remotesensing-logo

Journal Browser

Journal Browser

Deep Learning Based Target Detection and Recognition in Remote Sensing Images

A special issue of Remote Sensing (ISSN 2072-4292). This special issue belongs to the section "Remote Sensing Image Processing".

Deadline for manuscript submissions: closed (31 October 2023) | Viewed by 26774

Special Issue Editors

Aerospace Information Research Institute, Chinese Academy of Sciences, Beijing 100094, China
Interests: target detection; target recognition; target tracking; deep learning; remote sensing
1. College of Information Science and Technology, Beijing University of Chemical Technology, Beijing 100029, China
2. Interdisciplinary Research Center for Artificial Intelligence, Beijing University of Chemical Technology, Beijing 100029, China
Interests: image processing; artificial intelligence; remote sensing; high performance computing
Special Issues, Collections and Topics in MDPI journals
School of Remote Sensing and Information Engineering, Wuhan University, Wuhan, China
Interests: satellite image processing; satellite image analysis; remote sensing; image registration; image reconstruction; image restoration; cloud cover; missing data analysis; image mosaic; image fusion; image inpainting; multitemporal analysis
Special Issues, Collections and Topics in MDPI journals
Department of Electrical and Computer Engineering, Mississippi State University, Starkville, MS 39762, USA
Interests: machine learning; remote sensing
Special Issues, Collections and Topics in MDPI journals
German Aerospace Center, 82234 Wessling, Germany
Interests: data mining; image classification; image segmentation; data visualization
School of Automation, Northwestern Polytechnical University, Xi'an 710021, China
Interests: SAR target recognition; transfer learning; unsupervised learning
Special Issues, Collections and Topics in MDPI journals

Special Issue Information

Dear Colleagues,

Target detection and recognition is a fundamental task in remote sensing, and it plays a significant role in various applications. Tradition algorithms use manually designed features whose representation capability is limited. With the further development of deep learning (DL) techniques, DL-based target detection and recognition approaches have become increasingly popular. Despite substantial progress in the field of DL-based detectors and classifiers with automatically learned features, there are several remaining issues: 1) the performance of tiny targets or target detection in low-resolution images is not satisfactory due to limited information; 2) target detection and recognition with few training samples is still a challenge; techniques such as transfer learning, weakly supervised learning, self-supervised learning and meta learning are possible solutions requiring investigation; 3) current target detection and recognition models are more like black boxes; their interpretability needs to be further studied in order to advance their development in remote sensing images.

This Special Issue aims to provide a platform for researchers to discuss and provide solutions for the above-mentioned issues, contributing to the development of target detection and recognition in remote sensing images.

Topics of interest include, but are not limited to:

  • Deep-learning-based target detection, tracking and recognition in visible remote sensing images, infrared remote sensing images or synthetic aperture radar images.
  • Advanced remote sensing target detection and recognition techniques for addressing issues including few-shot learning, tiny target detection, fine-grained target recognition, etc.
  • Land cover and land use classification, change detection of remote sensing images with one sensor or multiple sensors.
  • Investigations on the physical interpretability of target detection and recognition models in remote sensing images.

Dr. Zongxu Pan
Prof. Dr. Fan Zhang
Dr. Xinghua Li
Dr. Bo Tang
Dr. Wei Yao
Dr. Zhongling Huang
Guest Editors

Manuscript Submission Information

Manuscripts should be submitted online at www.mdpi.com by registering and logging in to this website. Once you are registered, click here to go to the submission form. Manuscripts can be submitted until the deadline. All submissions that pass pre-check are peer-reviewed. Accepted papers will be published continuously in the journal (as soon as accepted) and will be listed together on the special issue website. Research articles, review articles as well as short communications are invited. For planned papers, a title and short abstract (about 100 words) can be sent to the Editorial Office for announcement on this website.

Submitted manuscripts should not have been published previously, nor be under consideration for publication elsewhere (except conference proceedings papers). All manuscripts are thoroughly refereed through a single-blind peer-review process. A guide for authors and other relevant information for submission of manuscripts is available on the Instructions for Authors page. Remote Sensing is an international peer-reviewed open access semimonthly journal published by MDPI.

Please visit the Instructions for Authors page before submitting a manuscript. The Article Processing Charge (APC) for publication in this open access journal is 2700 CHF (Swiss Francs). Submitted papers should be well formatted and use good English. Authors may use MDPI's English editing service prior to publication or during author revisions.

