检索结果-内蒙古大学图书馆

object-based Multi-Temporal and Multi-Source Land Cover Mapping Leveraging Hierarchical Class Relationships

REMOTE SENSING 2020年第17期12卷 2814-2814页

作者： Gbodjo, Yawogan Jean Eudes Ienco, Dino Leroux, Louise Interdonato, Roberto Gaetano, Raffaele Ndao, Babacar Univ Montpellier UMR TETIS INRAE F-34000 Montpellier France Univ Montpellier LIRMM F-34095 Montpellier France UPR AIDA Cirad Dakar Senegal Univ Montpellier Cirad AIDA F-34398 Montpellier France UMR TETIS Cirad F-34000 Montpellier France Ctr Suivi Ecol Dakar 15532 Senegal Univ Cheikh Anta Diop Dakar 15532 Senegal

European satellite missions Sentinel-1 (S1) and Sentinel-2 (S2) provide at high spatial resolution and high revisit time, respectively, radar and optical images that support a wide range of Earth surface monitoring tasks, such as Land Use/Land Cover mapping. A long-standing challenge in the remote sensing community is about how to efficiently exploit multiple sources of information and leverage their complementarity, in order to obtain the most out of radar and optical data. In this work, we propose to deal with land cover mapping in an object-based image analysis (OBIA) setting via a deep learning framework designed to leverage the multi-source complementarity provided by radar and optical satellite image time series (SITS). The proposed architecture is based on an extension of Recurrent Neural Network (RNN) enriched via a modified attention mechanism capable to fit the specificity of SITS data. Our framework also integrates a pretraining strategy that allows to exploit specific domain knowledge, shaped as hierarchy over the set of land cover classes, to guide the model training. Thorough experimental evaluations, involving several competitive approaches were conducted on two study sites, namely the Reunion island and a part of the Senegalese groundnut basin. Classification results, 79% of global accuracy on the Reunion island and 90% on the Senegalese site, respectively, have demonstrated the suitability of the proposal.

关键词： land cover classification multi-source remote sensing satellite image time series object based image analysis deep learning neural networks pretraining

来源：评论

学校读者我要写书评

暂无评论

Parcel level staple crop type identification based on newly defined red-edge vegetation indices and ORNN

引用

COMPUTERS AND ELECTRONICS IN AGRICULTURE 2023年第1期211卷

作者： Xu, Lu Ming, Dongping Zhang, Liang Dong, Dehui Qing, Yuanzhao Yang, Jianyu Zhou, Chenghu China Univ Geosci Beijing Sch Informat Engn Beijing 100083 Peoples R China Minist Nat Resources China Polytech Ctr Nat Resources Big data Beijing 100036 Peoples R China China Agr Univ Coll Land Sci & Technol Beijing 100083 Peoples R China Chinese Acad Sci Inst Geog Sci & Nat Resources Res State Key Lab Resources & Environm Informat Syst Beijing 100101 Peoples R China

It is of strategic importance for early estimating planting area and crop yield to identify staple crop types on parcel level in a remotely sensed way. Limited by remote sensor imaging mechanism and current sensor technology, remote sensing images usually cannot be of high resolution in both spatial and spectral aspects. So, it has been found in many studies that supplementary data or features such as synthetic aperture radar data or temporal characteristic of time series data are usually integrated for crop type identification, which demands lots of data and overlooks the most typical spectral characteristics. Thanks to the expanded detecting bands of several moderate resolution optical satellites, the red-edge bands reflecting specific spectral features of crops are available for precise classification. Nevertheless, only a few studies have concentrated on these bands and the potential value of red-edge bands has not been sufficiently stretched in practical use. In this study, two newly defined red-edge vegetation indices (REVIs) and their time series data have been used for distinguishing different crops efficiently. Firstly, according to the reflectance characteristics in optical images and the data distribution of crops in feature space, two kinds of variants of REVIs based on green band and red-edge band, namely, ReG_RVI and ReG_NDVI have been proposed. Then, the time series data of REVIs can be generated from remote sensing images and be fed into the recurrent neural network (RNN) as training data for crop type classification. Finally, object based image analysis and RNN are combined with the idea of majority voting to achieve accurate crop type identification on cultivated land parcel level. The experimental results of two study areas using time series images captured by Gaofen 6 wide field of view sensor show an overall accuracy of above 90 % in identifying crop types, which proves the effectiveness of the newly defined REVIs and the feasibility of the prop

关键词： Crop type Remote sensing Red-edge vegetation index Recurrent neural network object based image analysis

