检索结果-内蒙古大学图书馆

Optimized Segmentation based on the Weighted Aggregation Method for Loess Bank Gully Mapping

REMOTE SENSING 2020年第5期12卷 793-793页

作者： Ding, Hu Liu, Kai Chen, Xiaozheng Xiong, Liyang Tang, Guoan Qiu, Fang Strobl, Josef Nanjing Normal Univ Key Lab Virtual Geog Environm Minist Educ Nanjing 210023 Peoples R China South China Normal Univ Sch Geog Guangzhou 510631 Peoples R China Chinese Acad Sci Nanjing Inst Geog & Limnol Key Lab Watershed Geog Sci Nanjing 210008 Peoples R China Nanjing Normal Univ Sch Geog Nanjing 210023 Peoples R China Jiangsu Ctr Collaborat Innovat Geog Informat Reso Nanjing 210023 Peoples R China Univ Texas Dallas Dept Geospatial Sci Richardson TX 75080 USA Univ Salzburg Dept Geoinformat Z GIS A-5020 Salzburg Austria

The Chinese Loess Plateau suffers severe gully erosion. Gully mapping is a fundamental task for gully erosion monitoring in this region. Among the different gully types in the Loess Plateau, the bank gully is usually regarded as the most important source for the generation of sediment. However, approaches for bank gully extraction are still limited. This study put forward an integrated framework, including segmentation optimization, evaluation and Extreme Gradient Boosting (XGBoost)-based classification, for the bank gully mapping of Zhifanggou catchment in the Chinese Loess Plateau. The approach was conducted using a 1-m resolution digital elevation model (DEM), based on unmanned aerial vehicle (UAV) photogrammetry and WorldView-3 imagery. The methodology first divided the study area into different watersheds. Then, segmentation by weighted aggregation (SWA) was implemented to generate multi-level segments. For achieving an optimum segmentation, area-weighted variance (WV) and Moran's I (MI) were adopted and calculated within each sub-watershed. After that, a new discrepancy metric, the area-number index (ANI), was developed for evaluating the segmentation results, and the results were compared with the multi-resolution segmentation (MRS) algorithm. Finally, bank gully mappings were obtained based on the XGBoost model after fine-tuning. The experiment results demonstrate that the proposed method can achieve superior segmentation compared to MRS. Moreover, the overall accuracy of the bank gully extraction results was 78.57%. The proposed approach provides a credible tool for mapping bank gullies, which could be useful for the catchment-scale gully erosion process.

关键词： object-based image analysis gully mapping segmentation optimization unmanned aerial vehicle (UAV) XGBoost

来源：评论

学校读者我要写书评

暂无评论

Mapping Vegetation at Species Level with High-Resolution Multispectral and Lidar Data Over a Large Spatial Area: A Case Study with Kudzu

引用

REMOTE SENSING 2020年第4期12卷 609-609页

作者： Liang, Wanwan Abidi, Mongi Carrasco, Luis McNelis, Jack Tran, Liem Li, Yingkui Grant, Jerome Univ Tennessee Dept Entomol & Plant Pathol Knoxville TN 37996 USA North Carolina State Univ Ctr Geospatial Analyt Raleigh NC 27695 USA Univ Tennessee Dept Elect Engn & Comp Sci Knoxville TN 37996 USA Natl Inst Math & Biol Synth Knoxville TN 37996 USA Univ Tennessee Dept Ecol & Evolutionary Biol Knoxville TN 37996 USA Oak Ridge Natl Lab Oak Ridge TN 37831 USA Univ Tennessee Dept Geog Knoxville TN 37996 USA

