要:Knowledge and data are the two main elements that have characterized the development of remote sensing image interpretation for decades. With the continuous enrichment of sensor platforms and rapid breakthroughs in deep learning, big data, multi-modal, and long time-series methodologies, data-driven intelligent remote sensing image interpretation has become a hot research direction in recent years. However, in the deepening and expanding research and applications, the limitations of data-driven methods such as difficult reuse between different scenarios, strong training sample dependence, and weak interpretability are beginning to emerge. Various types of knowledge accumulated in the long-term remote sensing image interpretation practice have the characteristics of objective reality, certainty, scene adaptability, interpretability, etc., which can be complemented with data-driven approaches, and the dual-driven of knowledge and data is becoming a new direction of remote sensing image interpretation. This paper first reviews the major stages in the development of remote sensing image interpretation and the respective roles of knowledge and data in each of these stages. Then the main types of knowledge involved in remote sensing image interpretation are summarized and categorized into fourteen types. The fusion of knowledge and deep learning is an important path to achieve the dual-drive of knowledge and data, and this paper summarizes five categories and fifteen subcategories of knowledge and deep neural network fusion methods with relevant cases. From the perspective of knowledge types, this paper further provides an overview of existing applications of remote sensing interpretation with joint knowledge and data. The effectiveness and capability increment of fusing knowledge and data is demonstrated by the analyses of typical examples. Lastly, this paper gives a systematic prospect on the framework and key techniques for knowledge and data compound driven remote sensing image interpretation.  
      关键词:remote sensing image interpretation;knowledge driven;data driven;artificial intelligence;knowledge graphs;deep learning;natural resources;review   
      1806
      |
      1334
      |
      0
      <HTML>
      <L-PDF><Enhanced-PDF><Meta-XML>
      <引用本文> <批量引用> 55245150 false
      发布时间:2024-12-30
    • 在河流断面形态提取领域,专家系统梳理了近20年研究进展,提出了适用于缺资料地区或大范围流域断面形态提取的“空—天—地”一体化观测方案,为河流断面形态获取提供了新方向。
      XUE Yuan,QIN Chao,XU Mengzhen,FU Xudong,LI Dan,WU Baosheng,WANG Guangqian
      Vol. 28, Issue 11, Pages: 2719-2738(2024) DOI: 10.11834/jrs.20243315
      Progress and prospects in river cross section extraction based on multi-source multisource remote sensing
      摘要:Natural rivers carry water and materials within a certain boundary geometry. Research on rivers oftenfrequently involves extracting the geometric information of river surfaces and boundaries or hydraulic characteristics, such as flow velocity and discharge. Among these hydraulic attributes, geometric data that pertaining to river cross sections and other river features, which are easier to observe than the dynamic flow characteristics, are indispensable for conducting research on hydrological processes and material fluxes within a river system. Traditionally, the extraction of such data has relied heavily relied on field measurements, posing challenges in obtaining data for from inaccessible areas, such as mountainous regions, canyons, disaster-prone regionsareas, or expansive river basins. With the continuous advancement of multi-source remote sensing technology, which encompassing includes underwater remote sensing, near-earth Earthremote sensing, and satellite remote sensing, it has become possible to addressing the data scarcity in mountainous regions, canyons, and other areas has become possible by integrating multi-source remote sensing observations with limited ground measurements and establishing automatic extraction methods. Building upon the advancements made in the extraction of river cross section morphology over the past two decades, this paper study examines the strengths and limitations of current methods. This studyIt presented presents an integrated “air-space-ground” remote sensing data observation scheme, amalgamatedcombined with the corresponding automatic extraction methodologies, such as river surfaces extraction method, river width, extraction method and river water level extraction methods, to extract river information, particularly cross section morphology, in data-scarce or large-scale river basins. Furthermore, this study offered offers valuable insights into the future development trends by considering the existing technical progress in the field.  
      关键词:Cross section morphology;multi-source remote sensing;integrated “air-space-ground” observation;Automated extraction   
      1071
      |
      1008
      |
      0
      <HTML>
      <L-PDF><Enhanced-PDF><Meta-XML>
      <引用本文> <批量引用> 50080283 false
      发布时间:2024-12-30
    • 高分辨率遥感图像场景分类研究取得新成果,为土地监测、环境保护等领域提供解决方案。
      LI Zhi,GAO Lianru,ZHENG Ke,NI Li
      Vol. 28, Issue 11, Pages: 2739-2760(2024) DOI: 10.11834/jrs.20243519
      Research progress of high-resolution remote sensing image scene classification
      摘要:With the rapid advancement of remote sensing technology, the resolution of remote sensing satellites is improving, the number of spectral bands is increasing, and revisit periods are contracting. This progression empowers researchers to access more valuable data and information from remote sensing images. Concepts, such as remote sensing big data, remote sensing foundation models, and smart cities, have successively emerged in recent years, imposing increased demands on the intelligent extraction technology of massive remote sensing data, particularly regarding remote sensing image information.As an indispensable element of intelligent information extraction technology applied in fields, such as land use and cover, national land resource surveys, natural disaster observation, agricultural yield estimation, and forestry protection, remote sensing image classification exhibits substantial practical importance. Remote sensing image scene classification has been introduced in this context. The objective of scene classification in remote sensing images is to comprehensively and semantically categorize each given remote sensing image. This task entails summarizing and analyzing the extracted feature information at a high level and assigning different labels to areas of interest based on their features.In contrast with natural images, although they contain features, such as color, texture, and shape, remote sensing images encounter more challenges in classification due to the intricate scene content resulting from the overhead perspective, weak texture, and color information caused by low resolution. Nevertheless, as one of the technical means in remote sensing applications, remote sensing image scene classification technology plays a pivotal role in the development of practical application technologies.After years of development, numerous comprehensive review studies on remote sensing image scene classification have been conducted locally and abroad. However, the recent surge in remote sensing big data has introduced new challenges into scene classification. The ongoing evolution of deep learning technology, particularly the widespread application of Convolutional Neural Networks (CNNs) and transformers, has resulted in significant advancements in remote sensing image scene classification. In this context, self-supervised learning, as a method that is independent of annotated data, has become indispensable in the field of remote sensing image scene classification. Foundation models based on self-supervised learning have been successfully implemented in scene classification, presenting innovative solutions to this field. As the volume of remote sensing data continues to increase, the dataset scale for remote sensing image scene classification is expanding rapidly, giving rise to increasingly intricate classification tasks. Remote sensing image scene classification datasets are swiftly progressing toward the integration of multiple sources, the incorporation of multiple labels, and the inclusion of large-scale samples.Drawing from the findings of the current literature survey, this study systematically compiles a summary of deep learning methods within the domain of remote sensing image scene classification. Encompassing CNNs, visual transformers, and generative adversarial networks, this overview also introduces representative datasets and foundation models since the inception of scene classification. Several classical scene classification methods have undergone evaluation across various benchmark datasets. In addition, this study delves into primary challenges and prospects, paving the way for further research in the classification of scenes in remote sensing images.  
      关键词:high-resolution remote sensing image;image classification;scene classification;deep learning   
      229
      |
      115
      |
      0
      <HTML>
      <L-PDF><Enhanced-PDF><Meta-XML>
      <引用本文> <批量引用> 78379994 false
      发布时间:2024-12-30
    • 海滩视频监测技术为海岸带资源环境监测提供新方案,推动中国海岸动力地貌学精细化研究。
      QI Hongshuai,YIN Hang,CAI Feng,ZHANG Chi,LIU Gen,CAO Zhubin
      Vol. 28, Issue 11, Pages: 2761-2779(2024) DOI: 10.11834/jrs.20243311
      Advances and prospects of the a shore-based video monitoring technology on beach research
      摘要:The coastal zone, which is positioned at the dynamic interface of between land and sea, stands out as one of the most vibrant and crucial natural regions on Earth. Subject to the interactive forces of land, sea, and atmosphere, it's the geomorphic and ecological systems of the coastal zone are exceptionally delicatefragile. Beaches, which are characterized by the unique scenic attributes and susceptibility to coastal landform changes, act as the frontline defense against ocean hydrodynamics forces and serve as guardians of the coastal ecological environment. Simultaneously, they play a pivotal role in fostering green coastal economies. Strengthening the monitoring and research of beaches has become an imperative requirement for achieving harmonious coexistence between humans and nature. Traditionally, beach monitoring has predominantly relied on labor-intensive field measurements, resulting in time-consuming efforts and limited spatiotemporal resolution in data sampling. The current bottleneck in beach research lies in the challenge of acquiring multidimensional, high-quality, and long-duration datasets. Shore-based video monitoring, which is an optical remote sensing technology designed for coastal process observation and quantitative data extraction, provides a new and effective solution for continuous beach monitoring. This technology, born which was developed in the United States in the 1980 s with Argus as its paradigmatic representative, has proliferated globally over the past four decades. At present, It it is now universally acknowledged as a potent tool in the field of nearshore observation. While theAlthough research on shore-based video monitoring technology started relatively late in China, it has developed rapidly. In particular, the joint efforts of the Institute of Third Institute of Oceanography of China, Hohai University, and Xiamen University led to the independent development of the Coastal Shore-based Video Imagery Monitoring SystemCOSVIMS. The progress of shoreShore-based video monitoring technology has witnessed achieved significant progressstrides. Algorithmic advancements related to sandy coastlines, intertidal zone topography, nearshore wave propagation & and dissipation processes, nearshore currents, and bathymetry extraction have played a crucial roles in enhancing the capabilities of this technology. These algorithms not only ensure accuracy but also enable a more comprehensive understanding of coastal dynamics. The applicability of video monitoring technology spans covers various scenarios, including beach management, coastal engineering construction and assessment, storm surge process observation, and aeolian sand transport process monitoring. Looking ahead, the future Future trends of shore-based video monitoring technology are promising, especiallyparticularly in the context of the accelerating coastal urbanization in China. The contradictions Contradictions between the development and protection of coastal zone resources and the environment are becoming increasingly pronounced. The widespread adoption of shore-based video monitoring technology holds exhibits high practical significance and value for the research and protecting protection the of coastal zones along China. Breakthroughs in emerging technologies, such as artificial intelligence, 5G, big data, and Cloudcloud, coupled with the growing demand for coastal zone resource and environmental monitoring, are paving the way for greater development opportunities for shore-based video monitoring technology. In conclusion, the evolution of shore-based video monitoring technology signifies a transformative era in coastal research and management. As technology continues to advance, it opens new frontiers for understanding and preserving the delicate balance between coastal development and environmental conservation are opened. The journey development from direct measurements to advanced remote sensing exemplifies the evolution toward more efficient and sustainable coastal research methodologies.  
