检索结果-内蒙古大学图书馆

Recent Advances in Deep-Learning-Based SAR image Target Detection and recognition

IEEE JOURNAL OF SELECTED TOPICS IN APPLIED EARTH OBSERVATIONS AND remote sensing 2025年 18卷 6884-6915页

作者： Lang, Ping Fu, Xiongjun Dong, Jian Yang, Huizhang Yin, Junjun Yang, Jian Martorella, Marco Tsinghua Univ Dept Elect Engn Beijing 100084 Peoples R China Beijing Inst Technol Sch Integrated Circuits & Elect Beijing 100081 Peoples R China Nanjing Univ Sci & Technol Sch Elect & Opt Engn Nanjing 210094 Peoples R China Univ Sci & Technol Beijing Sch Comp & Commun Engn Beijing 100083 Peoples R China Univ Pisa Dept Informat Engn I-56127 Pisa Italy

Synthetic aperture radar (SAR) image target detection and recognition (SAR-TDR) tasks have become research hot spots in the remote sensing application. These targets include ships, vehicles, aircraft, oil tanks, bridges, and so on. However, with the rapid development of SAR technology and increasingly complex electromagnetic environment, complex characteristics of SAR images bring severe challenges to the accurate SAR-TDR via traditional physical models or manual feature-extraction-based machine learning methods. In recent years, deep learning (DL), as a powerful automatic feature extraction algorithm, has been widely used in the computer vision domain. More specifically, DL has also been introduced into the SAR-TDR tasks and has effectively achieved good performance in terms of accuracy, real-time processing, etc. With the rapid development of DL, SAR image processing, and practical requirements of SAR-TDR in civilian and military domains, it is crucial to conduct a systematic survey on SAR-TDR in the past few years. In this survey article, we mainly conduct a systematic overview of DL-based SAR-TDR literature on two tasks, i.e., target recognition (e.g., ground vehicles, ships, and aircraft) and target detection (e.g., ships, aircraft, change detection, sea surface oil spills, and oil tanks). More specifically, our related works about these topics are also presented to verify the effectiveness of DL-based methods. First, several DL methods (e.g., convolutional neural networks, recurrent neural networks, autoencoders, and generative adversarial networks), commonly used in SAR-TDR, are briefly introduced. Then, a systematic review of DL-based SAR-TDR (including our related works) is presented. Finally, the current challenges and future possible research directions are deeply analyzed and discussed.

关键词： Radar polarimetry Target recognition Synthetic aperture radar Marine vehicles Feature extraction Object detection Aircraft Surveys Accuracy Deep learning Automatic target recognition change detection (CD) deep learning (DL) SAR image interpretation target detection

来源：评论

学校读者我要写书评

暂无评论

Aircraft Target Detection in remote sensing images Based on Improved YOLOv7-Tiny Network

引用

IEEE ACCESS 2025年 13卷 48904-48922页

作者： Zhu, Ruizhe Jin, Hai Han, Yonghua He, Qiyang Mu, Haibo Zhejiang Sci Tech Univ Sch Informat Sci & Engn Hangzhou 310018 Peoples R China Hangzhou Hikvis Digital Technol Co Ltd Hangzhou 310018 Peoples R China

To address the challenges posed by complex backgrounds and target scale variations in remote sensing aircraft detection, we propose an improved YOLOv7-tiny network designed to boost both detection accuracy and the model's ability to generalize across varying conditions. Complex backgrounds hinder effective target feature extraction, while variations in target scale complicate the real-time detection of objects at different sizes. To address these issues, we introduce a multi-head dynamic joint self-attention mechanism that combines self-attention, channel attention, and spatial attention, thereby enhancing the model's ability to capture crucial target features. Furthermore, we design an enhanced feature extraction module that expands the receptive field and improves the model's ability to extract features from multi-scale targets. In addition, we constructed diverse remote sensing aircraft detection datasets and applied a range of data augmentation techniques to improve the robustness and adaptability of the model in real-world scenarios. The experimental results demonstrate that the enhanced YOLOv7-tiny model achieves a mAP of 93.41%, which is a 2.98% improvement over the original model. The model also achieves precision and recall rates of 90.44% and 87.81%, respectively, with a FPS rate of 23.31. These results highlight the model's excellence in both precision and real-time performance, making it highly effective for remote sensing aircraft detection and fine-grained feature recognition, particularly in complex environments where real-time processing is crucial.

关键词： Aircraft Feature extraction Atmospheric modeling remote sensing Accuracy Adaptation models Real-time systems Attention mechanisms Robustness YOLO Aircraft detection multi-scale target detection remote sensing images YOLOv7-tiny

来源：评论

学校读者我要写书评

暂无评论

Overview of remote sensing image fusion based on deep learning 12

Overview of remote sensing image fusion based on deep learni...