Mapping vegetation species is critical to facilitate related quantitative assessment, and mapping invasive plants is important to enhance monitoring and management activities. Integrating high-resolution multispectral remote-sensing (RS) images and lidar (light detection and ranging) point clouds can provide robust features for vegetation mapping. However, using multiple sources of high-resolution RS data for vegetation mapping on a large spatial scale can be both computationally and sampling intensive. Here, we designed a two-step classification workflow to potentially decrease computational cost and sampling effort and to increase classification accuracy by integrating multispectral and lidar data in order to derive spectral, textural, and structural features for mapping target vegetation species. We used this workflow to classify kudzu, an aggressive invasive vine, in the entire Knox County (1362 km(2)) of Tennessee (U.S.). object-based image analysis was conducted in the workflow. The first-step classification used 320 kudzu samples and extensive, coarsely labeled samples (based on national land cover) to generate an overprediction map of kudzu using random forest (RF). For the second step, 350 samples were randomly extracted from the overpredicted kudzu and labeled manually for the final prediction using RF and support vector machine (SVM). Computationally intensive features were only used for the second-step classification. SVM had constantly better accuracy than RF, and the producer's accuracy, user's accuracy, and Kappa for the SVM model on kudzu were 0.94, 0.96, and 0.90, respectively. SVM predicted 1010 kudzu patches covering 1.29 km(2) in Knox County. We found the sample size of kudzu used for algorithm training impacted the accuracy and number of kudzu predicted. The proposed workflow could also improve sampling efficiency and specificity. Our workflow had much higher accuracy than the traditional method conducted in this research, and could be easily im

关键词： detailed vegetation mapping kudzu mapping coarse label two-step classification object-based image analysis lidar point clouds sampling specificity

来源：评论

学校读者我要写书评

暂无评论

Assessing Google Earth Pro images for Detailed Conservation Diagnostics of Mangrove Communities

引用

JOURNAL OF COASTAL RESEARCH 2019年第Sup92期35卷 33-43页

作者： Garcia Calva, Liliana Golubov, Jordan del Carmen Mandujano, Maria Laura Lara-Dominguez, Ana Lopez-Portillo, Jorge Univ Autonoma Metropolitana Programa Doctorado Ciencias Biol & Salud Mexico City DF Mexico Univ Autonoma Metropolitana Lab Ecol & Taxon Vegetal Mexico City DF Mexico Univ Autonoma Metropolitana Inst Ecol Lab Genet & Ecol Mexico City DF Mexico Red Ecol Func Inst Ecol AC Xalapa Veracruz Mexico

Conservation of mangrove ecosystems is one of the major global environmental challenges of this century. To propose viable strategies for its conservation, it is necessary to produce an adequate cartography that enables, among other things, accurate differentiation among plant communities. Google Earth Pro (GEP) images were evaluated as a tool to classify different mangrove communities and were compared to RapidEye satellite (RE) images. The accuracy of the following classifications were compared: (1) one based on the normalized difference vegetation index (NDVI) of a RE image, (2) one based on a mosaic of 141 GEP images, both using object-based image analysis (OBIA) and (3) one based on the visual interpretation (VI) of the same GEP mosaic. It was possible to distinguish ten mangrove communities in 2436 ha, six of them dominated or codominated by the white mangrove Laguncularia racemosa, suggesting that this estuary has a strong riverine influence and low salinities. There was a 97% accuracy on the VI of GEP mosaic, significantly higher than the OBIA classifications done on RE (88%) or GEP (80%) images. These results indicate that GEP, an open access resource, provides similar accuracy when mapping mangrove communities compared to commercial images such as RE, and that the accuracy is greater when using VI. GEP images are very important tools to generate accurate and detailed large scale maps of mangrove communities and represent a great opportunity to move along in the challenge of identification, mapping and monitoring these valuable ecosystems.

关键词： image classification large scale mapping mangrove communities NDVI object-based image analysis RapidEye satellite image visual interpretation

来源：评论

学校读者我要写书评

暂无评论

object-based Mapping of Gullies Using Optical images: A Case Study in the Black Soil Region, Northeast of China

引用

REMOTE SENSING 2020年第3期12卷 487-487页

作者： Wang, Biwei Zhang, Zengxiang Wang, Xiao Zhao, Xiaoli Yi, Ling Hu, Shunguang Chinese Acad Sci Aerosp Informat Res Inst Beijing 100101 Peoples R China Univ Chinese Acad Sci Beijing 100049 Peoples R China