Keywords

  • target detection
  • target recognition
  • change detection
  • land and use and land cover classification
  • deep learning
  • remote sensing images

Published Papers (17 papers)

Order results
Result details
Select all
Export citation of selected articles as:

Research

Jump to: Other

21 pages, 7425 KiB  
Article
Unsupervised Domain-Adaptive SAR Ship Detection Based on Cross-Domain Feature Interaction and Data Contribution Balance
by Yanrui Yang, Jie Chen, Long Sun, Zheng Zhou, Zhixiang Huang and Bocai Wu
Remote Sens. 2024, 16(2), 420; https://doi.org/10.3390/rs16020420 - 21 Jan 2024
Viewed by 929
Abstract
Due to the complex imaging mechanism of SAR images and the lack of multi-angle and multi-parameter real scene SAR target data, the generalization performance of existing deep-learning-based synthetic aperture radar (SAR) image target detection methods are extremely limited. In this paper, we propose [...] Read more.
Due to the complex imaging mechanism of SAR images and the lack of multi-angle and multi-parameter real scene SAR target data, the generalization performance of existing deep-learning-based synthetic aperture radar (SAR) image target detection methods are extremely limited. In this paper, we propose an unsupervised domain-adaptive SAR ship detection method based on cross-domain feature interaction and data contribution balance. First, we designed a new cross-domain image generation module called CycleGAN-SCA to narrow the gap between the source domain and the target domain. Second, to alleviate the influence of complex backgrounds on ship detection, a new backbone using a self-attention mechanism to tap the potential of feature representation was designed. Furthermore, aiming at the problems of low resolution, few features and easy information loss of small ships, a new lightweight feature fusion and feature enhancement neck was designed. Finally, to balance the influence of different quality samples on the model, a simple and efficient E12IoU Loss was constructed. Experimental results based on a self-built large-scale optical-SAR cross-domain target detection dataset show that compared with existing cross-domain methods, our method achieved optimal performance, with the mAP reaching 68.54%. Furthermore, our method achieved a 6.27% improvement compared to the baseline, even with only 5% of the target domain labeled data. Full article
Show Figures

Figure 1

20 pages, 32219 KiB  
Article
A Lightweight Arbitrarily Oriented Detector Based on Transformers and Deformable Features for Ship Detection in SAR Images
by Bingji Chen, Fengli Xue and Hongjun Song
Remote Sens. 2024, 16(2), 237; https://doi.org/10.3390/rs16020237 - 07 Jan 2024
Viewed by 840
Abstract
Lightweight ship detection is an important application of synthetic aperture radar (SAR). The prevailing trend in recent research involves employing a detection framework based on convolutional neural networks (CNNs) and horizontal bounding boxes (HBBs). However, CNNs with local receptive fields fall short in [...] Read more.
Lightweight ship detection is an important application of synthetic aperture radar (SAR). The prevailing trend in recent research involves employing a detection framework based on convolutional neural networks (CNNs) and horizontal bounding boxes (HBBs). However, CNNs with local receptive fields fall short in acquiring adequate contextual information and exhibit sensitivity to noise. Moreover, HBBs introduce significant interference from both the background and adjacent ships. To overcome these limitations, this paper proposes a lightweight transformer-based method for detecting arbitrarily oriented ships in SAR images, called LD-Det, which excels at promptly and accurately identifying rotating ship targets. First, light pyramid vision transformer (LightPVT) is introduced as a lightweight backbone network. Built upon PVT v2-B0-Li, it effectively captures the long-range dependencies of ships in SAR images. Subsequently, multi-scale deformable feature pyramid network (MDFPN) is constructed as a neck network, utilizing the multi-scale deformable convolution (MDC) module to adjust receptive field regions and extract ship features from SAR images more effectively. Lastly, shared deformable head (SDHead) is proposed as a head network, enhancing ship feature extraction with the combination of deformable convolution operations and a shared parameter structure design. Experimental evaluations on two publicly available datasets validate the efficacy of the proposed method. Notably, the proposed method achieves state-of-the-art detection performance when compared with other lightweight methods in detecting rotated targets. Full article
Show Figures

