The rising need for precision farming and sustainable land management has catalyzed the requirement for sophisticated means of deriving practical data from remote sensing images.Image segmentation,or the process of di...The rising need for precision farming and sustainable land management has catalyzed the requirement for sophisticated means of deriving practical data from remote sensing images.Image segmentation,or the process of dividing the image into semantically relevant parts,has become a groundbreaking technology that allows resolving the problem of transitioning the pixel-level data to a parcel-level analysis.This review is a synthesis of the segmentation methods and their use in crop research and geospatial science.The architectures of pixel-based,object-based,and deep learning(convolutional neural networks,U-Net,Mask R-CNN,and Transformer models)are considered in terms of principles,capabilities,and limitations.Multi-spectral,hyperspectral,LiDAR,and SAR data are integrated to improve the efficiency of segmentation,allowing the possible delineation of fields,the classification of crops,health monitoring,monitoring of yields,and stress identification.In addition to agriculture,segmentation helps in land use and land cover mapping,identification of temporal change,monitoring of the environment,and is used in combination with GIS-based spatial modeling.Nevertheless,issues related to data heterogeneity,mixed pixels,computational requirements,and inadequate availability of labelled data still exist despite the major progress.The future directions involve multi-source data fusion,pixel-to-parcel pipeline automation,and predictive models based on AI,which are used to enhance its scalability,robustness,and the ability to monitor in real-time.This review makes it clear that the use of image segmentation as a tool in generating precision agriculture,sustainable land use,and informed geospatial.展开更多
Rapid urbanization and digital transformation are reshaping how cities address challenges related to security,governance,and sustainable development.Geospatial information technology(GIT)has become a base infrastructu...Rapid urbanization and digital transformation are reshaping how cities address challenges related to security,governance,and sustainable development.Geospatial information technology(GIT)has become a base infrastructure for smart cities,where the gathering,synthesis,and examination of spatially explicit information are used to deliver data to make decisions in cities.Even after its increasing significance,the current body of research on geospatial innovation is still divided into the spheres of urban security,spatial governance,and smart city development.Such fragmentation restricts the integration of theoretical work,as well as limits the possibility of developing coherent policies and governance institutions.This article includes a systematic and integrative review of innovation in geospatial information technology to analyze the relationship between technological,data-driven,and institutional innovation and urban security practices,spatial governance processes,and smart city initiatives.Based on peer-reviewed sources on urban studies,geography,planning,and information science,the review generalizes the main trends in real-time spatial analytics,artificial intelligence,participatory geospatial platforms,and urban digital twins.The review shows that geospatial systems facilitate anticipatory governance,cross-sector coordination,and evidence-based urban management,and that it provides an integrative conceptual lens on the existing literature on smart cities and urban governance,as it positions geospatial information technology as a socio-technical infrastructure,as opposed to a technical tool.The paper recognizes the voids in critical research and the directions into the future on how to build ethical,inclusive,and context-sensitive geospatial systems that can allow the creation of secure,governable,and sustainable urban futures.展开更多
Accurate fine-grained geospatial scene classification using remote sensing imagery is essential for a wide range of applications.However,existing approaches often rely on manually zooming remote sensing images at diff...Accurate fine-grained geospatial scene classification using remote sensing imagery is essential for a wide range of applications.However,existing approaches often rely on manually zooming remote sensing images at different scales to create typical scene samples.This approach fails to adequately support the fixed-resolution image interpretation requirements in real-world scenarios.To address this limitation,we introduce the million-scale fine-grained geospatial scene classification dataset(MEET),which contains over 1.03 million zoom-free remote sensing scene samples,manually annotated into 80 fine-grained categories.In MEET,each scene sample follows a scene-in-scene layout,where the central scene serves as the reference,and auxiliary scenes provide crucial spatial context for fine-grained classification.Moreover,to tackle the emerging challenge of scene-in-scene classification,we present the context-aware transformer(CAT),a model specifically designed for this task,which adaptively fuses spatial context to accurately classify the scene samples.CAT adaptively fuses spatial context to accurately classify the scene samples by learning attentional features that capture the relationships between the center and auxiliary scenes.Based on MEET,we establish a comprehensive benchmark for fine-grained geospatial scene classification,evaluating CAT against 11 competitive baselines.The results demonstrate that CAT significantly outperforms these baselines,achieving a 1.88%higher balanced accuracy(BA)with the Swin-Large backbone,and a notable 7.87%improvement with the Swin-Huge backbone.Further experiments validate the effectiveness of each module in CAT and show the practical applicability of CAT in the urban functional zone mapping.The source code and dataset will be publicly available at https://jerrywyn.github.io/project/MEET.html.展开更多