      关键词:video image;beach monitoring;Argus;beach nourishment;rip currents;storm surge;aeolian transport   
      361
      |
      747
      |
      0
      <HTML>
      <L-PDF><Enhanced-PDF><Meta-XML>
      <引用本文> <批量引用> 50079019 false
      发布时间:2024-12-30

      Data Articles

    • 在城市信息管理与防灾减灾领域,专家建立了高分辨率多光谱遥感影像细粒度建筑物特征集MFBFS,为建筑物结构细粒度提取研究提供数据支持。
      WANG Zhenqing,ZHOU Yi,WANG Futao,WANG Shixin,GAO Guorui,ZHU Jinfeng,WANG Ping,HU Kailong
      Vol. 28, Issue 11, Pages: 2780-2791(2024) DOI: 10.11834/jrs.20243526
      MFBFS: A fine-grained building feature set for high-resolution multispectral remote sensing images
      摘要:Building information extraction from remote sensing images plays an essential role in urban information management and disaster prevention and mitigation. This study establishes a fine-grained building feature set, namely, MFBFS, for high-resolution multispectral remote sensing images. MFBFS uses the domestically produced Gaofen-2 multispectral remote sensing images as data source and selects 21 districts and counties with concentrated buildings in various disaster zones in China, covering 3668 km2 as the study area. These regions include Yongjia County, Xuwen County, and Wanning City in the southeastern coastal disaster belt; Ning'an City, Kaiyuan City, Laiyuan County, Shouguang City, Xinxiang County, Lujiang County, Hengdong County, and Songbei District in the eastern disaster belt; Daning County, Enshi City, Tengchong City, and Shuicheng County in the central disaster belt; Kashgar City, Yizhou District, and Pingluo County in the northwest disaster belt; and Diebu County, Yushu City, and Milin County in the Qinghai–Tibetan disaster belt. To obtain high-quality and high-resolution remote sensing images, a series of preprocessing procedures was applied to the Gaofen-2 images. Initially, poor-quality images were removed, followed by radiometric and orthorectification corrections on multispectral and panchromatic images, respectively. Finally, the panchromatic images were fused to enhance the spatial resolution of the multispectral images, resulting in a spatial resolution of 0.8 m. Seventeen feature components were generated from four perspectives: spectral, texture, edge, and index. Spectral features include features from the blue, green, red, and near-infrared bands. Texture features consist of contrast, dissimilarity, homogeneity, correlation, angular second moment, local binary pattern, and histogram of oriented gradients. Edge features comprise first-order and multi-order edge characteristics. Index features include building, shadow, vegetation, and water indexes. MFBFS encompasses over 260000 building instances, ensuring high intra-class diversity in terms of size, shape, color, orientation, background, and structural type. These instances are classified into three structural types, namely, steel and reinforced concrete, masonry, and block stone structures, significantly reflecting the abilities of buildings to resist disasters and their usable lifespans. The fine-grained design will cause the task of extracting buildings through remote sensing to play a greater role, particularly in pre-disaster loss prediction and post-disaster loss assessment in the disaster field. Rigorous quality control processes and field inspections were conducted to ensure the high accuracy of ground truth values. This procedure involved adherence to interpretation standards and inviting data inspectors and remote sensing image experts to assess the quality of remote sensing images and corresponding ground truth values. Ultimately, 191 GB of high-quality feature and label data were obtained. Each of the 17 feature components comprises 11005 512×512-sized feature maps with a spatial resolution of 0.8 m, uniformly expanded to a value range of [0,1]. Initial deep learning experiments demonstrate the effectiveness of MFBFS. This feature set, available for download athttps://github.com/WangZhenqing-RS/MFBFS, provides robust data support for fine-grained building structure extraction research and promotes the development of domestic high-resolution remote sensing data applications.  
      关键词:High-resolution remote sensing;multispectral imagery;fine-grained categories;building extraction;feature sets   
      181
      |
      152
      |
      0
      <HTML>
      <L-PDF><Enhanced-PDF><Meta-XML>
      <引用本文> <批量引用> 61669457 false
      发布时间:2024-12-30
    • 最新研究揭示,中国沿海围填海面积急剧减少,退围还海、退围还湿措施显著提速,为海岸带保护和生态治理提供重要数据支持。
      XU Jinyong,WANG Xiao,ZUO Lijun,ZHANG Weiwei,YI Ling,LIU Fang,HU Shunguang,SUN Feifei,ZHANG Zengxiang
      Vol. 28, Issue 11, Pages: 2792-2800(2024) DOI: 10.11834/jrs.20243475
      Dataset of remote sensing dynamic monitoring of coastal reclamation in China (2010—2020)
      摘要:Reclamation is an important cause of coastline changes, coastal wetland degradation, and offshore marine pollution. The control and management of reclamation are related to the protection of the national coastal zone and the construction of an ecological civilization. At present, high-frequency remote sensing monitoring of national-scale coastal reclamation types and their spatial distribution is lacking, and the tracking and monitoring of reclamation management measures based on remote sensing methods have not yet been effectively performed. By using the method of integrated remote sensing dynamic monitoring of the coastline and reclamation, and based on Landsat time-series satellite images, the spatial distributions of China’s coastal reclamation in the periods of 2010—2015, 2015—2018, and 2018—2020, and the spatial distributions of the measures of returning enclosures to the sea and wetland in the corresponding periods were extracted. The data outcomes were stored in the ArcGIS Shapefile format, and the compressed data volumes totaled 680 KB. The remote sensing dynamic monitoring of national-scale time-series coastal reclamation that is synchronized with the time of coastline change and shared the same satellite image basis is highly significant for the protection of coastline resources and the evaluation of the effect of coastal ecological and environmental management. The results showed that during the second decade of the 21st century, the area of newly coastal reclamation in China declined sharply and that the growth rate of coastal reclamation was effectively controlled. Meanwhile, the rate of measures for returning enclosures to the sea and wetland, which was mostly manifested as the restoration of aquaculture pits and ponds to mudflats and sea surfaces, was increased abruptly, particularly in 2018—2020. Achievements were directly related to the unprecedented strengthening of national policies for reclamation control and coastal zone protection in 2018. The remote sensing monitoring dataset of coastal reclamation dynamics can provide basic data guarantee for national ocean and coastal zone management and scientific research, and important support for the realization of Sustainable Development Goal 14.5.  
      关键词:remote sensing;coastal reclamation;dynamic;coastal zone;China   
      106
      |
      82
      |
      0
      <HTML>
      <L-PDF><Enhanced-PDF><Meta-XML>
      <引用本文> <批量引用> 78381662 false
      发布时间:2024-12-30

      Intelligent Remote Sensing of Resources

    • JIANG Hou,YAO Ling,BAI Yongqing,ZHOU Chenghu
      Vol. 28, Issue 11, Pages: 2801-2814(2024) DOI: 10.11834/jrs.20243440
      Assessment of rooftop photovoltaic power generation potentials by using multisource remote sensing data
      摘要:Rooftop solar photovoltaic (PV) systems are becoming increasingly critical in the global shift toward sustainable energy. Despite their importance, the fragmented and small-scale spatial distribution of rooftop PV systems poses significant challenges to accurate and detailed regional potential assessments. This study aims to deal with these challenges by developing a comprehensive assessment framework that integrates multisource remote sensing data and advanced artificial intelligence algorithms. The objective is to provide a robust methodology for evaluating the potential of rooftop PV systems on a large scale.The assessment framework developed in this study leverages a combination of geostationary meteorological satellite imagery and deep learning inversion models to estimate hourly surface solar radiation. To accurately extract building outlines, high-resolution remote sensing images are processed using advanced image segmentation models. Furthermore, the framework integrates a geometric optical model to simulate the PV generation process. This holistic approach enables the precise revelation of spatial and temporal variations in solar energy resources. It also facilitates the investigation of the total available rooftop resources and the determination of PV power generation potential at meter-level resolution and hourly scales.The effectiveness of the framework was validated through a case study conducted in Jiangsu Province, China. The results demonstrated the scalability and applicability of the framework across different geographic locations and multiple temporal scales. The estimation results revealed that rooftop resources in Jiangsu Province could support a PV installed capacity of 236.25 GW, with an annual power generation potential of 303.81 TWh. This substantial output could meet 41.1% of the province’s total electricity consumption. The case study highlights the framework’s ability to provide detailed and accurate assessments of rooftop PV potential on a large scale.This study illustrates the feasibility and effectiveness of integrating multisource remote sensing observations for the spatiotemporal assessment of rooftop PV potential. The developed framework offers robust tools and technical support for advancing the transition to sustainable energy. By providing insights into the spatial and temporal variability of solar resources, this framework paves the way for the optimized utilization of rooftop PV systems. This research contributes to the broadening effort of achieving sustainable energy goals by enabling more precise and large-scale assessments of rooftop PV potential.  