Figure 1

18 pages, 5778 KiB  
Article
Semantic-Layout-Guided Image Synthesis for High-Quality Synthetic-Aperature Radar Detection Sample Generation
by Yi Kuang, Fei Ma, Fangfang Li, Yingbing Liu and Fan Zhang
Remote Sens. 2023, 15(24), 5654; https://doi.org/10.3390/rs15245654 - 07 Dec 2023
Viewed by 748
Abstract
With the widespread application and functional complexity of deep neural networks (DNNs), the demand for training samples is increasing. This elevated requirement also extends to DNN-based SAR object detection. Most public SAR object detection datasets are oriented to marine targets such as ships, [...] Read more.
With the widespread application and functional complexity of deep neural networks (DNNs), the demand for training samples is increasing. This elevated requirement also extends to DNN-based SAR object detection. Most public SAR object detection datasets are oriented to marine targets such as ships, while data sets oriented to land targets are relatively rare, though they are an effective way to improve the land object detection capability of deep models through SAR sample generation. In this paper, a synthesis generation collaborative SAR sample augmentation framework is proposed to achieve flexible and diverse high-quality sample augmentation. First, a semantic-layout-guided image synthesis strategy is proposed to generate diverse detection samples. The issues of object location rationality and object layout diversity are also addressed. Meanwhile, a pix2pixGAN network guided by layout maps is utilized to achieve diverse background augmentation. Second, a progressive training strategy of diffusion models is proposed to achieve semantically controllable SAR sample generation to further improve the diversity of scene clutter. Finally, a sample cleaning method considering distribution migration and network filtering is employed to further improve the quality of detection samples. The experimental results show that this semantic synthesis generation method can outperform existing sample augmentation methods, leading to a comprehensive improvement in the accuracy metrics of classical detection networks. Full article
Show Figures

Figure 1

28 pages, 10998 KiB  
Article
Faster and Lightweight: An Improved YOLOv5 Object Detector for Remote Sensing Images
by Jiarui Zhang, Zhihua Chen, Guoxu Yan, Yi Wang and Bo Hu
Remote Sens. 2023, 15(20), 4974; https://doi.org/10.3390/rs15204974 - 15 Oct 2023
Cited by 1 | Viewed by 2274
Abstract
In recent years, the realm of deep learning has witnessed significant advancements, particularly in object detection algorithms. However, the unique challenges posed by remote sensing images, such as complex backgrounds, diverse target sizes, dense target distribution, and overlapping or obscuring targets, demand specialized [...] Read more.
In recent years, the realm of deep learning has witnessed significant advancements, particularly in object detection algorithms. However, the unique challenges posed by remote sensing images, such as complex backgrounds, diverse target sizes, dense target distribution, and overlapping or obscuring targets, demand specialized solutions. Addressing these challenges, we introduce a novel lightweight object detection algorithm based on Yolov5s to enhance detection performance while ensuring rapid processing and broad applicability. Our primary contributions include: firstly, we implemented a new Lightweight Asymmetric Detection Head (LADH-Head), replacing the original detection head in the Yolov5s model. Secondly, we introduce a new C3CA module, incorporating the Coordinate Attention mechanism, strengthening the network’s capability to extract precise location information. Thirdly, we proposed a new backbone network, replacing the C3 module in the Yolov5s backbone with a FasterConv module, enhancing the network’s feature extraction capabilities. Additionally, we introduced a Content-aware Feature Reassembly (content-aware reassembly of features) (CARAFE) module to reassemble semantic similar feature points effectively, enhancing the network’s detection capabilities and reducing the model parameters. Finally, we introduced a novel XIoU loss function, aiming to improve the model’s convergence speed and robustness during training. Experimental results on widely used remote sensing image datasets such as DIOR, DOTA, and SIMD demonstrate the effectiveness of our proposed model. Compared to the original Yolov5s algorithm, we achieved a mean average precision (mAP) increase of 3.3%, 6.7%, and 3.2%, respectively. These findings underscore the superior performance of our proposed model in remote sensing image object detection, offering an efficient, lightweight solution for remote sensing applications. Full article
Show Figures

Figure 1

23 pages, 5171 KiB  
Article
Multi-Stage Multi-Scale Local Feature Fusion for Infrared Small Target Detection
by Yahui Wang, Yan Tian, Jijun Liu and Yiping Xu
Remote Sens. 2023, 15(18), 4506; https://doi.org/10.3390/rs15184506 - 13 Sep 2023
Viewed by 918
Abstract
The detection of small infrared targets with dense distributions and large-scale variations is an extremely challenging problem. This paper proposes a multi-stage, multi-scale local feature fusion method for infrared small target detection to address this problem. The method is based on multi-stage and [...] Read more.
The detection of small infrared targets with dense distributions and large-scale variations is an extremely challenging problem. This paper proposes a multi-stage, multi-scale local feature fusion method for infrared small target detection to address this problem. The method is based on multi-stage and multi-scale local feature fusion. Firstly, considering the significant variation in target sizes, ResNet-18 is utilized to extract image features at different stages. Then, for each stage, multi-scale feature pyramids are employed to obtain corresponding multi-scale local features. Secondly, to enhance the detection rate of densely distributed targets, the multi-stage and multi-scale features are progressively fused and concatenated to form the final fusion results. Finally, the fusion results are fed into the target detector for detection. The experimental results for the SIRST and MDFA demonstrate that the proposed method effectively improves the performance of infrared small target detection. The proposed method achieved mIoU values of 63.43% and 46.29% on two datasets, along with F-measure values of 77.62% and 63.28%, respectively. Full article
Show Figures