来源：评论

学校读者我要写书评

暂无评论

Urban Development in West Africa-Monitoring and Intensity analysis of Slum Growth in Lagos: Linking Pattern and Process

引用

REMOTE SENSING 2018年第7期10卷 1044-1044页

作者： Badmos, Olabisi S. Rienow, Andreas Callo-Concha, Daniel Greve, Klaus Juergens, Carsten Univ Bonn Ctr Dev Res Genscherallee 3 D-53113 Bonn Germany Univ Bonn Dept Geog Meckenheimer Allee 166 D-53115 Bonn Germany Ruhr Univ Bochum Dept Geog Univ Str 104Bldg UNI 104 D-44799 Bochum Germany

Upgrading all slums in Lagos by 2030 will an ambitious task, given that more than 70% of its residents resides in slums. Furthermore, there is no recent study identifying neither the slums nor their temporal growth/development pattern in Lagos that can backstop any slum management initiative. This study aims to contribute by applying object-based image analysis and intensity analysis to map and link patterns and processes of slum growth in Lagos. RapidEye imagery from 2009 and 2015 were used to create maps for each time point for six land use categories (water, vegetated area, open space, road, slum, and other urban). Intensity analysis was applied to quantify the annual intensity of changes at the category and transition level. An overall accuracy (and kappa coefficient) of 94% (0.9) and 89% (0.86) were achieved for the 2009 and 2015 land use and land cover maps, respectively. This study showed that slums in Lagos have increased spatially during the time interval studied, with a total net gain of 9.18 square kilometers, influenced by the increase in population, mainly due to in-migration to Lagos. However, this study also revealed that slums were actively losing and gaining land area between 2009 and 2015, with an annual gain and loss intensity of 10.08 and 6.41, respectively, compared to the uniform intensity of 3.15. The gain was due to poor maintenance of buildings and encroachment onto available spaces (water and open space), while the loss was attributed to gentrification and demolition processes. A systematic process of transition was observed between slums and other urban (and open space) areas in the interval studied, and this process was mainly influenced by the Lagos state government. This analysis is crucial for designing policy interventions to manage slum growth in Lagos.

关键词： slum object based image analysis intensity analysis Lagos

来源：评论

学校读者我要写书评

暂无评论

Random Forest Classification for Mangrove Land Cover Mapping Using Landsat 5 TM and Alos Palsar imageries

引用

Procedia Environmental Sciences 2015年 24卷 215-221页

作者： Romie Jhonnerie Vincentius P. Siregar Bisman Nababan Lilik Budi Prasetyo Sam Wouthuyzen Marine Technology Graduate Program Faculty of Fisheries and Marine Science Bogor Agricultural University Bogor Indonesia Departmen of Aquatic Resources Utilization Faculty of Fisheries and Marine Science Riau University. Indonesia Department of Marine Science and Technology Faculty of Fisheries and Marine Science Bogor Agricultural University Bogor Indonesia Departmen of Forest Resources Conservation and Ecotourism Faculty of Forestry Bogor Agricultural University Bogor Indonesia Research Center for Oceanography Indonesian Institute of Science Jl. Pasir Putih I Ancol Timur Jakarta Indonesia

The objective of this research was to evaluate the accuracy of random forest classification rule using object based image analysis (OBIA) application (eCognition Developer) and the results were compared with common pixel-based classification algorithm (maximum likelihood/ML) for mangrove land cover mapping in Kembung River, Bengkalis Island, Indonesia. Seven data input model derived from Landsat 5TM bands, ALOS PALSAR FBD, and spectral transformations (NDVI, NDWI, NDBI) were examined by both classifiers. Feature objects statistical parameters were selected and implemented on random forest classifier. Overall accuracy (OA) as well as user and producer accuracies and Kappa statistic were used to compare classification results. Our results showed that the more data model used produced higher overall accuracy and kappa statistics for RF classifier. For each data input model, random forest classifier has higher overall accuracy than maximum likelihood. The best mangrove discrimination in RF classifier was achieved when the combination of Landsat 5 TM, SAR, and spectral transformation were used, while in ML classifier, the best mangrove discrimination was achieved when the combination of Landsat 5 TM and ALOS PALSAR was used. The overall accuracy achieved by RF classifier was 81.1% and 0.76 for Kappa statistic. Meanwhile, for ML classifier, the overall accuracy achieved was 77.7% and 0.71 for Kappa statistic.