      关键词:Renewable energy;rooftop photovoltaics;remote sensing image segmentation;surface solar radiation inversion;carbon reduction   
      1346
      |
      851
      |
      0
      <HTML>
      <L-PDF><Enhanced-PDF><Meta-XML>
      <引用本文> <批量引用> 68061161 false
      发布时间:2024-12-30
    • 在长江中下游地区地表水体遥感监测领域,专家构建了多要素耦合和水体精细监测方法,为地表水资源管理提供技术支撑。
      GUO Shanchuan,DU Peijun,XIA Zilong,FANG Hong,TANG Pengfei
      Vol. 28, Issue 11, Pages: 2815-2827(2024) DOI: 10.11834/jrs.20243275
      Remote sensing extraction and change analysis of surface water body over the Middle and Lower Reaches of the Yangtze River based on a multi-element coupling framework
      摘要:Under global climate change and the long-term high-intensity exploitation of resources, surface water body in the middle and lower reaches of the Yangtze River region (MLRYRR) has become more complex, extreme, and hazardous in recent decades. However, technical inadequacy in remote sensing water extraction still exists because of the complexity of regional land cover, fragmented structures, and inconsistent feature characteristics. This study aims to propose a framework that combines multisource data fusion to extract a water body under complex geographical conditions at the basin scale. Accordingly, it shows the spatiotemporal pattern of surface water body over MLRYRR.First, the spectral features of several land objects with a water-like spectrum were explored based on multispectral remote sensing images. An enhanced remote sensing water index for large-scale spatial and temporal water extraction was introduced. This index can effectively differentiate water bodies with aquatic plants and vegetation in subtropical regions. Second, the proposed index was incorporated into an automatic water extraction model through the decision-level fusion of multisource geographic information data (i.e., topography, hydrology, and impervious surfaces). Third, considering the seasonal variation of surface water bodies, a frequency-based classification scheme was introduced to estimate the yearly distribution of stable and seasonal water at 30 m spatial resolution in MLRYRR from 1984 to 2020.On the basis of the proposed framework and the Google Earth Engine platform, the annual spatial distribution data of stable and seasonal water bodies at a spatial resolution of 30 m in MLRYRR from 1984 to 2020 were obtained. The produced data were validated using 9000 validation samples in different scenes (e.g., urban, agricultural, and lacustrine scenes) and achieved a recall of 98.4%. Results showed that the spatiotemporal distribution of surface water and its trends demonstrate regional heterogeneity, with the water area in Jiangsu and Zhejiang Provinces expanding at 35.1 km2·a-1 and 6.5 km2·a-1, respectively, and the water area in Anhui, Jiangxi, and Hunan Provinces decreasing at 46.53, 35.6, and 26 km2·a-1, respectively. Moreover, the results of the annual water body area in MLRYRR can spatially reflect the drought and flood situations in different watersheds. The change trends of the water area in Hubei Province and Shanghai were insignificant. The mode and intensity of human disturbance and geo-climatic factors were the driving factors of the pattern differentiation in water evolution.The proposed surface water extraction framework and data results contribute to improving our understanding of the spatiotemporal distribution, evolution processes, and environmental effects of surface water. The results can provide spatial data and monitoring techniques to support surface water resource spatial investigation and the optimization of resource allocation, coordinated development, disaster risk assessment, and early warning. Future studies will focus on the dynamic observation method on surface water bodies through the collaborative processing of optical and synthetic aperture radar images to break through limitations imposed by continuously cloudy and rainy conditions in subtropical regions.  
      关键词:subtropical remote sensing;water resources;water body extraction;time-series analysis;Google Earth Engine;the Middle and Lower Reaches of the Yangtze River Region   
      414
      |
      412
      |
      0
      <HTML>
      <L-PDF><Enhanced-PDF><Meta-XML>
      <引用本文> <批量引用> 50070346 false
      发布时间:2024-12-30
    • 在自然资源监测领域,专家构建了新增建设用地卫星遥感智能监测技术框架,验证了其可行性,并在土地执法督查中得到应用。
      LIU Lirong,TANG Xinming,GAN Yuhang,YOU Shucheng,LIU Ke,LUO Zhengyu
      Vol. 28, Issue 11, Pages: 2828-2837(2024) DOI: 10.11834/jrs.20244063
      Research on satellite remote sensing-based intelligent monitoring technologies for new construction land
      摘要:The accurate and frequent extraction of information regarding new construction land is essential in natural resource monitoring and land law enforcement supervision. To fulfill these practical demands, this study constructed a technical framework for the intelligent monitoring of new construction land via satellite remote sensing. The framework includes a “spatial-temporal-spectral-classified” monitoring hypercube, a base map generation for monitoring, a sample annotation iteration, component-based artificial intelligence (AI) change detection model establishment, parcel information filtering, and post-processing. To meet the demand for accurate applications in large areas and complex scenes, this study fully combined different AI algorithms and network structures, such as attention mechanism, domain adaptation, and visual transformers, to develop a component-based AI change detection model for improving the accuracy and reliability of the algorithm. Meanwhile, to address issues, such as misidentification during the automatic extraction of new construction land parcels, parcel fragmentation, and edge inaccuracy, geomorphological principles were comprehensively utilized to set constraints and investigate post-processing parcel refinement methods. Experiments by region and time were conducted on large-scale remote sensing monitoring of new construction land to verify the feasibility of the proposed concept of the “spatial-temporal-spectral-classified” monitoring hypercube. Moreover, through ablation analysis of the component-based AI change detection model, the advantages and disadvantages of the algorithms were compared and analyzed. In particular, the visual transformer module exhibits evident advantages in terms of the feature completeness, edge accuracy, and recall rate of new construction land extraction. On the basis of certain operational data of satellite image-based law enforcement and supervision, cloud cover filtering was conducted. Wrongly extracted parcels accounted for about 0.84%. In addition, after the post-processing parcel refinement method proposed in this study was adopted, the accuracy and practicability of the monitoring results were further enhanced. The satellite remote sensing-based technologies and methods for the intelligent monitoring of new construction land proposed in this study have been applied to natural resource monitoring, such as land law enforcement and supervision.  
      关键词:satellite remote sensing;artificial intelligence;newly constructed land;change detection   
      98
      |
      68
      |
      0
      <HTML>
      <L-PDF><Enhanced-PDF><Meta-XML>
      <引用本文> <批量引用> 78381390 false
      发布时间:2024-12-30
    • 最新研究揭示了全国19个城市群建设用地扩张的时空变化特征,为优化城市群空间格局、促进区域协调发展提供科学依据。
      ZHANG Tao,WANG Guanghui,ZHENG Lijuan,DAI Hailun,HE Yuhua,LIU Ting
      Vol. 28, Issue 11, Pages: 2838-2849(2024) DOI: 10.11834/jrs.20243399
      Remote sensing monitoring and spatiotemporal characteristic analysis of urban construction land expansion in the urban agglomerations of China in the past 30 years
      摘要:The analysis of the spatial and temporal changes of urban agglomeration construction land expansion is highly significant for optimizing the spatial pattern of urban agglomerations and promoting regional coordinated development.In this study, 19 urban agglomerations in China are selected as the research object. Based on remote sensing image data of medium-scale resolution in China from 1990 to 2020, the boundary vector of urban concentrated construction areas is extracted from the actual construction of a city by using the method of human-computer interaction. Results are compared with those of other related studies, and research is conducted from the aspects of urban agglomeration expansion process, development stage, expansion mode, and center of gravity migration law.The results show the following. (1) Compared with other research results, the urban boundary of this study exhibits higher accuracy and reliability. It is also closer to the real situation of a city. (2) In the past 30 years, the urban scale change curve of 19 urban agglomerations in China generally presents an “S” type, and the expansion process of urban agglomerations can be divided into three periods: slow, rapid, and stable expansion. Eastern urban agglomerations entered the rapid expansion period about 10 years earlier than western urban agglomerations. Most urban agglomerations entered a period of steady expansion after 2015. (3) By 2020, 19 urban agglomerations were in a high-level development stage, and the coordination degree of spatial expansion between internal central and peripheral cities was continuously improved. From the perspective of spatial distribution, 70% of the urban agglomerations in the eastern region are in the stage of decentralization, 56% of the urban agglomerations in the western region are in the stage of agglomeration attenuation, and eastern urban agglomerations are generally in a higher level of urban agglomeration development stage. (4) Based on the expansion scale of urban agglomerations in different directions, the spatial expansion modes of 19 urban agglomerations in China can be divided into circular, fan-shaped, and axial expansions. The spatial expansion mode is closely related to the number, location, and influence of core cities within an urban agglomeration. In addition, topographic conditions pose certain restrictions on the expansion direction of urban agglomerations. (5) The gravity center migration trajectories of urban agglomerations are different, and expansion direction may be affected by various factors, such as the location of core cities, the construction of new areas, traffic conditions, and terrain conditions. From the perspective of change in the center of gravity of urban agglomerations, the center of gravity of 84% of urban agglomerations was relatively stable in the past 30 years, i.e., basically located in the core city or the same city adjacent to the core city. The core city of urban agglomerations is attractive, and the expansion of the construction scale of a peripheral city is relatively average.The results of this study provide intuitive and accurate data for the study of urban expansion in China. Long-term and high-precision monitoring of urban construction scale can fully reflect the development process of urban agglomerations.  