Gully erosion is a widespread natural hazard. Gully mapping is critical to erosion monitoring and the control of degraded areas. The analysis of high-resolution remote sensing images (HRI) and terrain data mixed with developed object-based methods and field verification has been certified as a good solution for automatic gully mapping. Considering the availability of data, we used only open-source optical images (Google Earth images) to identify gully erosion through image feature modeling based on OBIA (object-based image analysis) in this paper. A two-end extrusion method using the optimal machine learning algorithm (Light Gradient Boosting Machine (LightGBM)) and eCognition software was applied for the automatic extraction of gullies at a regional scale in the black soil region of Northeast China. Due to the characteristics of optical images and the design of the method, unmanaged gullies and gullies harnessed in non-forest areas were the objects of extraction. Moderate success was achieved in the absence of terrain data. According to independent validation, the true overestimation ranged from 20% to 30% and was mainly caused by land use types with high erosion risks, such as bare land and farm lanes being falsely classified as gullies. An underestimation of less than 40% was adjacent to the correctly extracted gullied areas. The results of extraction in regions with geographical object categories of a low complexity were usually more satisfactory. The overall performance demonstrates that the present method is feasible for gully mapping at a regional scale, with high automation, low cost, and acceptable accuracy.

关键词： gully google earth images object-based image analysis machine learning methods the black soil region of northeast China

来源：评论

学校读者我要写书评

暂无评论

Seagrass cover expansion off Caja de Muertos Island, Puerto Rico, as determined by long-term analysis of historical aerial and satellite images (1950-2014)

引用

ECOLOGICAL INDICATORS 2020年 117卷 106561-106561页

作者： Leon-Perez, Mariana C. Armstrong, Roy A. Hernandez, William J. Aguilar-Perera, Alfonso Thompson-Grim, Jill Univ Puerto Rico Mayaguez Dept Marine Sci Bioopt Oceanog Lab Lajas PR USA Texas A&M Univ Harte Res Inst Gulf Mexico Studies Corpus Christi TX 78412 USA CUNY City Coll NOAA CESSRST Ctr New York NY USA Univ Autonoma Yucatan Dept Biol Marina Merida Yucatan Mexico

Seagrass ecosystems affected by climate change and anthropogenic disturbances require baseline characteriza-tion of their cover, distribution, and dynamics for effective conservation strategies. Caja de Muertos Island (CMI) is an important nature reserve in Puerto Rico where long-term trends in seagrass cover are unknown but where availability of historical aerial photographs can provide elements for analyzing seagrass dynamics. The purpose of this study was to understand those dynamics in CMI, thereby providing a baseline for determining the severity of future disturbances and inform conservation strategies. We quantified changes in seagrass spatial extent using object-based image analysis in a 2014 WorldView-2 image and historical aerial photographs from 1950 to 2010. Contrasting with worldwide seagrass declining trends, seagrass extent in CMI showed a positive rate of ex-pansion and increased 64%, which was mainly explained by the patchy seagrass cover category that was also the most persistent cover throughout the 64-year period analyzed. Findings suggest an increase in seagrass extent due to recolonization of seagrasses after impacts of tropical storm Baker in 1950, aided by other factors such as reduction in grazing pressure. Even though seagrass cover expanded, with other tropical storms and hurricanes, no negative effects were observed except between 1993 and 1999. The amount and frequency of changes in seagrass cover categories varied between zones and seemed to be linked to localized environmental conditions (e.g. wave exposure) and geomorphological characteristics. The present study represents a baseline of seagrass cover distribution and dynamics, offering valuable information for conservation strategies in Caja de Muertos Island Nature Reserve. In the absence of in situ long-term monitoring data, this study also shows how aerial photography and an object-based image analysis are effective for assessing trends and quantifying seagrass cover changes in cl

关键词： Seagrass dynamics Aerial photography WorldView-2 object-based image analysis Hurricanes Caribbean

来源：评论

学校读者我要写书评

暂无评论

Mapping Tea Plantations from VHR images Using OBIA and Convolutional Neural Networks

引用

REMOTE SENSING 2020年第18期12卷 2935-2935页

作者： Tang, Zixia Li, Mengmeng Wang, Xiaoqin Fuzhou Univ Key Lab Spatial Data Min & Informat Sharing Minist EducAcad Digital China Fujian Natl & Local Joint Engn Res Ctr Satellite Geospat Fuzhou 350108 Peoples R China