引用

SPIE 12th International Symposium on Multispectral image processing and pattern recognition, MIPPR 2023

作者： Wang, Qian Southwest Forestry University 300 Bailong Temple Kunming650024 China

ISBN: (数字)9781510674929

ISBN: (纸本)9781510674912

With the widespread application and rapid development of remote sensing technology, the quality requirements for remote sensing images are gradually improving. Currently, relying solely on one sensor is difficult to ensure high signal-to- noise ratio while capturing images with high spatial and spectral resolution. remote sensing image fusion technology obtains high-quality images by combining the spatial information and spectral (or spectral) information of different sensors. Meanwhile, in recent years, deep learning theory has developed rapidly and is widely applied in image processing such as remote sensing image fusion. Therefore, in order to gain a more systematic understanding of the current status of remote sensing image fusion based on deep learning and promote the development of remote sensing image fusion, this article first introduces commonly used remote sensing satellite images and traditional image fusion algorithms;Then, the remote sensing image fusion algorithm based on deep learning is emphasized, and the advantages and disadvantages of deep learning fusion methods (PanNet, LPPN, WSDFNet) are compared and analyzed;Finally, the future prospects of remote sensing image fusion methods based on deep learning are presented. © COPYRIGHT SPIE. Downloading of the abstract is permitted for personal use only.

关键词： image fusion

来源：评论

学校读者我要写书评

暂无评论

FDA-FFNet: A Feature-Distance Attention-Based Change Detection Network for remote sensing image

引用

IEEE JOURNAL OF SELECTED TOPICS IN APPLIED EARTH OBSERVATIONS AND remote sensing 2024年 17卷 2224-2233页

作者： Peng, Wenguang Shi, Wenzhong Zhang, Min Wang, Lukang Wuhan Univ Sch Remote Sensing & Informat Engn Wuhan 430079 Peoples R China Hong Kong Polytech Univ Smart Cities Res Inst Hong Kong Peoples R China Hong Kong Polytech Univ Dept Land Surveying & Geoinformat Hong Kong Peoples R China China Univ Min & Technol Sch Environm & Spatial Informat Xuzhou 221116 Peoples R China

Convolutional neural networks have demonstrated remarkable capability in extracting deep semantic features from images, leading to significant advancements in various image processing tasks. This success has also opened up new possibilities for change detection (CD) in remote sensing applications. But unlike the conventional image recognition tasks, the performance of AI models in CD heavily relies on the method used to fuse the features from two different phases of the image. The existing deep-learning-based methods for CD typically fuse features of bitemporal images using difference or concatenation techniques. However, these approaches often fail tails to prioritize potential change areas adequately and neglect the rich contextual information essential for discerning subtle changes, potentially leading to slower convergence speed and reduced accuracy. To tackle this challenge, we propose a novel feature fusion approach called feature-difference attention-based feature fusion CD network. This method aims to enhance feature fusion by incorporating a feature-difference attention-based feature fusion module, enabling a more focused analysis of change areas. Additionally, a deep-supervised attention module is implemented to leverage the deep surveillance module for cascading refinement of change areas. Furthermore, an atrous spatial pyramid pooling fast is employed to efficiently acquire multiscale object information. The proposed method is evaluated on two publicly available datasets, namely the WHU-CD and LEVIR-CD datasets. Compared with the state-of-the-art CD methods, the proposed method outperforms in all metrics, with an intersection over union of 92.49% and 85.56%, respectively.

关键词： Feature extraction image segmentation remote sensing Euclidean distance Surveillance Support vector machines Standards Attention-based change detection (CD) deep learning deep supervision multiscale feature

来源：评论

学校读者我要写书评

暂无评论

remote sensing image super-resolution reconstruction using depth-dense residual and sub-pixel convolutional model

Remote sensing image super-resolution reconstruction using d...

引用

2024 International Conference on image, Signal processing, and pattern recognition, ISPP 2024

作者： Liu, Haoguang Fu, Yuwen Wang, Zhoujie Chinese Flight Test Establishment Xian China

ISBN: (纸本)9781510680425

Existing super-resolution reconstruction algorithms for remote sensing images often struggle to fully extract and utilize features in complex scenes, and the reconstruction results are not optimal due to the influence of *** propose a reconstruction network model that combines deep dense residual module and sub-pixel *** model connects multiple dense residual modules through recursive linking and introduces a channel attention mechanism to extract multiscale features from the *** the feature reconstruction process, a sub-pixel convolution structure is introduced to reduce the impact of noise and enhance reconstruction *** model is tested on the UC Merced Land Use public dataset, and the results demonstrate that the reconstruction results of our proposed model algorithm have better peak signal-to-noise ratio (PSNR) and structural similarity (SSIM) at different scales compared to the current mainstream reconstruction algorithm *** is proven that the proposed model can significantly improve the reconstruction quality through comparative analysis experiment, meeting the needs for high-resolution remote sensing image processing. © 2024 SPIE.