      关键词:urban agglomeration;expansion process;expansion mode;spatiotemporal characteristics   
      1264
      |
      1847
      |
      0
      <HTML>
      <L-PDF><Enhanced-PDF><Meta-XML>
      <引用本文> <批量引用> 50080253 false
      发布时间:2024-12-30
    • 在国土资源监测领域,专家设计了基于空间注意力机制与多任务学习的地块分割模型Field-Net,提高了监测效率和时效性。
      TIAN Fuyou,CAO Yupei,ZHAO Hang,WU Bingfang,ZENG Hongwei,LIU Yazhou,QIN Xingli,ZHANG Miao,ZHU Liang,ZHU Weiwei
      Vol. 28, Issue 11, Pages: 2850-2864(2024) DOI: 10.11834/jrs.20243191
      Agricultural field segmentation using spatial attention mechanism and multi-task learning strategy
      摘要:The accurate identification of agricultural fields, which are the smallest units of agricultural farming, is crucial for monitoring land resources and arable land utilization. Manual mapping methods are time-consuming and expensive, and they are incapable of real-time or near-real-time updates. To enhance agricultural field delineation efficiency, we introduce Field-Net, a field segmentation model that leverages a spatial attention mechanism and multitask learning in this research.Field-Net is based on the UNet architecture. It combines a spatial attention mechanism and a multitask learning approach. In addition to the segmentation task, we incorporate boundary identification and distance to the field boundary as two additional tasks, enabling the model to learn representative features related to fields. The model’s performance was evaluated using Gaofen-1 and Ziyuan-3 satellite images with a spatial resolution of 2 m in Lijin County, Dongying City, Shandong Province. We labeled 3,480 tiles that measured 256×256 pixels in YanWo District, with 3000 used for training, 360 for validation, and 120 for the spatial generalization performance test.We initially analyzed loss weight for the three tasks, i.e., mask, boundary, and pixel-to-boundary distance, in multitask learning by using a gradient test. For multitask learning, loss weight should prioritize the mask segmentation task as the primary task, while other tasks should be considered secondary. Across the entire test set, Field-Net achieved an overall accuracy of 92.23% and an Intersection Over Union (IOU) of 87.05%. We compared Field-Net with four state-of-the-art architectures: DeepLab v3+, HRNet, LinkNet, and D-LinkNet. Field-Net outperformed all of them in semantic segmentation tasks, with an IOU that was 0.26% higher than Link-Net, the most accurate among the four compared models, and 7.59% higher than that of DeepLab v3+. In the spatial generalization performance test, the average IOU of the Field-Net model was 3.51% higher than that of the Link-Net model, and spatial generalization performance was significantly improved. Ablation tests demonstrated that the spatial attention mechanism and multitask learning strategy improved the F1 score by 1.01% and IOU by 1.6% compared with the ResUNet model. The multitask learning strategy led to an improvement of 0.18% in F1 score for Field-Net and an improvement of 0.21% in IOU.Although challenges remain in identifying contiguous fields due to unclear boundaries, future enhancements can incorporate multitemporal and high-resolution remote sensing images to improve field feature discrimination. Feature visualization analysis revealed that the spatial attention mechanism and multitask learning strategy enabled the model to learn clustered features at field boundaries and within plots, enhancing feature representativeness. Overall, the Field-Net model supports the field-level monitoring of cropland use, including nonagricultural applications, such as grain production, enhancing the efficiency and timeliness of land resource monitoring. In generating the field dataset of China, complex and fragmented croplands pose considerable challenges to this task. In the future, the problem of lack of samples for model training can be solved by accumulating field segmentation datasets from different regions by borrowing the paradigm of Image-Net, while a general model for channels, regions, and sensors should be constructed subsequently. In the future, with the arrival of the “large model” era of deep learning, constructing a model for the task of parcel segmentation is also necessary to segment every field from the perspective of model and dataset.  
      关键词:Filed segmentation;Field-Net model;Spatial attention mechanism;Multi-task learning;GF satellite data   
      497
      |
      586
      |
      0
      <HTML>
      <L-PDF><Enhanced-PDF><Meta-XML>
      <引用本文> <批量引用> 50733504 false
      发布时间:2024-12-30
    • 在农业资源调查领域,专家提出了多尺度时空全局注意力模型MSSTGAM,有效提高了遥感影像时间序列农作物分类精度。
      ZHANG Weixiong,TANG Ping,MENG Yu,ZHAO Lijun,ZHAO Zhitao,ZHANG Zheng
      Vol. 28, Issue 11, Pages: 2865-2877(2024) DOI: 10.11834/jrs.20243557
      Crop type classification of remote sensing image time series based on multi-scale spatial-temporal global attention model
      摘要:With the development of deep learning, the use of deep learning methods to obtain accurate crop classification results from remote sensing image time series has become a research hotspot. The automatic intelligent interpretation of fine types of crops by utilizing remote sensing image time series plays an important role in the fields of agricultural resource investigation, supervision, and planning.The classical time series classification of remote sensing images is based on pixel-based classification, and only the temporal information of the time series is utilized. The spatial information of the shape, size, and distribution of ground objects in the time series of remote sensing images also plays an important role in the classification crops, so it is beneficial to extract the hybrid spatial-temporal features of the time series by fully mining the spatial—temporal information of the time series. However, existing deep learning methods extract local spatial or local temporal information by using convolutional or recurrent neural networks, resulting in the inadequate utilization of spatial-temporal information, and consequently, low classification accuracy.In recent years, whether in the field of Natural Language Processing or Computer Vision, the self-attention mechanism has proven to be an effective method to fully utilize data information by attaining global attention. Thus, in this paper, we propose a multiscale spatial-temporal global attention model (MSSTGAM), which combines a spatial self-attention mechanism and a temporal self-attention mechanism to construct a multiscale spatial-temporal global attention mechanism and fully obtain the information of the remote sensing image time series for the fine classification of crop types. Specifically, MSSTGAM adopts SWIN Transformer to process the spatial information of remote sensing image time series to obtain output at different spatial scales, and uses lightweight temporal attention encoder (LTAE) to obtain spatial-temporal global features at the deepest spatial scale, and shares the temporal attention weights to other spatial scale through the temporal sharing block to obtain multi-scale spatial-temporal global attention features for fine classification of crop type.The proposed method is evaluated on the publicly available dataset PASTIS and customized Mississippi dataset. The overall classification accuracy of 83.4% and 86.7% was obtained on the two datasets, respectively. Moreover the proposed method achieves the best F1 scores in most crop types, especially for wheat crops, which have an improvement of 2.6% and 3.3% over existing methods on the two datasets, respectively. The quantitative results demonstrate the effectiveness and application value of MSSTGAM for fine classification of crop type. The visualization of the classification results shows that the classification results of the proposed method have better spatial consistency, and the further visual analysis of temporal attention weights points out the theoretical basis for the proposed method to obtain fine classification of crops.The findings of this study show that multiscale spatial-temporal global attention demonstrates significant theoretical and practical significance. MSSTGAM can capture the global spatial-temporal evolution of land cover, which is conducive to improving the spatial consistency and classification accuracy of fine crop types. It is more effective for the fine classification of crop types from remote sensing image time series.  
      关键词:remote sensing image time series;crop type classification;self-attention mechanism;global attention;multi-scale spatial-temporal feature   
      442
      |
      390
      |
      0
      <HTML>
      <L-PDF><Enhanced-PDF><Meta-XML>
      <引用本文> <批量引用> 62606628 false
      发布时间:2024-12-30
    • 在农业遥感领域,专家探索了样本量对作物分类效果的影响及模型空间外推可行性,为大尺度区域作物遥感精准分类提供高效经济方法。
      XIE Yan,ZENG Hongwei,TIAN Fuyou,ZHANG Miao,HU Yueran,QIN Xingli,WU Bingfang,ZHANG Youzhi,XIE Wenhuan
      Vol. 28, Issue 11, Pages: 2878-2895(2024) DOI: 10.11834/jrs.20244005
      A study of sample dependence and spatial extrapolation of models for crop remote sensing classification
      摘要:Reducing reliance on in situ crop type samples is critical for remotely sensed crop type classification over large areas. This study used Suihua, a major grain-producing city in Heilongjiang Province, as an example to investigate the effect of sample size on crop type classification and test the possibility of extrapolating supervised classification models trained on a small region onto a larger area. In particular, this study trained the crop type classification model in Beilin District and then extrapolated it to the entire Suihua. First, a parameter-optimized random forest model was trained and used to identify the spatial distribution of crops in Beilin District in 2022 by using Sentinel-2 remote sensing imagery from the sowing to the mid-tasseling of maize. Overall Accuracy (OA) gradually increased as the proportion of samples participating in the random forest training increased from 10% to 50% of the Gaussian Variate Generator (GVG) samples in Beilin District. The model achieved the best performance with a maximum OA of 94.6% when 50% of the GVG samples in Beilin District were used for crop classification, where maize, rice, and soybean had approximately 130 training samples. Thereafter, the performance of the model remained stable even as the number of in situ crop samples increased. The most important features in the classification of maize, soybean, and rice were REP at the tassel stage of maize, shortwave-infrared (SWIR)1 at the pod stage of soybean, and the Land Surface Water Index (LSWI) during the transplanting stage of rice. Second, we extrapolated the best trained model in Beilin District to classify crop types in the entire Suihua. The model extrapolation achieved an OA of 93.7% for crop type classification in Suihua. This value was only 1.3% lower than that of the model trained directly in Suihua. The similarity of the spatial and probability distribution maps of the crops between the Beilin and Suihua models indicated that the extrapolation of the crop classification model in a small area can achieve a comparable classification result with the crop classification model trained directly in a large area. Finally, we carefully examined the effects of distance, spatial representativeness and number of samples, and similarity of crop structure between small area and target expansion area on model extrapolation. Different crops exhibit varying sensitivities to distance, and the classification effect of rice is insensitive to changes in distance due to the significant differences between the LSWI and SWIR1 of rice and other crops. Meanwhile, the classification effects of maize and soybean exhibit an overall decreasing trend of change with increasing extrapolation distance. In summary, when building crop classification models in small regions with similar crop structures in the source and target areas, not only the number of samples should be considered, but also the representativeness of their spatial distribution. Such consideration will ensure that the model is adequately trained and can achieve better spatial extrapolation effect. The results of this study provide a cost-effective and efficient method for accurately classifying crops over large areas by using remote sensing. In addition, this study provides a scientific basis for developing crop sampling strategies, selecting sensitive bands, and determining the classification time window. It is also a valuable reference for the development of model extrapolation methods with higher robustness and generalizability.  