Tea is an important economic plant, which is widely cultivated in many countries, particularly in China. Accurately mapping tea plantations is crucial in the operations, management, and supervision of the growth and development of the tea industry. We propose an object-based convolutional neural network (CNN) to extract tea plantations from very high resolution remote sensing images. image segmentation was performed to obtain image objects, while a fine-tuned CNN model was used to extract deep image features. We conducted feature selection based on the Gini index to reduce the dimensionality of deep features, and the selected features were then used for classifying tea objects via a random forest. The proposed method was first applied to Google Earth images and then transferred to GF-2 satellite images. We compared the proposed classification with existing methods: object-based classification using random forest, Mask R-CNN, and object-based CNN without fine-tuning. The results show the proposed method achieved a higher classification accuracy than other methods and produced smaller over- and under-classification geometric errors than Mask R-CNN in terms of shape integrity and boundary consistency. The proposed approach, trained using Google Earth images, achieved comparable results when transferring to the classification of tea objects from GF-2 images. We conclude that the proposed method is effective for mapping tea plantations using very high-resolution remote sensing images even with limited training samples and has huge potential for mapping tea plantations in large areas.

关键词： tea plantation mapping object-based image analysis convolutional neural network very high resolution remote sensing

来源：评论

学校读者我要写书评

暂无评论

An object-based Supervised Nearest Neighbor Method for Extraction of Rhizophora in Mangrove Forest from LiDAR Data and Orthophoto 9

An Object-Based Supervised Nearest Neighbor Method for Extra...

引用

9th IEEE International Conference on System Engineering and Technology (ICSET)

作者： Sarraga Alon, Alvin Festijo, Enrique D. Juanico, Drandreb Earl O. Technol Inst Philippines Grad Sch Program Quezon City Philippines Technol Inst Philippines Res & Dev Management Off Manila Philippines

ISBN: (纸本)9781728107585

Mangroves are a tropical and sub-tropical dominant ecosystem that flourishes in the coastal areas of the Philippines and offer both ecological and economic benefits for coastal communities. However, even with its benefits, its forest cover has experienced substantial damages and mangrove species are critically approaching extinctions. With this implications, mangrove forest cover mapping is an important aid to environmental policies and processes that will contribute to its resources protection and preservation. This study aimed to produce a rhizophora-specific map in a mangrove forest of Magnesia, Virac, Catanduanes, Philippines using Orthophoto and the derivatives from LiDAR Data. LiDAR derivatives such as DSM, DTM, normalized DSM, and Slope using Geographic Information Systems (GIS) were used to analyze the features for rhizophora extraction. An object-based classification scheme was developed using eCognition to extract rhizophora mangroves. The scheme includes Geographic Coordinates Systems (GCS) transformation, Georeferencing, Multiresolution Segmentation, and Classification using Nearest-Neighbor algorithm. Accuracy of the produced map for Magnesia is 99.63% showing two classes namely Rhizophora and Non-Rhizophora.

关键词： object-based image analysis mangrove nearest neighbor LiDAR orthophoto

来源：评论

学校读者我要写书评

暂无评论

Application of Off-Nadir Satellite imagery in Earthquake Damage Assessment Using object-based HOG Feature Descriptor

Application of Off-Nadir Satellite Imagery in Earthquake Dam...

引用

Conference on Geo-information for Disaster Management (Gi4DM)

作者： Jabari, S. Krafczek, M. Dept Geodesy & Geomat Engn Fac Engn Fredericton NB Canada

One of the most crutial applications of very-high-resolution (VHR) satellite images is disaster management. In disaster management, time is of great importance. Therefore, it is vital to acquire satellite images as quickly as possible and benefit from automatic change detection to speed up the process. Automatic damage map generation is performed by overlaying the co-registered before and after images of the area of interest and, compring them to highlight the affected infrastructures. For speeding up image capture, satellites tilt their imaging sensor and take images from oblique angles. However, this kind of image acquisition causes severe geometric distortion in the images, which hinders image co-registration in automatic change detection. In this study, a Patch-Wise Co-Registration (PWCR) solution is used. In this algorithm, the before and after images are co-registered in a segment-by-segment manner. From the literature, this algorithm is followed by a spectral comparison to detect changes. However, due to the complicated structure of debris in damage detection applications, spectral comparison methods cannot perform well. In this work, we developed an object-based method using Histogram of Oriented Gradient descriptor to detect damges and compared our results to different existing spectral and textural change detection methods. The algorithm is tested on images from the 2010-Heidi earthquake, captured by DigitalGlobe. The achieved highly accurate results demonstrate the potential of using off-nadir remote sensing images for automatic urban damage detection possibly in early response systems as it speeds up the damage map generation by providing flexibility to utilize images taken from different anlges.