关键词： image reconstruction

来源：评论

学校读者我要写书评

暂无评论

RCSFN: A remote sensing image scene classification and recognition network based on rectangle convolutional self attention fusion

引用

SIGNAL image AND VIDEO processing 2024年第12期18卷 8739-8756页

作者： Hou, Jingjin Zhou, Houkui Yu, Huimin Hu, Haoji Zhejiang A&F Univ Sch Math & Comp Sci Hangzhou 311300 Peoples R China Zhejiang Prov Key Lab Forestry Intelligent Monitor Hangzhou 311300 Peoples R China Zhejiang Univ Coll Informat Sci & Elect Engn Hangzhou 310027 Peoples R China State Key Lab CAD & CG Hangzhou 310027 Peoples R China

remote sensing scene classification is a critical task in the processing and analysis of remote sensing images. Traditional methods typically use standard convolutional kernels to extract feature information. Although these methods have seen improvements, they still struggle to fully capture unique local details, thus affecting classification accuracy. Each category within remote sensing scenes has its unique local details, such as the rectangular features of buildings in schools or industrial areas, as well as bridges and roads in parks or squares. The most important features are often these rectangular structures and their spatial positions, which standard convolutional kernels find challenging to capture *** address this issue, we propose a remote sensing scene classification method based on a Rectangle Convolution Self-Attention Fusion Network (RCSFN) architecture. In the RCSFN network, the Rectangle Convolution Maximum Fusion (RCMF) module operates in parallel with the first 4 x 4 convolutional layer of VanillaNet-5. The RCMF module uses two different rectangular convolutional kernels to extract different receptive fields, enhancing the extraction of shallow local features through addition and fusion. This process, combined with the concatenation of the original input features, results in richer local detail ***, we introduce an Area Selection (AS) module that focuses on selecting feature information within local regions. The Sequential Polarisation Self-Attention (SPS) mechanism, integrated with the Mini Region Convolution (MRC) module through feature multiplication, enhances important features and improves spatial positional relationships, thereby increasing the accuracy of recognising categories with rectangular or elongated features. Experiments were carried out on AID and NWPU-RESISC45 data sets, and the overall classification accuracy was 96.56% and 92.46%, respectively. This shows that the RCSFN network model proposed in t

关键词： remote sensing Scene classification Local feature fusion Position enhancement Attention mechanism

来源：评论

学校读者我要写书评

暂无评论

remote sensing aircraft small object detection algorithm based on YOLOv5

Remote sensing aircraft small object detection algorithm bas...

引用

2024 International Conference on image, Signal processing, and pattern recognition, ISPP 2024

作者： Qiu, Yijuan Xue, Jiefeng Zhang, Jie Jiang, Ping Zhang, Gang Lei, Tao National Key Laboratory of Optical Field Manipulation Science and Technology Chinese Academy of Sciences Sichuan Chengdu610209 China Institute of Optics and Electronics Chinese Academy of Sciences Sichuan Chengdu610209 China University of Chinese Academy of Sciences Beijing100049 China Southwest University of Science and Technology 621010 China

ISBN: (纸本)9781510680425

With the rapid development of remote sensing technology, remote sensing images play an important role in the agricultural field, geological field, and natural disaster detection. The size of aircraft in complex scenes in remote sensing images is extremely small, and aircraft of different models have similar shapes. Therefore, improving the accuracy of aircraft target recognition is a challenging task. We propose an improved aircraft small target recognition method based on yolov5, which can improve the recognition accuracy of aircraft targets while ensuring the speed of the model. The specific content is as follows: To address the problem of lack of remote sensing aircraft training sets, we use existing public remote sensing images to combine with aircraft model images. Most of the aircraft models only occupy a dozen to twenty pixels in the 1k*1k image, and perform Scale the generated data set;in order to better combine features of different scales and obtain higher-level feature fusion, introduce the BiFPN module with more residual connections and more complete feature fusion;use the SE attention mechanism to learn the weights of different features, extract information that is more important for detection and improve model performance;in view of the small size of the aircraft, a detection method based on Wasserstein distance such as NWD (Normalized Wasserstein Distance) is selected as the loss function. © 2024 SPIE.