      关键词:crop classification;Sample Dependency;Model Extrapolation;Random Forest;Google Earth Engine   
      285
      |
      280
      |
      0
      <HTML>
      <L-PDF><Enhanced-PDF><Meta-XML>
      <引用本文> <批量引用> 56633090 false
      发布时间:2024-12-30

      Intelligent Remote Sensing of Ecological Environment

    • 最新研究突破了遥感生态指数RSEI在高分辨率应用的限制,为生态环境质量评估提供新方法。
      WANG Yingqi,HUANG Huiping,ZHU Wenlu,YANG Guang,YU Kun
      Vol. 28, Issue 11, Pages: 2896-2909(2024) DOI: 10.11834/jrs.20243529
      Construction and application of high-resolution remote sensing ecological index
      摘要:The Remote Sensing Ecological Index (RSEI) is the most widely used ecological environment quality assessment model. In general, the four indicators (greenness, wetness, heat, and dryness) of RSEI are calculated from Landsat images to construct an index that comprehensively reflects the ecological environment condition in pixel units. High-resolution remote sensing images generally lack the short-wave and thermal infrared bands involved in the calculation of RSEI; therefore, the application of RSEI to high-resolution ecological environment quality assessment is limited. The advantages of high-resolution remote sensing data cannot be fully utilized due to the limitation of spectral resolution, which is undoubtedly wasteful.To solve the problem of mismatch between high-resolution remote sensing image bands and the bands required for RSEI calculation, this study established a multi-resolution band fusion model with scale-invariant features. On the basis of Landsat 8 and Gaofen (GF)-2 remote sensing images, short-wave infrared bands and surface temperature with high resolution (4 m) were generated utilizing the statistical relationship between bands. The high-resolution RSEI (HRSEI) was constructed based on the principle of RSEI, filling the gap of RSEI research at the fine scale. This method was applied to Fan County in Henan Province. The results showed the following.(1) High-resolution short-wave infrared band and surface temperature can be generated by utilizing the multi-resolution band fusion technique. The correlation coefficients between the fitted and original images were higher than 0.7, indicating that the machine learning model based on the random forest algorithm was effective. The obtained high-resolution band/product can be used in the subsequent ecological environment quality evaluation work. This method can effectively compensate for the disadvantage posed by the band absence of high-resolution images, breaking through the limitation of RSEI application at the fine scale and expanding the application scenario of high-resolution remote sensing data.(2) The calculation results of the first principal component of HRSEI showed that the loadings of greenness and wetness were positive, while those of heat and dryness were negative, indicating that greenness and wetness promoted ecological environment quality, whereas heat and dryness impeded it. The above results are consistent with the objective actual pattern and coincided with the trend of the RSEI results. The Pearson correlation coefficient showed that HRSEI and RSEI were highly correlated (R=0.74). The contrast and information entropy of HRSEI for the three typical areas (built-up, village, and beach areas) were greater than those of RSEI. By maintaining high relevance and consistency, the information abundance presented by HRSEI generated from 4 m GF-2 data is significantly higher than that from 30 m Landsat data.(3) The results of HRSEI in 2016 and 2023 showed that the ecological environment quality of Fan County had been generally improved. However, some areas where ecological environment quality deteriorated remained. Two major factors contributed to the deterioration. First, urbanization led to the expansion of built-up land, with previously cultivated or forested land being changed to impervious surfaces. Second, villages near the Yellow River carried out demolition of old villages and construction of resettlement areas due to the policy of relocation and reclamation in the Yellow River beach area. In particular, the lack of timely reclamation after the demolition of old villages seriously expanded the scope of deterioration of ecological environment quality.  
      关键词:high resolution;Remote sensing ecological index;Ecological environment quality;RSEI;Band fusion   
      376
      |
      755
      |
      0
      <HTML>
      <L-PDF><Enhanced-PDF><Meta-XML>
      <引用本文> <批量引用> 56615810 false
      发布时间:2024-12-30
    • 在植被物候研究领域,专家基于Google Earth Engine平台,利用ESTARFM算法融合Landsat 8影像与MODIS产品,生成高时间分辨率EVI序列,验证了其在物候监测能力上相比MODIS数据的提升效果,并探讨了影响融合效果的因素,为精细化的植被动态监测和生态系统研究提供理论支撑和数据参考。
      SONG Jie,ZHANG Zhao,HAN Jichong
      Vol. 28, Issue 11, Pages: 2910-2926(2024) DOI: 10.11834/jrs.20232646
      Accurately retrieving vegetation phenology at high spatial and temporal resolutions based on GEE and multi-source remote sensing data fusion
      摘要:Vegetation phenology is considered the most direct and sensitive indicator for assessing environmental changes. At present, costly and limited in situ observations at field scales are impractical for phenology monitoring. Alternatively, various aspects of phenology have been successfully characterized using remote sensing images. Image fusion has become a breakthrough in deriving fine-resolution phenological metrics due to the trade-off between the spatial and temporal resolutions of satellite sensors. Most existing studies have retrieved phenology with sparse time series (with intervals of >8 days); consequently, they fail to capture small intra-annual variations in phenology. In addition, a few scholars have used relatively dense time series, but obtained less robust results. Thus, extracting phenological metrics from fused results at fine temporal resolutions and comprehensively validating them are necessary to increase our knowledge regarding the fusion method.In this study, we implemented a spatiotemporal fusion approach, called the Enhanced Spatial and Temporal Adaptive Reflectance Fusion Model (ESTARFM), on the Google Earth Engine (GEE) platform. By leveraging this model, we generated a 30 m time series of the Enhanced Vegetation Index (EVI) with different resolutions (1, 3, 5, 7, 9, and 11 days) by fusing Landsat 8 and Moderate Resolution Imaging Spectroradiometer (MODIS) images in 2018. We selected four PhenoCam sites in North America. Among which, the predominant vegetation types are paddy rice, deciduous forests, corn, and shrubs (hereinafter referred to as Cam1—Cam4, respectively). We used Savitzky-Golay filtering and the maximum separation method to estimate phenological metrics, such as Start of Season (SOS), End of Season (EOS), and Length of Season (LOS). We evaluated the effect of temporal resolution on the phenometrics derived from the fusion results with field data (from the PhenoCam project). Then, we used the daily fused time series for further illustration. The fused images were validated via true Landsat 8 observations with sufficient unmasked pixels. Finally, we used MODIS data to assess the improvement in accuracy of the derived phenometrics fusion results.We generated 133, 273, 126, and 288 images via ESTARFM at daily resolution at Cam1—Cam4, respectively. These fused images obtained more spatial phenological details than low-resolution images (e.g., MODIS), and the errors of phenometrics derived from the fusion results generally increased with temporal resolution. The errors became more evident when time resolution reached 7 days. We conducted research and discussions under daily resolution. The correlational results prove that the fused images can accurately capture reliable spatial patterns with spatial efficiency ranging from 0.14 to 0.74, and they are temporally consistent with field observations (root mean square error: 0.01—0.02, r: 0.73—0.95). The derived phenological metrics are relatively accurate, with mean errors of 4.25, 4.75, and 7.5 days for SOS, EOS, and LOS, respectively. Compared with MODIS data, ESTARFM evidently reduces the errors of fusion-derived phenometrics for deciduous forests and shrubs.This research deployed a new framework on the GEE platform to estimate phenological metrics derived from satellite images fused by ESTARFM and assessed accuracy with field data in North America. Our findings revealed the reliability of the EVI time series with fine resolution generated by ESTARFM. This framework helps us obtain authentic details of spatiotemporal phenological distributions. Our study will surely provide important supportive theories and data basis for issues concerning vegetation dynamics and ecosystem.  