Agricultural drought,characterized by insufficient soil moisture crucial for crop growth,poses significant chal lenges to food security and economic sustainability,particularly in water-scarce regions like Senegal.Thi...Agricultural drought,characterized by insufficient soil moisture crucial for crop growth,poses significant chal lenges to food security and economic sustainability,particularly in water-scarce regions like Senegal.This study addresses this issue by developing a comprehensive geospatial monitoring system for agricultural drought using the Regional Hydrologic Extremes Assessment System(RHEAS).This system,with a high-resolution of 0.05°,effectively simulates daily soil moisture and generates the Soil Moisture Deficit Index(SMDI)-based agricultural drought monitoring.The SMDI derived from the RHEAS has effectively captured historical droughts in Senegal over the recent 30 years period from 1993 to 2022.The SMDI,also provides a comprehensive understanding of regional variations in drought severity(S),duration(D),and frequency(F),through S-D-F analysis to identify key drought hotspots across Senegal.Findings reveal a distinct north-south gradient in drought conditions,with the northern and central Senegal experiencing more frequent and severe droughts.The study highlights that Senegal experiences frequent short-duration droughts with high severity,resulting in extensive spatial impact.Addition ally,increasing trends in drought severity and duration suggest evolving climate change effects.These findings emphasize the urgent need for sustainable interventions to mitigate drought impacts on agricultural productiv ity.Specifically,the study identifies recurrent and intense drought hotspots affecting yields of staple crops like maize and rice,as well as cash crops like peanuts.The developed high-resolution drought monitoring system for Senegal not only identifies hotspots but also enables prioritizing sustainable approaches and adaptive strategies,ultimately sustaining agricultural productivity and resilience in Senegal’s drought-prone regions.展开更多
The evolving landscape of geospatial science is marked by a fundamental shift-from static spatial sensing to dynamic spatial intelligence.This transformation is driven not only by advances in data acquisition and comp...The evolving landscape of geospatial science is marked by a fundamental shift-from static spatial sensing to dynamic spatial intelligence.This transformation is driven not only by advances in data acquisition and computation but also by the growing demand for intelligent systems that automate perception,support decision-making,and adapt across diverse environments.Three recent studies published in Revue Internationale de Géomatique offer valuable insights into this trajectory,highlighting how methodological innovation in remote sensing(RS)and geographic information system(GIS)is laying the foundation for the next generation of smart geospatial applications.展开更多
Nowadays,spatiotemporal information,positioning,and navigation services have become critical components of new infrastructure.Precise positioning technology is indispensable for determining spatiotemporal information ...Nowadays,spatiotemporal information,positioning,and navigation services have become critical components of new infrastructure.Precise positioning technology is indispensable for determining spatiotemporal information and providing navigation services.展开更多
为实现遥感数字图像的快速显示,开发了基于QT和GDAL(Geospatial Data Abstraction Library)的遥感图像快速显示程序,并以ERDAS IMAGINE标准数据格式IMG图像文件为例,给出了Windows系统下IMG图像显示和坐标实时显示程序的环境配置过程、...为实现遥感数字图像的快速显示,开发了基于QT和GDAL(Geospatial Data Abstraction Library)的遥感图像快速显示程序,并以ERDAS IMAGINE标准数据格式IMG图像文件为例,给出了Windows系统下IMG图像显示和坐标实时显示程序的环境配置过程、开发框架、建立流程以及功能模块的实现。程序采用开源GDAL类库,以图形用户界面框架QT作为开发工具,Visual Studio 2008作为开发平台,采取多线程分块处理方法提取IMG数据中波段信息和坐标信息,实现IMG图像的快速显示以及坐标的实时显示。同时结合QT和GDAL进行开源程序的开发,改变了依赖于宿主软件进行二次开发的程序开发模式。研究结果表明,对于1.8 GByte遥感图像的显示,采用多线程分块处理方法,与单纯使用Raster IO()函数相比,处理时间缩短了2.7 s,提高了图像读取和显示效率以及程序开发的自主性,满足了大数据量的应用需求。展开更多
采用熔盐顶部籽晶法从K 2 Mo 3 O 10-B 2 O 3助熔剂中生长出尺寸为20 mm的优质GdAl 3(BO 3)4(简称GAB)和Nd^3+激活的自变频激光晶体。确定了GAB晶体的透光波长范围、折射率和倍频系数随波长的变化,结果表明其在整个透光范围内均可实现...采用熔盐顶部籽晶法从K 2 Mo 3 O 10-B 2 O 3助熔剂中生长出尺寸为20 mm的优质GdAl 3(BO 3)4(简称GAB)和Nd^3+激活的自变频激光晶体。确定了GAB晶体的透光波长范围、折射率和倍频系数随波长的变化,结果表明其在整个透光范围内均可实现相位匹配。测定了Nd^3+∶GAB晶体在室温下的偏振吸收、荧光光谱和荧光寿命,进行了光谱计算,测试了晶体的自变频激光性能,实现了紫外-可见光-红外-中红外多波段激光输出。展开更多
基金supported under the 2024 Foshan City Self-Funded Science and Technology Innovation Project“Research on Image Segmentation Technology Based on Convolutional Neural Networks in Crop Images”(Project Number:2420001004686).
