检索结果-内蒙古大学图书馆

Semantic Uncertainty-Awared for Semantic Segmentation of remote sensing images

IET image processing 2025年第1期19卷

作者： Qiu, Xiangfeng Zhang, Zhilin Luo, Xin Zhang, Xiang Yang, Youcheng Wu, Yundong Su, Jinhe Jimei Univ Sch Comp Engn Xiamen Peoples R China Xiamen Kingtop Informat Technol Co Ltd Xiamen Peoples R China Ropeok Technol Grp Co Ltd Xiamen Peoples R China

remote sensing image segmentation is crucial for applications ranging from urban planning to environmental monitoring. However, traditional approaches struggle with the unique challenges of aerial imagery, including complex boundary delineation and intricate spatial relationships. To address these limitations, we introduce the semantic uncertainty-aware segmentation (SUAS) method, an innovative plug-and-play solution designed specifically for remote sensing image analysis. SUAS builds upon the rotated multi-scale interaction network (RMSIN) architecture and introduces the prompt refinement and uncertainty adjustment module (PRUAM). This novel component transforms original textual prompts into semantic uncertainty-aware descriptions, particularly focusing on the ambiguous boundaries prevalent in remote sensing imagery. By incorporating semantic uncertainty, SUAS directly tackles the inherent complexities in boundary delineation, enabling more refined segmentations. Experimental results demonstrate SUAS's effectiveness, showing improvements over existing methods across multiple metrics. SUAS achieves consistent enhancements in mean intersection-over-union (mIoU) and precision at various thresholds, with notable performance in handling objects with irregular and complex boundaries-a persistent challenge in aerial imagery analysis. The results indicate that SUAS's plug-and-play design, which leverages semantic uncertainty to guide the segmentation task, contributes to improved boundary delineation accuracy in remote sensing image analysis.

关键词： computer vision image recognition image segmentation remote sensing

来源：评论

学校读者我要写书评

暂无评论

Lightweight remote sensing super-resolution with multi-scale graph attention network

引用

pattern recognition 2025年 160卷

作者： Wang, Yu Shao, Zhenfeng Lu, Tao Huang, Xiao Wang, Jiaming Zhang, Zhizheng Zuo, Xiaolong Wuhan Univ State Key Lab Informat Engn Surveying Mapping & Re Wuhan 430079 Peoples R China Wuhan Inst Technol Sch Comp Sci & Engn Hubei Key Lab Intelligent Robot Wuhan 430073 Peoples R China Emory Univ Dept Environm Sci Atlanta GA 30322 USA

remote sensing Super-Resolution (RS-SR) constitutes a pivotal component in the domain of remote sensing image analysis, aimed at enhancing the spatial resolution of low-resolution imagery. Recent advancements have seen deep learning techniques achieving substantial progress in the RS-SR field. Notably, Graph Neural Networks (GNNs) have emerged as a potent mechanism for processing remote sensing images, adept at elucidating the intricate inter-pixel relationships within images. Nevertheless, a prevalent limitation among existing GNN-based methodologies is their disregard for the high computational demands, which circumscribes their applicability in environments with limited computational resources. This paper introduces a streamlined RS-SR framework, leveraging a Multi-Scale Graph Attention Network (MSGAN), designed to effectively balance computational efficiency with high performance. The core of MSGAN is a novel multi-scale graph attention module, integrating graph attention block and multi-scale lattice block structures, engineered to comprehensively assimilate both localized and extensive spatial information in remote sensing images. This enhances the framework's overall efficacy and resilience in RS-SR tasks. Comparative experimental analyses demonstrate that MSGAN delivers competitive results against state-of-the-art methods while reducing parameter count and computational overhead, presenting a promising avenue for deployment in scenarios with limited computational resources.