关键词： Random forest classification object based image analysis pixel based classification Landsat TM ALOS PALSAR mangrove Kembung River Bengkalis island

来源：评论

学校读者我要写书评

暂无评论

Forest Habitat Change Dynamics in a Riparian Wetland

引用

Procedia Environmental Sciences 2011年 7卷 371-376页

作者： Szilvia Kollár Zoltán Vekerdy Béla Márkus Faculty of Geoinformatics University of West Hungary Székesfehérvár 8000 Hungary Department of Water Resources ITC Enschede 7500 AE The Netherlands

Our study aims at monitoring ecosystem dynamics in the Szigetköz Danube floodplain (Hungary), where the wetland communities have been seriously affected by the abrupt hydrological changes after the Danube-diversion in 1992. Wetland habitats were mapped in the last years for a chosen area, based on a comparable use of archived analogue and digital aerial images and available silvicultural and botanical inventories. During the object-based image analysis the classification was mainly based on texture measures. The resulting map time series show habitat borders for each year and present the changes separately for each class. The method is suitable for the characterization of spatial and temporal structure of wetland land cover in a realistic way.

关键词： remote sensing object based image analysis wetland change detection aerial imagery texture based classification

来源：评论

学校读者我要写书评

暂无评论

Fast Extraction of Roads for Emergencies with Segmentation of Satellite imagery

引用

Procedia - Social and Behavioral Sciences 2016年 223卷 903-908页

作者： Vincenzo Barrile Giuliana Bilotta DICEAM - Faculty of Engineering “Mediterranea” University of Reggio Calabria Via Graziella Feo di Vito 89100 Reggio Calabria Italy Department of Planning University IUAV of Venice Santa Croce 191 Tolentini 30135 Venice Italy

Among the problems that national and international organizations often face in rescue operations following earthquakes there is the knowledge of the roads that allow aid to reach its destination. During the Haiti earthquake on 2010-01-12 and the Nepal earthquake on 2015-04-25 proved to be very useful the work of volunteers creating maps of the roads on site, useful in case of emergency. Though away from the site of interest, with such crowdsourcingactions,volunteers can contribute to the efforts of mapping inaccessible or little known regions by tracing images from aerial photographs or satellite images, in some cases made available free of charge. The objective of this research, still in progress, is to experience for obtaining very quickly the road network, by extracting it automatically or semi-automatically from satellite data with the object based image analysis, with a segmentation based on Mathematical Morphology.

关键词： Segmentation Satellite imagery Mathematical morphology Gis object based image analysis

来源：评论

学校读者我要写书评

暂无评论

Mapping stony rise landforms using a novel remote sensing, geophysical, and machine learning approach

引用

GEOMORPHOLOGY 2024年 450卷

作者： Fraser, Shaye Soto-Berelov, Mariela Holden, Lucas Hewson, Robert Webb, John Jones, Simon RMIT Univ Sch Sci Melbourne Australia Univ Twente Fac Geoinformat Sci & Earth Observat ITC Dept Appl Earth Sci Enschede Netherlands La Trobe Univ Discipline Ecol & Environm Melbourne Vic Australia

A basalt lava plain covers much of central and western Victoria in Australia. Many of the younger flows (<1 Ma) have hummocky surfaces with abundant tumuli and lava rises, known locally as stony rises due to the protruding basalt boulders. Stony rises are noteworthy geologically but have ecological importance and are culturally significant to Aboriginal Australians. Ecologically, stony rises host many native plant species. Culturally, stony rises provide evidence of Aboriginal land use and occupation through the presence of artefact deposits. Currently, mapping of stony rises is carried out by conducting field surveys on a study-by-study basis. This could be performed more rapidly using remote sensing datasets such as LiDAR, satellite and aerial imagery, and geophysical data. This research mapped stony rises in an area that is experiencing rapid urban development using a combination of remotely sensed datasets that characterise the morphology of stony rises through a Machine Learning approach that utilises object based image analysis (OBIA) segmentation and Random Forests (RF) classification techniques. The research successfully identified stony rises in the landscape to an accuracy of 89.97 %, with 314 potential new stony rise sites identified. The research identified the most important predictor variables were slope, local elevation, the concentrations of thorium and uranium (obtained from airborne radiometric data), aeromagnetic data (total magnetic intensity), the Normalised Difference Water Index (NDWI), and the Clay Mineral Ratio. The results from this study lay a foundation for standardising the definition of a stony rise within Australia. Furthermore, by utilising various predictor variables, this study highlights the success of a multifaceted approach into detecting a specific landform at a landscape level.