文摘The rising need for precision farming and sustainable land management has catalyzed the requirement for sophisticated means of deriving practical data from remote sensing images.Image segmentation,or the process of dividing the image into semantically relevant parts,has become a groundbreaking technology that allows resolving the problem of transitioning the pixel-level data to a parcel-level analysis.This review is a synthesis of the segmentation methods and their use in crop research and geospatial science.The architectures of pixel-based,object-based,and deep learning(convolutional neural networks,U-Net,Mask R-CNN,and Transformer models)are considered in terms of principles,capabilities,and limitations.Multi-spectral,hyperspectral,LiDAR,and SAR data are integrated to improve the efficiency of segmentation,allowing the possible delineation of fields,the classification of crops,health monitoring,monitoring of yields,and stress identification.In addition to agriculture,segmentation helps in land use and land cover mapping,identification of temporal change,monitoring of the environment,and is used in combination with GIS-based spatial modeling.Nevertheless,issues related to data heterogeneity,mixed pixels,computational requirements,and inadequate availability of labelled data still exist despite the major progress.The future directions involve multi-source data fusion,pixel-to-parcel pipeline automation,and predictive models based on AI,which are used to enhance its scalability,robustness,and the ability to monitor in real-time.This review makes it clear that the use of image segmentation as a tool in generating precision agriculture,sustainable land use,and informed geospatial.
基金project supported by the Scientific Research Fund of the Zhejiang Provincial Education Department(grant number Y202456064).
文摘Rapid urbanization and digital transformation are reshaping how cities address challenges related to security,governance,and sustainable development.Geospatial information technology(GIT)has become a base infrastructure for smart cities,where the gathering,synthesis,and examination of spatially explicit information are used to deliver data to make decisions in cities.Even after its increasing significance,the current body of research on geospatial innovation is still divided into the spheres of urban security,spatial governance,and smart city development.Such fragmentation restricts the integration of theoretical work,as well as limits the possibility of developing coherent policies and governance institutions.This article includes a systematic and integrative review of innovation in geospatial information technology to analyze the relationship between technological,data-driven,and institutional innovation and urban security practices,spatial governance processes,and smart city initiatives.Based on peer-reviewed sources on urban studies,geography,planning,and information science,the review generalizes the main trends in real-time spatial analytics,artificial intelligence,participatory geospatial platforms,and urban digital twins.The review shows that geospatial systems facilitate anticipatory governance,cross-sector coordination,and evidence-based urban management,and that it provides an integrative conceptual lens on the existing literature on smart cities and urban governance,as it positions geospatial information technology as a socio-technical infrastructure,as opposed to a technical tool.The paper recognizes the voids in critical research and the directions into the future on how to build ethical,inclusive,and context-sensitive geospatial systems that can allow the creation of secure,governable,and sustainable urban futures.
基金supported by the National Natural Science Foundation of China(42030102,42371321).
文摘Accurate fine-grained geospatial scene classification using remote sensing imagery is essential for a wide range of applications.However,existing approaches often rely on manually zooming remote sensing images at different scales to create typical scene samples.This approach fails to adequately support the fixed-resolution image interpretation requirements in real-world scenarios.To address this limitation,we introduce the million-scale fine-grained geospatial scene classification dataset(MEET),which contains over 1.03 million zoom-free remote sensing scene samples,manually annotated into 80 fine-grained categories.In MEET,each scene sample follows a scene-in-scene layout,where the central scene serves as the reference,and auxiliary scenes provide crucial spatial context for fine-grained classification.Moreover,to tackle the emerging challenge of scene-in-scene classification,we present the context-aware transformer(CAT),a model specifically designed for this task,which adaptively fuses spatial context to accurately classify the scene samples.CAT adaptively fuses spatial context to accurately classify the scene samples by learning attentional features that capture the relationships between the center and auxiliary scenes.Based on MEET,we establish a comprehensive benchmark for fine-grained geospatial scene classification,evaluating CAT against 11 competitive baselines.The results demonstrate that CAT significantly outperforms these baselines,achieving a 1.88%higher balanced accuracy(BA)with the Swin-Large backbone,and a notable 7.87%improvement with the Swin-Huge backbone.Further experiments validate the effectiveness of each module in CAT and show the practical applicability of CAT in the urban functional zone mapping.The source code and dataset will be publicly available at https://jerrywyn.github.io/project/MEET.html.