Graphical abstract

20 pages, 9544 KiB  
Article
Adaptive Adjacent Layer Feature Fusion for Object Detection in Remote Sensing Images
by Xuesong Zhang, Zhihui Gong, Haitao Guo, Xiangyun Liu, Lei Ding, Kun Zhu and Jiaqi Wang
Remote Sens. 2023, 15(17), 4224; https://doi.org/10.3390/rs15174224 - 28 Aug 2023
Cited by 3 | Viewed by 1078
Abstract
Object detection in remote sensing images faces the challenges of a complex background, large object size variations, and high inter-class similarity. To address these problems, we propose an adaptive adjacent layer feature fusion (AALFF) method, which is developed on the basis of RTMDet. [...] Read more.
Object detection in remote sensing images faces the challenges of a complex background, large object size variations, and high inter-class similarity. To address these problems, we propose an adaptive adjacent layer feature fusion (AALFF) method, which is developed on the basis of RTMDet. Specifically, the AALFF method incorporates an adjacent layer feature fusion enhancement (ALFFE) module, designed to capture high-level semantic information and accurately locate object spatial positions. ALFFE also effectively preserves small objects by fusing adjacent layer features and employs involution to aggregate contextual information in a wide spatial range for object essential features extraction in complex backgrounds. Additionally, the adaptive spatial feature fusion (ASFF) module is introduced to guide the network to select and fuse the crucial features to improve the adaptability to objects with different sizes. The proposed method achieves mean average precision (mAP) values of 77.1%, 88.9%, and 95.7% on the DIOR, HRRSD, and NWPU VHR-10 datasets, respectively. Notably, our approach achieves mAP75 values of 60.8% and 79.0% on the DIOR and HRRSD datasets, respectively, surpassing the state-of-the-art performance on the DIOR dataset. Full article
Show Figures

Figure 1

20 pages, 4502 KiB  
Article
MFSFNet: Multi-Scale Feature Subtraction Fusion Network for Remote Sensing Image Change Detection
by Zhiqi Huang and Hongjian You
Remote Sens. 2023, 15(15), 3740; https://doi.org/10.3390/rs15153740 - 27 Jul 2023
Viewed by 1143
Abstract
Change detection plays a crucial role in remote sensing by identifying surface modifications between two sets of temporal remote sensing images. Recent advancements in deep learning techniques have yielded significant achievements in this field. However, there are still some challenges: (1) Existing change [...] Read more.
Change detection plays a crucial role in remote sensing by identifying surface modifications between two sets of temporal remote sensing images. Recent advancements in deep learning techniques have yielded significant achievements in this field. However, there are still some challenges: (1) Existing change feature fusion methods often introduce redundant information. (2) The complexity of network structures leads to a large number of parameters and difficulties in model training. To overcome these challenges, this paper proposes a Multi-Scale Feature Subtraction Fusion Network (MFSF-Net). It comprises two primary modules: the Multi-scale Feature Subtraction Fusion (MFSF) module and the Feature Deep Supervision (FDS) module. MFSF enhances change features and reduces redundant pseudo-change features. FDS provides additional supervision on different scales of change features in the decoder, improving the training efficiency performance of the network. Additionally, to address the problem of imbalanced samples, the Dice loss strategy is introduced as a means to mitigate this issue. Through comprehensive experiments, MFSF-Net achieves an F1 score of 91.15% and 95.64% on LEVIR-CD and CDD benchmark datasets, respectively, outperforming six state-of-the-art algorithms. Moreover, it attains an improved balance between model complexity and performance, showcasing the efficacy of the proposed approach. Full article
Show Figures

Graphical abstract

21 pages, 30903 KiB  
Article
Multi-Oriented Enhancement Branch and Context-Aware Module for Few-Shot Oriented Object Detection in Remote Sensing Images
by Haozheng Su, Yanan You and Sixu Liu
Remote Sens. 2023, 15(14), 3544; https://doi.org/10.3390/rs15143544 - 14 Jul 2023
Viewed by 873
Abstract
For oriented object detection, the existing CNN-based methods typically rely on a substantial and diverse dataset, which can be expensive to acquire and demonstrate limited capacity for generalization when faced with new categories that lack annotated samples. In this case, we propose MOCA-Net, [...] Read more.
For oriented object detection, the existing CNN-based methods typically rely on a substantial and diverse dataset, which can be expensive to acquire and demonstrate limited capacity for generalization when faced with new categories that lack annotated samples. In this case, we propose MOCA-Net, a few-shot oriented object detection method with a multi-oriented enhancement branch and context-aware module, utilizing a limited number of annotated samples from novel categories for training. Especially, our method generates multi-oriented and multi-scale positive samples and then inputs them into an RPN and the detection head as a multi-oriented enhancement branch for enhancing the classification and regression capabilities of the detector. And by utilizing the context-aware module, the detector can effectively extract contextual information surrounding the object and incorporate it into RoI features in an adaptive manner, thereby improving its classification capability. As far as we know, our method is the first to attempt this in this field, and comparative experiments conducted on the public remote sensing dataset DOTA for oriented object detection showed that our method is effective. Full article
Show Figures