      关键词:vegetation phenology;ESTARFM;maximum separation method;PhenoCam;Google Earth Engine (GEE)   
      565
      |
      426
      |
      0
      <HTML>
      <L-PDF><Enhanced-PDF><Meta-XML>
      <引用本文> <批量引用> 43004278 false
      发布时间:2024-12-30
    • 最新研究利用遥感技术,开发自适应地形卷积算法,有效提高高山松碳储量估测精度。
      TENG Chenkai,XIAO Yueyao,ZHANG Jialong,HE Yunrun,CHEN Chaoqing
      Vol. 28, Issue 11, Pages: 2927-2942(2024) DOI: 10.11834/jrs.20244006
      Estimation of <italic style="font-style: italic">Pinus densata</italic> carbon storage based on Landsat time series data and ATC filtering algorithm
      摘要:Time-series remote sensing data have applications in the accurate estimation of forest carbon storage, providing data support for a deeper understanding of the carbon cycle process of forest ecosystems, scientific management, and protection of forest resources. However, considerable noise is present in remote sensing time-series data. To enhance the accuracy of estimating carbon storage, a filtering algorithm was developed to reduce the interference of noise in Landsat time-series data from high-altitude areas. The algorithm was developed based on the continuous inventory fixed plot data of the National Forest Inventory in the Shangri-La area for the years 1987, 1992, 1997, 2002, 2007, 2012, and 2017, and Landsat time-series images from 1987 to 2017. In this study, the Adaptive Topography Convolution (ATC) algorithm was developed using Python. This algorithm considers the effects of terrain factors on image quality, removes noise from the image while retaining as much details as possible, and uses Savitzky–Golay filtering and median filtering to filter Landsat time-series data. By using the random forest regression (RFR) algorithm, a carbon storage estimation model for Pinus densata in Shangri La City was constructed. The optimal estimation model was selected to invert and map the carbon storage of P. densata in 1987, 1992, 1997, 2002, 2007, 2012, and 2017. The results showed the following. (1) In accordance with the mean absolute error of the image quality evaluation index, image quality is best after filtering with the ATC algorithm. In addition, the peak signal-to-noise ratio value of the time-series data filtered by the ATC algorithm is relatively high, indicating an improvement in data quality. (2) When using the RFR algorithm, the filtered data showed higher fitting and prediction accuracy than the original data. (3) When using the RFR algorithm, the time-series data filtered based on the ATC algorithm exhibit the best estimation accuracy when selecting the top 10 feature factors with contribution and the feature factors with cumulative contribution reaching 70% for modeling. (4) The estimation model constructed based on the ATC-filtered time series and remote sensing features (number of features 10) and the random forest algorithm exhibits the best performance in research, with a determination coefficient (R2) of 0.867, a root mean square error (RMSE) of 15.527/(t/hm2), a prediction accuracy (P) of 73.54%, and a relative RMSE (rRMSE) of 41.14%. 5) The carbon storage inversion results of Shangri La Pinus densata based on the optimal estimation model are as follows: 6.77 million tons (1987), 7.16 million tons (1992), 7.22 million tons (1997), 4.36 million tons (2002), 7.20 million tons (2007), 7.11 million tons (2012), and 7.53 million tons (2017). From the inversion results, the carbon storage of Shangri-La P. densata exhibited a gradually increasing trend during the period of 1987—1997. However, carbon storage exhibited a significant fluctuating trend from 2002 to 2017. The use of the ATC filtering method can effectively remove noise in the time-series images of high-altitude areas, reducing the uncertainty of time-series images and improving the accuracy of the remote sensing estimation of P. densata carbon storage.  
      关键词:Landsat time series;Filter;Pinus Densata;carbon storage;ATC   
      263
      |
      166
      |
      0
      <HTML>
      <L-PDF><Enhanced-PDF><Meta-XML>
      <引用本文> <批量引用> 56625805 false
      发布时间:2024-12-30
    • 在高分辨率遥感影像建筑物提取领域,ILGS-Net模型结合CNN与Transformer,有效融合局部细节与全局上下文特征,提高建筑物边界定位精度。
      LIU Yuxin,MENG Yu,DENG Yupeng,CHEN Jingbo,LIU Diyou
      Vol. 28, Issue 11, Pages: 2943-2953(2024) DOI: 10.11834/jrs.20243307
      Integration of CNN and transformer for high-resolution remote sensing image building extraction: A dual-stream network
      摘要:Convolutional Neural Networks (CNNs) and transformers have emerged as pivotal tools in the field of building extraction tasks in high-resolution remote sensing images. Although these techniques have found widespread applications, challenges persist for CNNs in effectively modeling long-range spatial dependencies, frequently leading to complications, such as the emergence of internal holes in the extracted building structures. Conversely, transformers exhibit limitations in capturing spatial local details, potentially resulting in the production of blurry building edges and the oversight of smaller structures. In response to these challenges, this study presents an innovative dual-stream network model tailored for building extraction in high-resolution remote sensing images, called the network for the integration of local and global features stream (ILGS-Net). ILGS-Net is designed to capitalize on the strengths of CNNs and transformers. The model incorporates multilevel local-global feature fusion modules to seamlessly blend the intricate local details and expansive global context features of buildings. An edge loss function is integrated into the objective function, contributing to the refinement of building boundary localization precision.The proposed ILGS-Net endeavors to address the shortcomings of existing methodologies by efficiently combining the unique attributes of CNNs and transformers. The multilevel local-global feature fusion modules play a pivotal role in striking a harmonious balance between capturing the fine-grained local details and incorporating the broader global context features of buildings. Simultaneously, the inclusion of an edge loss function serves as a guiding mechanism in model training, augmenting the precision of building boundary localization. Extensive experiments conducted across three high-resolution building datasets consistently demonstrate the superior performance of the proposed ILGS-Net compared with benchmark methods described in this paper. Notably, the proposed method achieves a remarkable increase of an average of 1% in the intersection over union across all three datasets.In conclusion, ILGS-Net emerges as a groundbreaking dual-stream network model that is specifically designed for building extraction in high-resolution remote sensing images. By seamlessly integrating CNNs and transformers, along with the implementation of multilevel local-global feature fusion and the inclusion of an edge loss function, the model adeptly addresses challenges associated with spatial dependencies and local details, resulting in a marked improvement in the accuracy of building extraction. The experimental results underscore the efficacy of the proposed method, making it a promising and influential approach for achieving high-precision building extraction in high-resolution remote sensing images. The confluence of advanced methodologies and innovative techniques within ILGS-Net marks a significant step forward in the field of remote sensing image analysis. As technology continues to evolve, ILGS-Net represents a pivotal contribution that exhibits promise for further advancements in building extraction accuracy, providing a solid foundation for continuous research and application in the field of high-resolution remote sensing imagery analysis.In the future, the success of ILGS-Net prompts further exploration and research. Investigating the potential of similar integrative approaches in other remote sensing tasks exhibits promise. In addition, refining and expanding the current model architecture to accommodate varying scales and complexities of urban landscapes are a logical progression. Future work should focus on translating these advancements into tangible benefits for decision-makers and stakeholders in urban development and disaster response.  
      关键词:remote sensing;building extraction;deep learning;dual-stream network;edge loss;local-global feature fusion   
      755
      |
      672
      |
      0
      <HTML>
      <L-PDF><Enhanced-PDF><Meta-XML>
      <引用本文> <批量引用> 50051209 false
      发布时间:2024-12-30
    • 记者从城市绿地遥感监测领域获悉,专家提出了一种复杂城市环境下的资源三号遥感影像绿地提取方法,有效提高了城市绿地提取精度,为城市可持续发展提供科学依据。
      DU Xiaoyu,WANG Guanghui,LU Chen,YAN Zhigang,ZHANG Tao
      Vol. 28, Issue 11, Pages: 2954-2969(2024) DOI: 10.11834/jrs.20243521
      Green space extraction method from ZY-3 remote sensing images in complex urban environment
      摘要:Urban green space is an important part of the urban ecosystem, providing multiple functions such as ecology, landscape, culture, and health benefits. Obtaining the spatial distribution of urban green space accurately can offer a scientific basis and support for sustainable urban development. However, using ZY-3 remote sensing images and the Normalized Difference Vegetation Index (NDVI) to extract urban green spaces, urban background features such as water, blue-roofed buildings, and shadows may be confused with urban green space. To address this issue, we proposed an extraction method of urban green space for ZY-3 remote sensing images in complex urban environments. This method introduced extraction features of green space that are robust to low-brightness pixels, and designed object-level extraction features of shadow in the Hue-Saturation-Intensity(HSI) color space. Then it applied different features to perform threshold segmentation inside and outside shadow areas, achieving the extraction of urban green space. The accuracy of the proposed method was evaluated using the 10 cities including Nanjing, Wuhan, Urumqi, and Shenyang, as research areas. The results showed that: (1) The overall precision, recall, F1 value, and IoU (Intersection of Union) of the proposed method were 93.28% and 92.60%,92.91 and 86.76%. (2) The proposed extraction feature of green space robust to low-brightness pixels outperformed RVI, NDVI, and DVI. Compared with related models such as Deeplab V3+, Segformer and UPerNet, the proposed method had better overall performance, and its extraction accuracy was superior to the product Urban Green Space-1 m(UGS-1 m); (3) From the perspective of local extraction details, the proposed method can effectively distinguish green spaces from water and blue-roofed houses, and extract vegetation inside the shadow areas of buildings. The proposed method is of great significance for quickly and efficiently conducting operational remote sensing monitoring of urban green space.  