关键词： remote sensing Multi-scale network Lightweight network Super-resolution Graph attention network

来源：评论

学校读者我要写书评

暂无评论

BAIDet: remote sensing image object detector based on background and angle information

引用

SIGNAL image AND VIDEO processing 2024年第12期18卷 9295-9304页

作者： Yu, Jiangfeng Sun, Lin Song, Shuhua Guo, Guolong Chen, Kai Shandong Univ Sci & Technol Coll Geodesy & Geomat Qingdao 266590 Peoples R China GEOVIS Wisdom Technol Co Ltd Qingdao 266100 Peoples R China Yangtze Univ Sch Geosci Wuhan 430000 Peoples R China Lingnan Normal Univ Sch Math & Stat Zhanjiang 524048 Peoples R China 91001 Troops Beijing 100309 Peoples R China

remote sensing object detection, with large differences in object size, arbitrary orientation and tight arrangement, leads to difficulties in object recognition and localization. Therefore, a remote sensing image object Detector (BAIDet) based on Background and Angle Information is proposed in this paper. Firstly, a large convolutional kernel global attention module is designed to fully utilize the global information of remote sensing images by expanding the receptive field. And obtain the edge information of ground objects through deformable convolution. Secondly, an angle-sensitive probabilistic intersection-over-union loss function (AS-ProbIoU Loss) is developed for bounding box regression for oriented object detection. Finally, experimental results on four remote sensing image datasets, DOTA, HRSC 2016, UCAS-AOD, and DIOR-R, demonstrated the effectiveness of this method.

关键词： Rotated object detection Deep convolution neural networks Angle loss Spatial attention

来源：评论

学校读者我要写书评

暂无评论

Hyperspectral remote sensing image processing and Information Extraction Technology Research in Geological recognition Application

Hyperspectral Remote Sensing Image Processing and Informatio...

引用

2022 IEEE Asia-Pacific Conference on image processing, Electronics and Computers, IPEC 2022

作者： Bi, Xiaojia Jiang, Liangwen Dong, Qing Wang, Dong Engineering Group. Co. Ltd China Railway Eryuan Chengdu China Academy of Sciences Aerospace Information Research Institute Chinese Chengdu China

ISBN: (纸本)9781665409025

Application of hyperspectral remote sensing technology in geological identification has highlighted its advantages, but the identification accuracy of metal mineral cations is limited. In this paper, the dominant spectrum of domestic HJ-1A data and Hyperion data are used for collaborative processing, realize multi-source hyperspectral image fusion and information extraction and recognition, mapping of mineral alteration information in the Qinghai-Tibet Plateau. The adopted method improves the identification accuracy of metal mineral cations in the visible light and near-infrared range. After comparative verification analysis and statistical evaluation of results, it is confirmed that this method has important application value for the identification of metal mineral cations in complex geological environment. © 2022 IEEE.

关键词： Minerals

来源：评论

学校读者我要写书评

暂无评论

Water area extraction method based on Res2Net-Unet 24

Water area extraction method based on Res2Net-Unet

引用

1st International Conference on image processing Machine Learning and pattern recognition

作者： Liu, Xiaoqun Li, Ning Commun Univ China Nanjing Peoples R China Vimicro Technology Corp Beijing Peoples R China

ISBN: (纸本)9798400707032

Regarding the issues that traditional water area extraction algorithms for remote sensing images require preset thresholds, are vulnerable to interference from mountain or vegetation shadows, and have poor robustness at different spatiotemporal scales, a model based on the Res2Net-UNet is proposed to extract water area from remote sensing images. The model adopts the UNet's encoder-decoder architecture, deepens the model's level with the ResNet network, and performs multi-scale feature extraction and fine-grained feature expression by combining Res2Net. Moreover, the model integrates an attention mechanism to increase the weight of key features and fully utilize the spatial and channel information in the images. After that, the model adds a skip connection to better fuse the encoder and decoder to preserve the detailed features of the image. Experimental findings on the GF-1 dataset demonstrate that the Res2Net-UNet model significantly enhances the accuracy of water extraction in remote sensing images.

关键词： water area extraction remote sensing image Res2Net-Unet attention mechanism

来源：评论

学校读者我要写书评

暂无评论

Hierarchical recognition for Urban Villages Fusing Multiview Feature Information

引用

IEEE JOURNAL OF SELECTED TOPICS IN APPLIED EARTH OBSERVATIONS AND remote sensing 2025年 18卷 3344-3355页

作者： Wang, Zhenkang Xia, Nan Hua, Song Liang, Jiale Ji, Xiankai Wang, Ziyu Wang, Jiechen Nanjing Univ Jiangsu Prov Key Lab Geog Informat Technol Nanjing 210023 Peoples R China Nanjing Univ Sch Geog & Ocean Sci Nanjing 210023 Peoples R China