Graphical abstract

24 pages, 2641 KiB  
Article
ESarDet: An Efficient SAR Ship Detection Method Based on Context Information and Large Effective Receptive Field
by Yimin Zhang, Chuxuan Chen, Ronglin Hu and Yongtao Yu
Remote Sens. 2023, 15(12), 3018; https://doi.org/10.3390/rs15123018 - 09 Jun 2023
Cited by 1 | Viewed by 1557
Abstract
Ship detection using synthetic aperture radar (SAR) has been extensively utilized in both the military and civilian fields. On account of complex backgrounds, large scale variations, small-scale targets, and other challenges, it is difficult for current SAR ship detection methods to strike a [...] Read more.
Ship detection using synthetic aperture radar (SAR) has been extensively utilized in both the military and civilian fields. On account of complex backgrounds, large scale variations, small-scale targets, and other challenges, it is difficult for current SAR ship detection methods to strike a balance between detection accuracy and computation efficiency. To overcome those challenges, ESarDet, an efficient SAR ship detection method based on contextual information and a large effective receptive field (ERF), is proposed. We introduce the anchor-free object detection method YOLOX-tiny as a baseline model and make several improvements to it. First, CAA-Net, which has a large ERF, is proposed to better merge the contextual and semantic information of ships in SAR images to improve ship detection, particularly for small-scale ships with complex backgrounds. Further, to prevent the loss of semantic information regarding ship targets in SAR images, we redesign a new spatial pyramid pooling network, namely A2SPPF. Finally, in consideration of the challenge posed by the large variation in ship scale in SAR images, we design a novel convolution block, called A2CSPlayer, to enhance the fusion of feature maps from different scales. Extensive experiments are conducted on three publicly available SAR ship datasets, DSSDD, SSDD, and HRSID, to validate the effectiveness of the proposed ESarDet. The experimental results demonstrate that ESarDet has distinct advantages over current state-of-the-art (SOTA) detectors in terms of detection accuracy, generalization capability, computational complexity, and detection speed. Full article
Show Figures

Figure 1

20 pages, 2940 KiB  
Article
A Class-Incremental Learning Method for SAR Images Based on Self-Sustainment Guidance Representation
by Qidi Pan, Kuo Liao, Xuesi He, Zhichun Bu and Jiyan Huang
Remote Sens. 2023, 15(10), 2631; https://doi.org/10.3390/rs15102631 - 18 May 2023
Viewed by 934
Abstract
Existing deep learning algorithms for synthetic aperture radar (SAR) image recognition are performed with offline data. These methods must use all data to retrain the entire model when new data are added. However, facing the real application environment with growing data, retraining consumes [...] Read more.
Existing deep learning algorithms for synthetic aperture radar (SAR) image recognition are performed with offline data. These methods must use all data to retrain the entire model when new data are added. However, facing the real application environment with growing data, retraining consumes much time and memory space. Class-Incremental Learning (CIL) addresses this problem that deep learning faces in streaming data. The goal of CIL is to enable the model to continuously learn new classes without using all data to retrain the model while maintaining the ability to recognize previous classes. Most of the CIL methods adopt a replay strategy to realize it. However, the number of retained samples is too small to carry enough information. The replay strategy is still trapped by forgetting previous knowledge. For this reason, we propose a CIL method for SAR images based on self-sustainment guidance representation. The method uses the vision transformer (ViT) structure as the basic framework. We add a dynamic query navigation module to enhance the model’s ability to learn the new classes. This module stores special information about classes and uses it to guide the direction of feature extraction in subsequent model learning. In addition, the method also comprises a structural extension module to defend the forgetting of old classes when the model learns new knowledge. It is constructed to maintain the representation of the model in previous classes. The model will learn under the coordinated guidance of old and new information. Experiments on the Moving and Stationary Target Acquisition and Recognition (MSTAR) dataset show that our method performs well with remarkable advantages in CIL tasks. This method has a better accuracy rate and performance dropping rate than state-of-the-art methods under the same setting and maintains the ability of incremental learning with fewer replay samples. Additionally, experiments on a popular image dataset (CIFAR100) also demonstrate the scalability of our approach. Full article
Show Figures