      关键词:remote sensing;urban green space;high-resolution remote sensing;ZY-3;Normalized Difference Vegetation Index (NDVI);shadow extraction   
      75
      |
      59
      |
      0
      <HTML>
      <L-PDF><Enhanced-PDF><Meta-XML>
      <引用本文> <批量引用> 78381659 false
      发布时间:2024-12-30
    • 在水质监测领域,专家建立了基于LBFGS-MLP模型的非光学特性水质参数反演体系,为评估城市河流水体状况提供理论依据。
      HE Ruyan,LYU Zijun,JIA Sen
      Vol. 28, Issue 11, Pages: 2970-2983(2024) DOI: 10.11834/jrs.20243509
      Inversion of non-optical water quality parameters of hyperspectral remote sensing based on LBFGS- accelerated multi-layer perceptron network
      摘要:Water is the source of life, the foundation of survival, a necessity for production, and the basis of ecology. However, under the dual pressures of human activities and climate change, aquatic ecosystems are facing increasingly severe challenges, particularly the serious problem of water pollution, which directly threatens the physical and mental health of residents. Water quality monitoring plays a crucial role in water pollution control, which precisely evaluates the health of water bodies and promptly adjusts control strategies, ensuring the stability and health of water environmental quality. Hyperspectral remote sensing exhibits significant potential in water quality monitoring. With the rapid development of Unmanned Aerial Vehicles (UAVs) and hyperspectral technology, UAVs equipped with hyperspectral sensors have considerably improved in terms of spectral and spatial resolutions. Accordingly, water quality parameter inversion by using hyperspectral remote sensing has gradually become a research hotspot. However, current research predominantly focuses on optical water quality parameters, with relatively less emphasis on nonoptical parameters, which also reflect the effect of human activities on water bodies. In this study, an urban river in a certain village in Guangdong Province is selected as the study area,. An experiment that involves UAV for hyperspectral remote sensing image acquisition and simultaneous water sample collection is conducted. Then, we propose a multilayer perceptron (MLP) network model accelerated by the limited-memory Broyden–Fletcher–Goldfarb–Shanno (LBFGS) method, called LBFGS-MLP, for the inversion of nonoptical water quality parameters. The parameters include Total Phosphorus (TP), Total Nitrogen (TN), and ammonia nitrogen (NH3-N), which are important indicators for measuring the nutritional status of water bodies. Through Pearson correlation analysis, spectral bands related to the three nonoptical water quality parameters (TP, TN, and NH3-N) are selected as model input. Subsequently, on the basis of exploring the effect of different network depths and optimization algorithms on model performance, the LBFGS optimization algorithm is adopted to accelerate the MLP network, and the loss function is mean squared error. Finally, the LBFGS-MLP model is applied to spatially analyze the concentrations of TP, TN, and NH3-N in the study area. Overall, the LBFGS-MLP model demonstrates significantly better accuracy on the training and testing datasets for the concentrations of TP, TN, and NH3-N compared with the random forest, CatBoost, and XGBoost models, particularly in the inversion of TN and NH3-N concentrations. The model’s coefficients of determination are 0.71, 0.82, and 0.72, while the mean absolute errors are 0.0118, 0.0394, and 0.0601 mg/L, respectively. The concentrations of TP, TN, and NH3-N in the study area are mostly distributed at 0.1—0.3, 2—5, and 0.1—0.4 mg/L, respectively, which are consistent with the survey results. Through this study, the effectiveness of the MLP algorithm in the inversion of nonoptical water quality parameters is verified, providing a theoretical basis and reference for a more comprehensive assessment of urban river water body condition.  
      关键词:Non-Optical Water Quality Parameters;machine learning;hyperspectral remote sensing;concentration inversion   
      196
      |
      159
      |
      0
      <HTML>
      <L-PDF><Enhanced-PDF><Meta-XML>
      <引用本文> <批量引用> 61845502 false
      发布时间:2024-12-30

      Intelligent Remote Sensing for Disaster Prevention and Mitigation

    • 在森林草原火灾防控领域,专家基于多源遥感数据,分析了2001至2021年四川省火灾时空特征,为预警防控提供先验信息。
      JIAO Miao,QUAN Xingwen,HE Binbin,YAO Jinsong
      Vol. 28, Issue 11, Pages: 2984-3001(2024) DOI: 10.11834/jrs.20243082
      Remote sensing-based spatial-temporal characteristics of forest grassland fires in Sichuan Province from 2001 to 2021
      摘要:In recent years, the problem of forest wildfires in Sichuan Province has emerged as a matter of great concern. These wildfires have occurred with alarming frequency, presenting a formidable threat not only to the local ecological security but also to the lives and property of the people and the courageous rescue workers who put themselves in harm’s way. This in-depth study is committed to conducting a comprehensive exploration of the temporal and spatial characteristics of forest and grassland fires in Sichuan Province over the extensive period from 2001 to 2021. The ultimate aim is to provide highly valuable and actionable information that can serve as a solid foundation for making well-informed decisions regarding fire prevention and control strategies. By understanding these characteristics, it is hoped that effective measures can be implemented to minimize the occurrence and impact of wildfires, thereby safeguarding the delicate balance of the ecosystem and protecting the well-being of the local population.This extensive research is firmly grounded in a diverse array of multi-source remote sensing fire products, such as MCD64A1, Fire_CCI51, and MCD14ML. Through meticulous extraction of effective fire points, a wealth of regional fire data is painstakingly obtained. Leveraging the power of a sophisticated geographic information system, the temporal trend and spatial distribution of forest and grassland fires are thoroughly examined. This involves analyzing patterns over time and identifying areas of concentration or dispersion. Additionally, mathematical statistics and an adaptable fuzzy neural network are skillfully employed to meticulously analyze the complex relationship between climatic, combustible, and topographical environmental factors and the occurrence of fires. By using these advanced techniques, researchers can gain a deeper understanding of the underlying causes and contributing factors of wildfires, enabling more targeted prevention and response efforts.Research findings reveal that from 2001 to 2014, both the frequency of fires and the area affected by them demonstrated an upward trend. This indicates a growing concern for fire management and prevention. Fires occurred with notable frequency from January to May, suggesting a seasonal pattern that can be used to inform preventive measures during these high-risk months. In terms of the spatial distribution of grassland fires, it exhibits a distinct heterogeneity, with a concentration mainly in the southwest region of Sichuan Province. This spatial pattern may be influenced by a combination of factors such as vegetation type, climate, and human activities. Intriguingly, in the northeast of China, grassland fires have witnessed a remarkable increase in recent times. This finding highlights the need for a broader understanding of fire dynamics on a national scale. In the correlation analysis of various influencing factors, a high degree of correlation is observed between forest fires and fuel water content. This suggests that changes in fuel moisture levels can have a significant impact on the likelihood and severity of forest fires. Environmental variables are clearly identified as the primary driving factors behind the temporal and spatial characteristics of forest fires. For grassland fires, although there is a strong correlation with meteorological factors, it is reasonably speculated that human factors also exert a substantial influence on the characteristics of grassland fires. This could include activities such as land use changes, agricultural practices, and accidental ignitions.Based on the detailed analysis of the temporal and spatial characteristics of fires in Sichuan Province, this study provides a solid and reliable decision-making basis for formulating forest and grassland fire prevention and control policies, establishing early warning systems, and enhancing monitoring efforts in this region. By understanding the patterns and drivers of wildfires, policymakers and fire management agencies can develop more effective strategies to protect the precious ecological environment and safeguard the lives and property of the people. This will undoubtedly contribute to a more sustainable future for the region, ensuring that the beauty and biodiversity of Sichuan’s forests and grasslands are preserved for generations to come.  
      关键词:remote sensing;Sichuan province;MCD64A1;Fire_CCI51;MCD14ML;forest grassland fire;spatial distribution;time trend;spatio-temporal characteristics   
      606
      |
      642
      |
      0
      <HTML>
      <L-PDF><Enhanced-PDF><Meta-XML>
      <引用本文> <批量引用> 54702672 false
      发布时间:2024-12-30
    • 据最新研究,云南省红河州冰雹灾害时空分布特征与规律被揭示,专家提出多级格网冰雹灾害遥感监测模型,为农业防雹提供解决方案。
      SHAO Xiaodong,JIANG Yangming,HUANG Kun,WANG Futao,WANG Tuo,ZHAO Huihui,HOU Qiuqiang,RUAN Haiming,GUAN Qunrong
      Vol. 28, Issue 11, Pages: 3002-3015(2024) DOI: 10.11834/jrs.20243483
      Multi-grid remote sensing monitoring method of hail disaster and the temporal-spatial distribution characteristics in Honghe
      摘要:Hail has occurred frequently and caused significant losses to local agricultural production in Honghe Prefecture, Yunnan Province, since 1961. Hail disaster distribution data at the county or weather station scale, which are obtained by using a statistical analysis method, cannot meet the requirements of agricultural hail prevention. Several hail disaster remote sensing monitoring methods, which are limited by single remote sensing data sources and the characteristics of designing for the global scale, lack applicability in mountainous areas. To capture the spatial and temporal distribution characteristics of hail and build a hail remote sensing monitoring model at the parcel level, this study used hail record data from hail suppression operation stations from 2009 to 2022 and conducted research on a multisource data fusion approach based on Ross Li and STARFM. It then proposed a multilevel grid normalized vegetation index standardization model and a hail remote sensing monitoring recognition index RNDVI_M. The Kneed method was used to extract the trend turning points of RNDVI_M as the threshold for extracting hail disaster areas. Then, the phenomenon universality verification method was applied to verify the effectiveness of the RNDVI_M threshold and evaluate the accuracy of hail monitoring. On the basis of hail survey data from 2009 to 2022, the maximum relative error is 9.08%, the average error is 5.62%, and the standard deviation is 1.66%. The spatial overlay analysis and spatial correlation analysis methods were used to quantitative analyzed hail frequency in different disaster-prone environments, such as landform types, terrain undulations, slopes, and terrain types, at the level of cultivated land plots. The proposed hail disaster risk assessment model calculates the spatial distribution characteristics of hail risk caused by natural conditions, such as climate, meteorology, terrain, and topography. Hail disasters in mountainous areas are significantly correlated with altitude and exhibit moderate correlation with slope and undulation. Hailstones typically move along mountain ranges and valleys, making farmlands along these valleys susceptible to hail disasters.The advantages of this model are as follows. (1) Parameter adaptation for multilevel grid models is used to improve model adaptability under 3D climate conditions of mountainous areas, increasing the accuracy of hail monitoring and risk assessment from county scale to cultivated land plot scale. (2) Spatial correlation quantitative analysis is conducted between the spatial distribution of hail disasters and terrain, such as altitude, slope, undulation, river valleys, valleys, and ridges at the scale of cultivated land plots. (3) The hail susceptibility assessment model is constructed at the cultivated land plot scale. Research results contribute to the rational adjustment of the crop planting structure, the planning and layout of artificial hail control operation points, and the reduction of hail disaster losses.  