Urban village (UV) renovation is crucial for urban renewal, with effective UV recognition serving as a prerequisite. While existing studies on UV recognition predominantly rely on high-resolution remote sensing images (RSI), and few integrate street view images (SVI), which could cause confusion in regions with similar planar features, such as old residential area and industrial parks. This article proposed a hierarchical framework for UV recognition which integrated multiview images. The spectral, textural, and structural features were extracted from Google RSI by machine-learning classifiers for each segmented block. The deep-learning method was applied to SVI to capture the architectural feature at each viewpoint. The rule-constrained fusion was conducted to combine the block-level and point-level UV recognition results. Taking a typical high-density megacity Nanjing as the study area, a high recognition overall accuracy (OA) and Kappa of 95.04% and 0.860 were achieved, identifying 172 UVs covering an area of 27.93 km(2) by 2020. The results demonstrated an "urban village ring" pattern in the city, with central urban areas showing a "multicenter and multicluster" spatial distribution, while suburban areas exhibited "large and concentrated" characteristics. Compared with results from single-view of RSI, the complementarity with SVI for multiview features increased the OA and Kappa by 3.34% and 0.079, which could effectively distinguish the old industrial parks. We believe that our proposed hierarchical framework is essential to the scientific and accurate UV recognition, which can guide the urban management and high-quality development.

关键词： Urban areas Feature extraction image segmentation image recognition remote sensing Character recognition Accuracy Internet Earth Deep learning Bottom-up and top-down street view sustainable urban renewal urban village (UV)

来源：评论

学校读者我要写书评

暂无评论

Refined change detection in heterogeneous low-resolution remote sensing images for disaster emergency response

引用

ISPRS JOURNAL OF PHOTOGRAMMETRY AND remote sensing 2025年 220卷 139-155页

作者： Wang, Di Ma, Guorui Zhang, Haiming Wang, Xiao Zhang, Yongxian Wuhan Univ State Key Lab Informat Engn Surveying Mapping & Re Wuhan 430079 Peoples R China Chengdu Univ Sch Architecture & Civil Engn Chengdu 610106 Peoples R China Tsinghua Univ State Key Lab Hydrosci & Engn Beijing 100084 Peoples R China

Heterogeneous remote sensing images Change Detection (HRSICD) is a significant challenge in remote sensing image processing, with substantial application value in rapid natural disaster response. However, significant differences in imaging modalities often result in poor comparability of their features, affecting the recognition accuracy. To address the issue, we propose a novel HRSICD method based on image structure relationships and semantic information. First, we employ a Multi-scale Pyramid Convolution Encoder to efficiently extract the multi-scale and detailed features. Next, the Cross-domain Feature Alignment Module aligns the structural relationships and semantic features of the heterogeneous images, enhancing the comparability between heterogeneous image features. Finally, the Multi-level Decoder fuses the structural and semantic features, achieving refined identification of change areas. We validated the advancement of proposed method on five publicly available HRSICD datasets. Additionally, zero-shot generalization experiments and real-world applications were conducted to assess its generalization capability. Our method achieved favorable results in all experiments, demonstrating its effectiveness. The code of the proposed method will be made available at https://github. com/Lucky-DW/HRSICD.

关键词： Heterogeneous remote sensing images Change detection Refined identification Low-resolution remote sensing images

来源：评论

学校读者我要写书评

暂无评论

Application of motion trajectory recognition based on remote sensing image optical processing in optimizing swimming training schemes

引用

Optical and Quantum Electronics 2024年第2期56卷 264-264页

作者： Jiang, Wei Sports Department of Jiangsu University Jiangsu Zhenjiang 212013 China

The optimization of swimming training scheme is of great significance to improve the training efficiency and competitive performance of swimmers. However, the traditional swimming training programs often lack of personalized and scientific, can not meet the characteristics and needs of different athletes. The aim of this paper is to realize the automatic recognition of motion trajectory through optical processing technology, so as to improve the swimming training scheme. The optical processing technology of remote sensing image is used to collect video images during swimming training, and the optical processing algorithm is used to extract the trajectory information of swimmers in the images. Then image processing and pattern recognition technology are used to identify and analyze the trajectory, so as to obtain accurate trajectory data. Finally, based on the analysis results, the swimming training scheme is optimized and personalized training suggestions are provided. Through the experiment and result analysis, the effectiveness of optical processing in the optimization of swimming training scheme is verified. Through the automatic identification and analysis of the movement trajectory, the swimmer’s skill and ability level can be more accurately understood, thus providing a scientific basis for the formulation and adjustment of training programs. The optical processing can also monitor and record the swimmers' trajectory changes in real time, providing real-time feedback and guidance for coaches to further improve the training effect. © 2023, The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature.