Figure 1

26 pages, 9769 KiB  
Article
Few-Shot PolSAR Ship Detection Based on Polarimetric Features Selection and Improved Contrastive Self-Supervised Learning
by Weixing Qiu, Zongxu Pan and Jianwei Yang
Remote Sens. 2023, 15(7), 1874; https://doi.org/10.3390/rs15071874 - 31 Mar 2023
Cited by 5 | Viewed by 1780
Abstract
Deep learning methods have been widely studied in the field of polarimetric synthetic aperture radar (PolSAR) ship detection over the past few years. However, the backscattering of manmade targets, including ships, is sensitive to the relative geometry between target orientation and radar line [...] Read more.
Deep learning methods have been widely studied in the field of polarimetric synthetic aperture radar (PolSAR) ship detection over the past few years. However, the backscattering of manmade targets, including ships, is sensitive to the relative geometry between target orientation and radar line of sight, which makes the diversity of polarimetric and spatial features of ships. The diversity of scattering leads to a relative increase in the scarcity of PolSAR-labeled samples, which are difficult to obtain. To solve the abovementioned issue and extract the polarimetric and spatial features of PolSAR images better, this paper proposes a few-shot PolSAR ship detection method based on the combination of constructed polarimetric input data selection and improved contrastive self-supervised learning (CSSL) pre-training. Specifically, eight polarimetric feature extraction methods are adopted to construct deep learning network input data with polarimetric features. The backbone is pre-trained with un-labeled PolSAR input data through an improved CSSL method without negative samples, which enhances the representation capability by the multi-scale feature fusion module (MFFM) and implements a regularization strategy by the mix-up auxiliary pathway (MUAP). The pre-trained backbone is applied to the downstream ship detection network; only a few labeled samples are used for fine-tuning and the construction method of polarimetric input data with the best detection effect is studied. The comparison and ablation experiment results on the self-established PolSAR ship detection dataset verify the superiority of the proposed method, especially in the case of few-shot learning. Full article
Show Figures

Graphical abstract

25 pages, 22178 KiB  
Article
An Anchor-Free Detection Algorithm for SAR Ship Targets with Deep Saliency Representation
by Jianming Lv, Jie Chen, Zhixiang Huang, Huiyao Wan, Chunyan Zhou, Daoyuan Wang, Bocai Wu and Long Sun
Remote Sens. 2023, 15(1), 103; https://doi.org/10.3390/rs15010103 - 24 Dec 2022
Cited by 5 | Viewed by 2076
Abstract
Target detection in synthetic aperture radar (SAR) images has a wide range of applications in military and civilian fields. However, for engineering applications involving edge deployment, it is difficult to find a suitable balance of accuracy and speed for anchor-based SAR image target [...] Read more.
Target detection in synthetic aperture radar (SAR) images has a wide range of applications in military and civilian fields. However, for engineering applications involving edge deployment, it is difficult to find a suitable balance of accuracy and speed for anchor-based SAR image target detection algorithms. Thus, an anchor-free detection algorithm for SAR ship targets with deep saliency representation, called SRDet, is proposed in this paper to improve SAR ship detection performance against complex backgrounds. First, we design a data enhancement method considering semantic relationships. Second, the state-of-the-art anchor-free target detection framework CenterNet2 is used as a benchmark, and a new feature-enhancing lightweight backbone, called LWBackbone, is designed to reduce the number of model parameters while effectively extracting the salient features of SAR targets. Additionally, a new mixed-domain attention mechanism, called CNAM, is proposed to effectively suppress interference from complex land backgrounds and highlight the target area. Finally, we construct a receptive-field-enhanced detection head module, called RFEHead, to improve the multiscale perception performance of the detection head. Experimental results based on three large-scale SAR target detection datasets, SSDD, HRSID and SAR-ship-dataset, show that our algorithm achieves a better balance between ship target detection accuracy and speed and exhibits excellent generalization performance. Full article
Show Figures