基金supported by the NASA(Grant No.80NSSC21K0403)USAID Kansas State University subcontract KSU-A20-0163-S035 with Michigan State University.
文摘Agricultural drought,characterized by insufficient soil moisture crucial for crop growth,poses significant chal lenges to food security and economic sustainability,particularly in water-scarce regions like Senegal.This study addresses this issue by developing a comprehensive geospatial monitoring system for agricultural drought using the Regional Hydrologic Extremes Assessment System(RHEAS).This system,with a high-resolution of 0.05°,effectively simulates daily soil moisture and generates the Soil Moisture Deficit Index(SMDI)-based agricultural drought monitoring.The SMDI derived from the RHEAS has effectively captured historical droughts in Senegal over the recent 30 years period from 1993 to 2022.The SMDI,also provides a comprehensive understanding of regional variations in drought severity(S),duration(D),and frequency(F),through S-D-F analysis to identify key drought hotspots across Senegal.Findings reveal a distinct north-south gradient in drought conditions,with the northern and central Senegal experiencing more frequent and severe droughts.The study highlights that Senegal experiences frequent short-duration droughts with high severity,resulting in extensive spatial impact.Addition ally,increasing trends in drought severity and duration suggest evolving climate change effects.These findings emphasize the urgent need for sustainable interventions to mitigate drought impacts on agricultural productiv ity.Specifically,the study identifies recurrent and intense drought hotspots affecting yields of staple crops like maize and rice,as well as cash crops like peanuts.The developed high-resolution drought monitoring system for Senegal not only identifies hotspots but also enables prioritizing sustainable approaches and adaptive strategies,ultimately sustaining agricultural productivity and resilience in Senegal’s drought-prone regions.
文摘The evolving landscape of geospatial science is marked by a fundamental shift-from static spatial sensing to dynamic spatial intelligence.This transformation is driven not only by advances in data acquisition and computation but also by the growing demand for intelligent systems that automate perception,support decision-making,and adapt across diverse environments.Three recent studies published in Revue Internationale de Géomatique offer valuable insights into this trajectory,highlighting how methodological innovation in remote sensing(RS)and geographic information system(GIS)is laying the foundation for the next generation of smart geospatial applications.
文摘Nowadays,spatiotemporal information,positioning,and navigation services have become critical components of new infrastructure.Precise positioning technology is indispensable for determining spatiotemporal information and providing navigation services.
文摘为实现遥感数字图像的快速显示,开发了基于QT和GDAL(Geospatial Data Abstraction Library)的遥感图像快速显示程序,并以ERDAS IMAGINE标准数据格式IMG图像文件为例,给出了Windows系统下IMG图像显示和坐标实时显示程序的环境配置过程、开发框架、建立流程以及功能模块的实现。程序采用开源GDAL类库,以图形用户界面框架QT作为开发工具,Visual Studio 2008作为开发平台,采取多线程分块处理方法提取IMG数据中波段信息和坐标信息,实现IMG图像的快速显示以及坐标的实时显示。同时结合QT和GDAL进行开源程序的开发,改变了依赖于宿主软件进行二次开发的程序开发模式。研究结果表明,对于1.8 GByte遥感图像的显示,采用多线程分块处理方法,与单纯使用Raster IO()函数相比,处理时间缩短了2.7 s,提高了图像读取和显示效率以及程序开发的自主性,满足了大数据量的应用需求。
文摘采用熔盐顶部籽晶法从K 2 Mo 3 O 10-B 2 O 3助熔剂中生长出尺寸为20 mm的优质GdAl 3(BO 3)4(简称GAB)和Nd^3+激活的自变频激光晶体。确定了GAB晶体的透光波长范围、折射率和倍频系数随波长的变化,结果表明其在整个透光范围内均可实现相位匹配。测定了Nd^3+∶GAB晶体在室温下的偏振吸收、荧光光谱和荧光寿命,进行了光谱计算,测试了晶体的自变频激光性能,实现了紫外-可见光-红外-中红外多波段激光输出。