关键词： off-nadir satellite imagery damage detection object-based image analysis HOG feature descriptor

来源：评论

学校读者我要写书评

暂无评论

Monitoring mangrove association changes in the Can Gio biosphere reserve and implications for management

引用

REMOTE SENSING APPLICATIONS-SOCIETY AND ENVIRONMENT 2019年 13卷 298-305页

作者： Pham, Lien T. H. Vo, Tuan Q. Thanh Duc Dang Nguyen, Uyen T. N. Univ Waikato Dept Geog Tourism & Environm Planning Private Bag 3105 Hamilton New Zealand Vietnam Natl Univ Ho Chi Minh City Univ Sci Fac Environm Sci 227 Nguyen Cu StrDist 5 Hcmc Vietnam Can Tho Univ Dept Land Resources 3-2 St Can Tho Vietnam Singapore Univ Technol & Design Engn Syst & Design Pillar 8 Somapah Rd Singapore Singapore Thuy Loi Univ Inst Water & Environm Res 2 Truong Sa Str Hcmc Vietnam Univ Waikato Sch Sci & Engn Private Bag 3105 Hamilton New Zealand

Mangrove forests are useful in protecting coastal lines, reducing atmospheric carbon dioxide (CO2), and providing shelters and foods for ecosystems, but they are among the top threatened habitats worldwide caused by human activities, climate change, and coastal-related natural disasters such as tsunami, flood, storm, etc. Understanding quantitative and spatial information of mangrove associations is essential for managing mangrove forests effectively. The main aim of this research is to investigate the changes of mangrove forest associations in tropical regions and possible reasons of changes via the case study of the Can Gio biosphere reserve in Viet Nam by using remote sensing datasets (SPOT 4 and 5 imagery), object-based image analysis and Support Vector Machine classifier. The classification results showed that the highest overall accuracy of this combination was 81.6%, given 420 reference objects, confirming the applicability of using remote sensing on assessing the development of mangrove forests. For the case study, we found that there was a decrease of the Avicennia alba - Sonneratia alba (Association I) area by 20.1% caused by the development of aquaculture and other human activities, but there was an increase of the Rhizophora apiculata (Association II) area by 34.8%. These trends may result in the exacerbated risk of soil erosion and make the region become more vulnerable to climate change and tropical storm events. Management policies should be revised to harmonise the benefits of economic development, environmental protection, and species diversity.

关键词： Mangrove associations Support Vector Machine SPOT imagery Can Gio biosphere reserve Human activities object-based image analysis

来源：评论

学校读者我要写书评

暂无评论

Monitoring Artificial Surface Expansion in Ecological Redline Zones by Multi-temporal VHR images 10

Monitoring Artificial Surface Expansion in Ecological Redlin...

引用

10th International Workshop on the analysis of Multitemporal Remote Sensing images (MultiTemp)

作者： Lin, Cong Zhang, Peng Bai, Xuyu Wang, Xin Du, Peijun Nanjing Univ Sch Geog & Ocean Sci Nanjing Peoples R China

ISBN: (纸本)9781728146157

Multi-temporal remote sensing dataset allows for cost- and time-efficient monitoring of land-cover vital to the conservation of ecosystems and biodiversity. The Very High Resolution (VHR) images provide the needed spatial resolution which is important for government management and decision making. In this paper, we propose a workflow for detecting artificial surface expansion in ecological redline zones based on object-based image analysis and transfer matrix. The new increased artificial surface regions are further obtained by rule optimization in the preliminary results. Finally, 854 new artificial surface patches with a total area of more than 1.93 square kilometers were detected. The proposed approach quantifies the effects of human activities without high cost and time-consuming process, which is of great significance for the protection and management of ecological redline zones.

关键词： ecological redline zone disturbance detecting object-based image analysis very high-resolution image

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：