Graphical abstract

25 pages, 5239 KiB  
Article
Novel Asymmetric Pyramid Aggregation Network for Infrared Dim and Small Target Detection
by Guangrui Lv, Lili Dong, Junke Liang and Wenhai Xu
Remote Sens. 2022, 14(22), 5643; https://doi.org/10.3390/rs14225643 - 08 Nov 2022
Cited by 7 | Viewed by 1406
Abstract
Robust and efficient detection of small infrared target is a critical and challenging task in infrared search and tracking applications. The size of the small infrared targets is relatively tiny compared to the ordinary targets, and the sizes and appearances of the these [...] Read more.
Robust and efficient detection of small infrared target is a critical and challenging task in infrared search and tracking applications. The size of the small infrared targets is relatively tiny compared to the ordinary targets, and the sizes and appearances of the these targets in different scenarios are quite different. Besides, these targets are easily submerged in various background noise. To tackle the aforementioned challenges, a novel asymmetric pyramid aggregation network (APANet) is proposed. Specifically, a pyramid structure integrating dual attention and dense connection is firstly constructed, which can not only generate attention-refined multi-scale features in different layers, but also preserve the primitive features of infrared small targets among multi-scale features. Then, the adjacent cross-scale features in these multi-scale information are sequentially modulated through pair-wise asymmetric combination. This mutual dynamic modulation can continuously exchange heterogeneous cross-scale information along the layer-wise aggregation path until an inverted pyramid is generated. In this way, the semantic features of lower-level network are enriched by incorporating local focus from higher-level network while the detail features of high-level network are refined by embedding point-wise focus from lower-level network, which can highlight small target features and suppress background interference. Subsequently, recursive asymmetric fusion is designed to further dynamically modulate and aggregate high resolution features of different layers in the inverted pyramid, which can also enhance the local high response of small target. Finally, a series of comparative experiments are conducted on two public datasets, and the experimental results show that the APANet can more accurately detect small targets compared to some state-of-the-art methods. Full article
Show Figures

Graphical abstract

18 pages, 15642 KiB  
Article
A New Ship Detection Algorithm in Optical Remote Sensing Images Based on Improved R3Det
by Jianfeng Li, Zongfeng Li, Mingxu Chen, Yongling Wang and Qinghua Luo
Remote Sens. 2022, 14(19), 5048; https://doi.org/10.3390/rs14195048 - 10 Oct 2022
Cited by 5 | Viewed by 1999
Abstract
The task of ship target detection based on remote sensing images has attracted more and more attention because of its important value in civil and military fields. To solve the problem of low accuracy in ship target detection in optical remote sensing ship [...] Read more.
The task of ship target detection based on remote sensing images has attracted more and more attention because of its important value in civil and military fields. To solve the problem of low accuracy in ship target detection in optical remote sensing ship images due to complex scenes and large-target-scale differences, an improved R3Det algorithm is proposed in this paper. On the basis of R3Det, a feature pyramid network (FPN) structure is replaced by a search architecture-based feature pyramid network (NAS FPN) so that the network can adaptively learn and select the feature combination update and enrich the multiscale feature information. After the feature extraction network, a shallow feature is added to the context information enhancement (COT) module to supplement the small target semantic information. An efficient channel attention (ECA) module is added to make the network gather in the target area. The improved algorithm is applied to the ship data in the remote sensing image data set FAIR1M. The effectiveness of the improved model in a complex environment and for small target detection is verified through comparison experiments with R3Det and other models. Full article
Show Figures

Graphical abstract

28 pages, 4616 KiB  
Article
A Few-Shot Learning Method for SAR Images Based on Weighted Distance and Feature Fusion
by Fei Gao, Jingming Xu, Rongling Lang, Jun Wang, Amir Hussain and Huiyu Zhou
Remote Sens. 2022, 14(18), 4583; https://doi.org/10.3390/rs14184583 - 14 Sep 2022
Cited by 10 | Viewed by 2256
Abstract
Convolutional Neural Network (CNN) has been widely applied in the field of synthetic aperture radar (SAR) image recognition. Nevertheless, CNN-based recognition methods usually encounter the problem of poor feature representation ability due to insufficient labeled SAR images. In addition, the large inner-class variety [...] Read more.
Convolutional Neural Network (CNN) has been widely applied in the field of synthetic aperture radar (SAR) image recognition. Nevertheless, CNN-based recognition methods usually encounter the problem of poor feature representation ability due to insufficient labeled SAR images. In addition, the large inner-class variety and high cross-class similarity of SAR images pose a challenge for classification. To alleviate the problems mentioned above, we propose a novel few-shot learning (FSL) method for SAR image recognition, which is composed of the multi-feature fusion network (MFFN) and the weighted distance classifier (WDC). The MFFN is utilized to extract input images’ features, and the WDC outputs the classification results based on these features. The MFFN is constructed by adding a multi-scale feature fusion module (MsFFM) and a hand-crafted feature insertion module (HcFIM) to a standard CNN. The feature extraction and representation capability can be enhanced by inserting the traditional hand-crafted features as auxiliary features. With the aid of information from different scales of features, targets of the same class can be more easily aggregated. The weight generation module in WDC is designed to generate category-specific weights for query images. The WDC distributes these weights along the corresponding Euclidean distance to tackle the high cross-class similarity problem. In addition, weight generation loss is proposed to improve recognition performance by guiding the weight generation module. Experimental results on the Moving and Stationary Target Acquisition and Recognition (MSTAR) dataset and the Vehicle and Aircraft (VA) dataset demonstrate that our proposed method surpasses several typical FSL methods. Full article
Show Figures