关键词： Motion path recognition Optical processing remote sensing image Swimming Training scheme

来源：评论

学校读者我要写书评

暂无评论

Target recognition in ISAR image via Range Profile Perturbation Imaging

引用

IEEE TRANSACTIONS ON GEOSCIENCE AND remote sensing 2024年 62卷

作者： Dong, Ganggang Liu, Hongwei Xidian Univ Natl Key Lab Radar Signal Proc Xian Peoples R China

The deep learning-based target recognition methods have achieved great success in recent years. The major advantage lies in the ability to learn the high-level representation from the input data adaptively. Yet, they relied on large amounts of training data to cover the complete distribution space of samples. It is infeasible to be met in the practical applications. The fitting ability and the learning power were therefore deteriorated. To solve the problem, a new target augmentation method is proposed in this article. The original complex-valued image is first recast into the phase history. The range profiles are then randomly perturbed, such as the shift in cyclic, corruption in range bins, and the drop of range bins. The perturbed range profiles are used for target imaging, forming the new inverse synthetic aperture radar (ISAR) images. The diversity of the training dataset can be enhanced, and the learning effectiveness of the deep model can be improved accordingly. To verify the proposed method, several rounds of experiments are performed. The results demonstrate the advantages of the proposed method in comparison to state-of-the-art.

关键词： Radar imaging Imaging Radar Target recognition Radar cross-sections Training History Spaceborne radar Geoscience and remote sensing Feature extraction Deep learning inverse synthetic aperture radar (ISAR) imaging target recognition

来源：评论

学校读者我要写书评

暂无评论

Joint Classification of Hyperspectral and LiDAR Data via Multiprobability Decision Fusion Method

引用

remote sensing 2024年第22期16卷 4317-4317页

作者： Chen, Tao Chen, Sizuo Chen, Luying Chen, Huayue Zheng, Bochuan Deng, Wu China West Normal Univ Sch Comp Sci Nanchong 637002 Peoples R China Civil Aviat Univ China Sch Elect Informat & Automat Tianjin 300300 Peoples R China China West Normal Univ Inst Artificial Intelligence Nanchong 637002 Peoples R China China West Normal Univ Key Lab Optimizat Theory & Applicat Nanchong 637002 Peoples R China Southwest Jiaotong Univ State Key Lab Rail Transit Vehicle Syst Chengdu 610031 Peoples R China

With the development of sensor technology, the sources of remotely sensed image data for the same region are becoming increasingly diverse. Unlike single-source remote sensing image data, multisource remote sensing image data can provide complementary information for the same feature, promoting its recognition. The effective utilization of remote sensing image data from various sources can enhance the extraction of image features and improve the accuracy of feature recognition. Hyperspectral remote sensing (HSI) data and light detection and ranging (LiDAR) data can provide complementary information from different perspectives and are frequently combined in feature identification tasks. However, the process of joint use suffers from data redundancy, low classification accuracy and high time complexity. To address the aforementioned issues and improve feature recognition in classification tasks, this paper introduces a multiprobability decision fusion (PRDRMF) method for the combined classification of HSI and LiDAR data. First, the original HSI data and LiDAR data are downscaled via the principal component-relative total variation (PRTV) method to remove redundant information. In the multifeature extraction module, the local texture features and spatial features of the image are extracted to consider the local texture and spatial structure of the image data. This is achieved by utilizing the local binary pattern (LBP) and extended multiattribute profile (EMAP) for the two types of data after dimensionality reduction. The four extracted features are subsequently input into the corresponding kernel-extreme learning machine (KELM), which has a simple structure and good classification performance, to obtain four classification probability matrices (CPMs). Finally, the four CPMs are fused via a multiprobability decision fusion method to obtain the optimal classification results. Comparison experiments on four classical HSI and LiDAR datasets demonstrate that the method pro

关键词： HSI data LiDAR data image classification principal component-relative total variation (PRTV) multiprobability decision fusion method

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：