关键词： Training aircraft

来源：评论

学校读者我要写书评

暂无评论

AquaPile-YOLO: Pioneering Underwater Pile Foundation Detection with Forward-Looking Sonar image processing

引用

remote sensing 2025年第3期17卷 360-360页

作者： Xu, Zhongwei Wang, Rui Cao, Tianyu Guo, Wenbo Shi, Bo Ge, Qiqi Tongji Univ Dept Informat & Commun Engn Shanghai 201804 Peoples R China China State Shipbldg Corp Haiying Enterprise Grp C Wuxi 214061 Peoples R China Shanghai Jiao Tong Univ Dept Automat Shanghai 200240 Peoples R China

Underwater pile foundation detection is crucial for environmental monitoring and marine engineering. Traditional methods for detecting underwater pile foundations are labor-intensive and inefficient. Deep learning-based image processing has revolutionized detection, enabling identification through sonar imagery analysis. This study proposes an innovative methodology, named the AquaPile-YOLO algorithm, for underwater pile foundation detection. Our approach significantly enhances detection accuracy and robustness by integrating multi-scale feature fusion, improved attention mechanisms, and advanced data augmentation techniques. Trained on 4000 sonar images, the model excels in delineating pile structures and effectively identifying underwater targets. Experimental data show that the model can achieve good target identification results in similar experimental scenarios, with a 96.89% accuracy rate for underwater target recognition.

关键词： AquaPile-YOLO multi-scale feature fusion deep learning sonar image underwater target recognition attention mechanism

来源：评论

学校读者我要写书评

暂无评论

Deep semantic-aware remote sensing image deblurring

引用

SIGNAL processing 2023年第1期211卷

作者： Song, Zhenbo Zhang, Zhenyuan Fang, Feiyi Fan, Zhaoxin Lu, Jianfeng Nanjing Univ Sci & Technol Sch Comp Sci & Engn Nanjing 210094 Peoples R China Renmin Univ China Sch Informat Beijing 100872 Peoples R China

This paper addresses the problem of blind deblurring of single remote sensing (RS) images with deep neural networks. Most existing deep learning-based methods are migrated from natural image deblurring models, disregarding the domain gap to remote sensing images. Besides, the image deblurring problem is typically considered as an independent low-level image pre-processing, taking no account of downstream tasks, such as classification and segmentation. In this paper, we first present a novel decoder with a par-allel fusion stream for fusing multi-scale RS features and expanding the receptive field. Then, to generate high-quality sharp RS images, we propose to calculate the perceptual loss on an RS-pre-trained network instead of computing on VGG19 pre-trained on natural images i.e. imageNet. For bridging the RS im-age deblurring results to the downstream recognition tasks, we further propose a semantic loss, which is calculated on the last-layer feature map of an RS segmentation network. With extensive experiments con-ducted on public RS image datasets, we demonstrate that the proposed method improves results for RS image deblurring and achieves competitive performance both qualitatively and quantitatively. Moreover, downstream recognition experiments validate the superior quality of the recovered images over existing methods.& COPY;2023 Elsevier B.V. All rights reserved.

关键词： remote sensing image deblurring Deep learning Semantic supervision

来源：评论

学校读者我要写书评

暂无评论

Interchannel Antenna pattern Correction for Azimuth Digital Beamforming in Airborne SAR

引用

IEEE GEOSCIENCE AND remote sensing LETTERS 2025年 22卷

作者： Castillo, Juan Pablo Navarro Scheiber, Rolf Jaeger, Marc Moreira, Alberto German Aerosp Ctr DLR Microwaves & Radar Inst D-82234 Wessling Germany

High-resolution wide swath (HRWS) synthetic aperture radar (SAR) systems are normally designed to have identical antenna patterns in each receive channel. Nevertheless, due to different factors, this condition might not be satisfied, resulting in a multichannel radar system with different antenna patterns. This letter studies the impact of these relative antenna differences on the performance of a state-of-the-art azimuth motion-adaptive image reconstruction for a real airborne SAR sensor with multiple azimuth channels on receive. The performance of the reconstruction algorithm is evaluated both when these relative differences are neglected and when they are accounted for in the multichannel reconstruction process. Additionally, the range dependence of the antenna pattern as a function of wavenumber and the use of range block processing to minimize the impact of this dependence are analyzed. The results confirm the relevance of including these additional steps in the reconstruction, extending the understanding of SAR systems using digital beamforming (DBF) in azimuth.

关键词： Antennas Azimuth Synthetic aperture radar Reconstruction algorithms Receiving antennas image reconstruction Antenna radiation patterns Filters Doppler effect Spaceborne radar Airborne synthetic aperture radar (SAR) antenna pattern azimuth SAR image reconstruction digital beamforming (DBF) range block processing

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

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

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：