Figure 1

19 pages, 10022 KiB  
Article
MEA-Net: A Lightweight SAR Ship Detection Model for Imbalanced Datasets
by Yiyu Guo and Luoyu Zhou
Remote Sens. 2022, 14(18), 4438; https://doi.org/10.3390/rs14184438 - 06 Sep 2022
Cited by 8 | Viewed by 1962
Abstract
The existing synthetic aperture radar (SAR) ship datasets have an imbalanced number of inshore and offshore ship targets, and the number of small, medium and large ship targets differs greatly. At the same time, the existing SAR ship detection models in the application [...] Read more.
The existing synthetic aperture radar (SAR) ship datasets have an imbalanced number of inshore and offshore ship targets, and the number of small, medium and large ship targets differs greatly. At the same time, the existing SAR ship detection models in the application have a huge structure and require high computing resources. To solve these problems, we propose a SAR ship detection model named mask efficient adaptive network (MEA-Net), which is lightweight and high-accuracy for imbalanced datasets. Specifically, we propose the following three innovative modules. Firstly, we propose a mask data balance augmentation (MDBA) method, which solves the imbalance of sample data between inshore and offshore ship targets by combining mathematical morphological processing and ship label data to greatly improve the ability of the model to detect inshore ship targets. Secondly, we propose an efficient attention mechanism (EAM), which effectively integrates channel features and spatial features through one-dimensional convolution and two-dimensional convolution, to improve the feature extraction ability of the model for SAR ship targets. Thirdly, we propose an adaptive receptive field block (ARFB), which can achieve more effective multi-scale detection by establishing the mapping relationship between the size of the convolution kernel and the channel of feature map, to improve the detection ability of the model for ship targets of different sizes. Finally, MEA-Net is deployed on the Jeston Nano edge computing device of the 2 GB version. We conducted experimental validation on the SSDD and HRSID datasets. Compared with the baseline, the AP of MEA-Net increased by 2.18% on the SSDD dataset and 3.64% on the HRSID dataset. The FLOPs and model parameters of MEA-Net were only 2.80 G and 0.96 M, respectively. In addition, the FPS reached 6.31 on the Jeston Nano, which has broad application prospects. Full article
Show Figures

Graphical abstract

Other

Jump to: Research

16 pages, 4163 KiB  
Technical Note
Comparative Analysis of Remote Sensing Storage Tank Detection Methods Based on Deep Learning
by Lu Fan, Xiaoying Chen, Yong Wan and Yongshou Dai
Remote Sens. 2023, 15(9), 2460; https://doi.org/10.3390/rs15092460 - 07 May 2023
Cited by 1 | Viewed by 1758
Abstract
Since the Industrial Revolution, methane has become the second most important greenhouse gas component after CO2 and the second most important culprit of global warming, leading to serious climate change problems such as droughts, fires, floods, and glacial melting. While most of [...] Read more.
Since the Industrial Revolution, methane has become the second most important greenhouse gas component after CO2 and the second most important culprit of global warming, leading to serious climate change problems such as droughts, fires, floods, and glacial melting. While most of the methane in the atmosphere comes from emissions from energy activities such as petroleum refining, storage tanks are an important source of methane emissions during the extraction and processing of crude oil and natural gas. Therefore, the use of high-resolution remote sensing image data for oil and gas production sites to achieve efficient and accurate statistics for storage tanks is important to promote the strategic goals of “carbon neutrality and carbon peaking”. Compared with traditional statistical methods for studying oil storage tanks, deep learning-based target detection algorithms are more powerful for multi-scale targets and complex background conditions. In this paper, five deep learning detection algorithms, Faster RCNN, YOLOv5, YOLOv7, RetinaNet and SSD, were selected to conduct experiments on 3568 remote sensing images from five different datasets. The results show that the average accuracy of the Faster RCNN, YOLOv5, YOLOv7 and SSD algorithms is above 0.84, and the F1 scores of YOLOv5, YOLOv7 and SSD algorithms are above 0.80, among which the highest detection accuracy is shown by the SSD algorithm at 0.897 with a high F1 score, while the lowest average accuracy is shown by RetinaNet at only 0.639. The training results of the five algorithms were validated on three images containing differently sized oil storage tanks in complex backgrounds, and the validation results obtained were better, providing more accurate references for practical detection applications in remote sensing of oil storage tank targets in the future. Full article
Show Figures

Graphical abstract

Back to TopTop