      关键词:Hail disaster;hail remote sensing identification index (RNDVI_M);Hail Disasters Remote Sensing Monitoring;Temporal and spatial distribution of hail disasters;Honghe   
      1073
      |
      269
      |
      0
      <HTML>
      <L-PDF><Enhanced-PDF><Meta-XML>
      <引用本文> <批量引用> 67910010 false
      发布时间:2024-12-30
    • 在矿区地面沉降预测领域,专家提出了SFF-PredRNN模型,为矿区沉降灾害防治提供有效数据支撑。
      GUO Xiaowei,CHEN Tao
      Vol. 28, Issue 11, Pages: 3016-3031(2024) DOI: 10.11834/jrs.20243488
      Spatial and temporal prediction of ground subsidence in mining areas considering seasonal characteristics
      摘要:Mining can cause severe ground subsidence, which is frequently accompanied by widespread and uneven characteristics, posing considerable threats to production and life in a mining community. The timely and accurate monitoring and prediction of ground subsidence in mining areas are crucial for mitigating its adverse effects. However, traditional spatiotemporal prediction models for ground subsidence often experience difficulty in capturing comprehensive spatiotemporal information and learning the intricate features associated with this phenomenon.To address these challenges, this study incorporates a temporal decomposition strategy into a deep learning network model, resulting in the development of the Spatiotemporal Forecasting Framework (SFF-PredRNN) model. This innovative approach considers seasonal displacement features, enhancing the model’s ability to accurately capture complex spatiotemporal patterns. By integrating this advanced methodology, the SFF-PredRNN model offers improved predictive capabilities, allowing for effective mitigation measures against ground subsidence and its associated risks.The focus of this study is the Micun coal mine located in Xinmi City, a region characterized by extensive mineral resource extraction and distinct seasonal variations in rainfall. The summer season contributes significantly to the annual rainfall, accounting for 60.9%. Certain mining areas within this region have experienced notable ground subsidence issues. By using the small baseline set interference technique algorithm, ground subsidence data from 2018 to 2021 were collected for the study area. The analysis revealed distinct spatial differences in subsidence patterns, particularly in the Mengzhuang and Zhangpocun coal mines at the center and the Wangzhuang coal mine in the southwest. These areas exhibited severe ground subsidence problems, with the maximum subsidence reaching 256 mm, while the surrounding regions did not experience significant ground subsidence. A spatiotemporal dataset of ground subsidence was constructed based on the collected information, and the developed SFF-PredRNN model was employed for prediction. The model’s accuracy was assessed using metrics, such as mean absolute error, root mean square deviation, peak signal-to-noise ratio, and structural similarity index measure. Meanwhile, to assist in verifying the advantages of the model in the spatiotemporal prediction of the mine area, we selected a profile line that crossed the mine area in the horizontal and vertical directions and chose equal spacing to take out a certain number of subsidence points. Then, we extracted the subsidence values predicted by the model through these points and verified the results. The results demonstrated that the SFF-PredRNN model, as proposed in this study, exhibited superior accuracy in predicting subsidence for the years 2019, 2020, and 2021. This finding highlights the model’s strengths in the temporal and spatial predictions of ground subsidence. The predictions for the upcoming year indicated a continued trend of subsidence in the mining areas of Mengzhuang, Wangzhuang, and Zhangpocun, with an expected maximum cumulative subsidence of 274.3 mm. The spatial distribution of settlement in the study area remained consistent with previous patterns.In conclusion, the SFF-PredRNN model proposed in this study exhibits good performance in the spatiotemporal prediction of ground subsidence, and thus, it can be used as an effective method for the spatiotemporal prediction of ground subsidence. This study provides effective methodological guidance for the prevention and early warning of ground subsidence disasters in mining areas. In the future, we can improve the prediction model by integrating more data on ground subsidence influencing factors to realize more accurate spatiotemporal prediction on a large scale.  
      关键词:remote sensing;InSAR;ground subsidence;spatio-temporal prediction;seasonality;SFF-PredRNN   
      162
      |
      153
      |
      0
      <HTML>
      <L-PDF><Enhanced-PDF><Meta-XML>
      <引用本文> <批量引用> 56625186 false
      发布时间:2024-12-30
    • 新密市地面沉降预测研究取得进展,专家构建TSC-LSTM模型,为城市地面沉降研究提供支持。
      ZHAO Hewen,CHEN Tao
      Vol. 28, Issue 11, Pages: 3032-3044(2024) DOI: 10.11834/jrs.20243530
      Ground subsidence prediction model in Xinmi City based on TSC—LSTM
      摘要:Predicting land subsidence is crucial for conducting in-depth analyses of and providing early warnings for urban land subsidence patterns. However, traditional numerical prediction models frequently experience difficulty in accurately capturing the intricate characteristics of land subsidence data, leading to less precise predictions. This study focuses on Xinmi City and endeavors to improve the accuracy of land subsidence prediction by combining time series feature extraction methods with time series prediction techniques.In this study, 232 interference images provided by HyP3 were utilized to acquire land subsidence information in Xinmi City from January 2018 to December 2022, employing Small Baseline Subset (SBAS)-Interferometric Synthetic Aperture radar (InSAR) technology. Recognizing the challenge of achieving high accuracy in directly predicting land subsidence data, this study developed a land subsidence prediction model that integrates trend and seasonal characteristics by using a Long Short-term Memory (LSTM) network, namely, the Trend Seasonal Characteristics-LSTM (TSC-LSTM). The TSC-LSTM model capitalizes on the strengths of weighted regression seasonal trend decomposition (STL) in extracting time series features from settlement data and the LSTM model for addressing the vanishing gradient problem in time series prediction.This fusion of techniques allows for a precise analysis of land subsidence data and enables highly accurate predictions. Distinguishing itself from the conventional LSTM model, the TSC-LSTM model refrains from directly inputting ground subsidence data. Instead, it employs STL to meticulously extract trend and seasonal characteristics from land subsidence data. This approach maximizes the utilization of characteristic information inherent in land subsidence data. Subsequently, these features are fed into the LSTM model for prediction. This unique methodology reduces noise interference and significantly enhances the accuracy of model predictions.This research leverages time-series InSAR data for Xinmi City from 2018 to 2022. It employs the TSC-LSTM model, deep learning architectures (recurrent neural network and LSTM), and conventional machine learning algorithms (multilayer perceptron and support vector regression) to forecast the cumulative subsidence data for five subsidence centers by using SBAS-InSAR. This study identifies the two most optimal models and validates their efficacy in single-point prediction scenarios, utilizing domain-specific terminologies. Research findings indicate the following. (1) Between 2018 and 2022, Xinmi City experienced a land subsidence rate that ranged from -60.3 mm to 51.96 mm per annum, resulting in the identification of five distinct land subsidence center areas. Among these, the highest cumulative settlement and uplift reached 304.9 mm and 197.68 mm, respectively. The universality of the TSC-LSTM model across diverse datasets has been corroborated, demonstrating its high precision, exceptional generalization capability, and stable high performance in the prediction of land subsidence, employing specialized terminologies. (2) The TSC-LSTM model exhibited exceptional performance in predicting the five subsidence center areas. The R² values for the TSC-LSTM model range from 0.9985 to 0.9992, significantly surpassing the second-best model, i.e., LSTM, which has an R² range of 0.9662 to 0.9872. Moreover, the root mean square error values for the prediction accuracy of the TSC-LSTM model are less than 2 mm, achieving a range of 1.2426 mm to 1.7403 mm . (3) Single-point prediction results demonstrate the superior ability of the TSC-LSTM model to accurately capture local changes in the cumulative settlement data. The TSC-LSTM model proposed in this study outperforms the traditional LSTM model in terms of prediction accuracy and model stability, providing robust support for in-depth research on urban land subsidence.  
      关键词:Ground settlement prediction;TSC—LSTM;SBAS—InSAR;Cumulative settlement data decomposition;Xinmi City;LSTM   
      445
      |
      380
      |
      0
      <HTML>
      <L-PDF><Enhanced-PDF><Meta-XML>
      <引用本文> <批量引用> 54662966 false
      发布时间:2024-12-30
    0