关键词： Stony rises Basalt outcrop Remote sensing Machine learning object based image analysis Random forests

来源：评论

学校读者我要写书评

暂无评论

Mapping burial mounds based on UAV-derived data in the Suusamyr Plateau, Kyrgyzstan

引用

JOURNAL OF ARCHAEOLOGICAL SCIENCE 2020年 123卷 105251-105251页

作者： Sarasan, Adriana Ardelean, Adrian-Cristian Balarie, Andrei Wehrheim, Ruben Tabaldiev, Kubatbek Akmatov, Kunbolot Natl Museum Banat Dept Archaeol Timisoara 300223 Romania Free Univ Berlin D-14195 Berlin Germany Kyrgyz Turkish Manas Univ Fac Humanities Dept Hist Bishkek Kyrgyzstan

Located in the northern part of the Tian Shan Mountains, at an average elevation of over 2200 m.a.s.l., the Suusamyr Plateau is home to numerous archaeological sites. Nonetheless, there is little to no information available regarding the spatial characteristics and preservation conditions of these complex burial grounds. During the recent years, unmanned aerial vehicles (UAVs) have offered the possibility of fast acquisition of high -resolution images that facilitate the identification and documentation of archaeological sites and features. Given these advantages over 1500 ha (29 individual sites) were investigated within the Suusamyr Plateau, using a DJI Phantom 4 Pro quadcopter. based on the UAV derived dataset, over 600 burial mounds were identified via manual mapping. However, as this is a time consuming and subjective process, a new method that integrates the multi-scale topographic analysis and the geomorphons approach is introduced here for the semi-automated extraction of burial mounds. Following an object-based image analysis (OBIA) routine, the proposed methodology was developed within a single site (T1), while the transferability degree of the final model was subsequently assessed within two other investigated sites (T2, T3). The accuracy of the obtained model was determined using the F1-Score (harmonic mean of precision and recall). The highest detection rate of 100% was achieved for test site T2, while the lowest value of 87.5% was obtained for site T3.

关键词： Unmanned aerial vehicle Digital surface model Geomorphons object based image analysis Burial mounds Suusamyr plateau

来源：评论

学校读者我要写书评

暂无评论

Soft image Segmentation Model

Soft Image Segmentation Model

引用

2012年遥感中的计算机视觉国际会议

作者： Xuehong Chen Jin Chen Yasushi Yamaguchi State Key Laboratory of Earth Surface Processes and Resource Ecology Beijing Normal University Beiji Graduate School of Environmental Studies Nagoya University Nagoya Japan

ISBN: (纸本)9781467312721

With the spatial resolution of remote sensing data increases,the information of spatial features become more and more important for interpretation of remote sensing ***,object based image analysis methods receive more and more *** segmentation is the most important step for object based *** segmentation algorithms have been developed,however,there are always uncertainties or errors in image ***,such uncertainty information was largely neglected in previous *** this paper,a soft segmentation model aiming to describe the uncertainties in the segmentation procedure was developed based on the multi-resolution segmentation,which is a bottom-up approach and follows a pair-wise object merging *** each merging step,the soft segmentation model calculates probabilities of several adjacent sub-objects merged into super-objects on the next *** combining the probability at each merging step,the final probability of each pixel merged into the super-objects on the top level can be acquired.A case study of an IKONOS image was conducted for validating the effectiveness of the proposed *** result shows that the soft image segmentation model is able to represent the reliability of the hard segmentation and the existence of mixed pixels on the borders of the adjacent *** uncertainty information of segmentation may help to further understand the segmented result.

关键词： soft segmentation hard segmentation object based image analysis

来源：评论

学校读者我要写书评

暂无评论

MONITORING LAND USE CHANGES ASSOCIATED WITH URBANIZATION IN THE UPPER BHIMA BASIN, MAHARASHTRA, INDIA

MONITORING LAND USE CHANGES ASSOCIATED WITH URBANIZATION IN ...

引用

IEEE International Geoscience and Remote Sensing Symposium

作者： Dipak R. Samal Shirish S. Gedam Centre of Studies in Resources Engineering (CSRE)

ISBN: (纸本)9781479911127

Urbanization is the most dramatic form of irreversible land transformation which occurs due to various anthropogenic activities. Multi-temporal Landsat images acquired from different sensors like MSS, TM and ETM+ between 1972-2009 as well as digital elevation dataset is used to study the influence of urbanization on its surrounding land use/cover pattern in the upper Bhima basin. An object based standard nearest neighbor (NN) classification technique is employed for precise classification of various land use classes in the basin. The study revealed that rapid built up area expansion driven by population growth, demand for water from city region leads to construction of new dams in high rainfall region and mass conversion of waste land to agricultural land are the major changes being noticed in the basin due to urbanization.

关键词： Urbanization Bhima basin object based image analysis GIS

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：