Remote Sensing

Research

Jump to: Other

19 pages, 6307 KiB

Open AccessArticle

Measurement of Downwelling Radiance Using a Low-Cost Compact Fourier-Transform Infrared System for Monitoring Atmospheric Conditions

by Haklim Choi and Jongjin Seo

Remote Sens. 2024, 16(7), 1136; https://doi.org/10.3390/rs16071136 - 25 Mar 2024

Viewed by 1173

Abstract

Temperature and water vapor play crucial roles in the Earth’s climate system, and it is important to understand and monitor the variation in the thermodynamic profile within the lower troposphere. Among various observation platforms for understanding the vertical structure of temperature and humidity, [...] Read more.

Temperature and water vapor play crucial roles in the Earth’s climate system, and it is important to understand and monitor the variation in the thermodynamic profile within the lower troposphere. Among various observation platforms for understanding the vertical structure of temperature and humidity, ground-based Fourier-transform infrared (FTIR) can provide detailed information about the lower troposphere by complementing the limitations of radiosonde or satellite methods. However, these ground-based systems have limitations in terms of cost, operation, and mobility. Herein, we introduce a cost-effective and easily deployable FTIR observation system designed to enhance monitoring capabilities for atmospheric conditions. The atmospheric downwelling radiance spectrum of sky is measured by applying a real-time radiative calibration using a blackbody. From the observed radiance spectrum, the thermodynamic profile (temperature and the water vapor mixing ratio) of the lower troposphere was retrieved using an algorithm based on the optimal estimation method (OEM). The retrieved vertical structure results in the lower troposphere were similar to the fifth-generation reanalysis database (ERA-5) of the European Center for Medium-range Weather Forecasts (ECMWF) and the National Centers for Environmental Prediction final analysis (NCEP FNL). This provides a potential possibility for monitoring atmospheric conditions by a compact FTIR system. Full article

(This article belongs to the Special Issue Remote Sensing and Machine Learning of Signal and Image Processing)

► Show Figures

Figure 1

20 pages, 7722 KiB

Open AccessArticle

Frequency Agile Anti-Interference Technology Based on Reinforcement Learning Using Long Short-Term Memory and Multi-Layer Historical Information Observation

by Weihao Shi, Shanhong Guo, Xiaoyu Cong, Weixing Sheng, Jing Yan and Jinkun Chen

Remote Sens. 2023, 15(23), 5467; https://doi.org/10.3390/rs15235467 - 23 Nov 2023

Viewed by 1696

Abstract

In modern electronic warfare, radar intelligence has become increasingly crucial when dealing with complex interference environments. This paper combines radar agile frequency technology with reinforcement learning to achieve adaptive frequency hopping for radar anti-jamming. Unlike traditional reinforcement learning with Markov decision processes (MDPs), [...] Read more.

In modern electronic warfare, radar intelligence has become increasingly crucial when dealing with complex interference environments. This paper combines radar agile frequency technology with reinforcement learning to achieve adaptive frequency hopping for radar anti-jamming. Unlike traditional reinforcement learning with Markov decision processes (MDPs), the interaction between radar and jammers occurs within the partially observable Markov decision processes (POMDPs). In this context, the partial observation information available to the agent does not strictly satisfy the Markov property. This paper uses multiple layers of historical observation information to solve this problem. Historical observations can be viewed as a time series, and time-sensitive networks are employed to extract the temporal information embedded within the observations. In addition, the reward function is optimized to facilitate the faster learning of the agent in the jammer sweep environment. This simulation shows that the optimization of the agent state, network structure, and reward function can effectively help the radar to resist jamming. Full article

(This article belongs to the Special Issue Remote Sensing and Machine Learning of Signal and Image Processing)

► Show Figures

Graphical abstract

25 pages, 2100 KiB

Open AccessArticle

Few-Shot High-Resolution Range Profile Ship Target Recognition Based on Task-Specific Meta-Learning with Mixed Training and Meta Embedding

by Yingying Kong, Yuxuan Zhang, Xiangyang Peng and Henry Leung

Remote Sens. 2023, 15(22), 5301; https://doi.org/10.3390/rs15225301 - 9 Nov 2023

Cited by 2 | Viewed by 1493

Abstract

High-resolution range profile (HRRP), characterized by its high availability and rich target structural information, has been extensively studied. However, HRRP-based target recognition methods using closed datasets exhibit limitations when it comes to identifying new classes of targets. The scarcity of samples for new [...] Read more.

High-resolution range profile (HRRP), characterized by its high availability and rich target structural information, has been extensively studied. However, HRRP-based target recognition methods using closed datasets exhibit limitations when it comes to identifying new classes of targets. The scarcity of samples for new classes leads to overfitting during the deep learning process, and the similarity in the scattering structures of different ships, combined with the significant structural differences among samples of the same ship, contribute to a high level of confusion among targets. To address these challenges, this paper proposed Task-Specific Mate-learning (TSML) for few-shot HRRP. Firstly, a Task-Adaptive Mixed Transfer (TAMT) strategy is proposed, which combines basic learning with meta-learning, to reduce the likelihood of overfitting and enhance adaptability for recognizing new classes of ships. Secondly, a Prototype Network is introduced to enable the recognition of new classes of targets with limited samples. Additionally, a Space-Adjusted Meta Embedding (SAME) is proposed based on the Prototype Network. This embedding function, designed for HRRP data, modifies the distances between samples in meta-tasks by increasing the distances between samples from different ships and decreasing the distances between samples from the same ship. The proposed method is evaluated based on an actual measured HRRP dataset and the experimental results prove that the proposed method can more accurately recognize the unknown ship classes with a small number of labels by learning the known classes of ships. In addition, the method has a degree of robustness to the number of training samples and a certain generalization ability, which can produce improved results when applied to other backbones. Full article

(This article belongs to the Special Issue Remote Sensing and Machine Learning of Signal and Image Processing)

► Show Figures

Graphical abstract

21 pages, 9032 KiB

Open AccessArticle

Faster and Better: A Lightweight Transformer Network for Remote Sensing Scene Classification

by Xinyan Huang, Fang Liu, Yuanhao Cui, Puhua Chen, Lingling Li and Pengfang Li

Remote Sens. 2023, 15(14), 3645; https://doi.org/10.3390/rs15143645 - 21 Jul 2023

Cited by 18 | Viewed by 3267

Abstract

Remote sensing (RS) scene classification has received considerable attention due to its wide applications in the RS community. Many methods based on convolutional neural networks (CNNs) have been proposed to classify complex RS scenes, but they cannot fully capture the context in RS [...] Read more.

Remote sensing (RS) scene classification has received considerable attention due to its wide applications in the RS community. Many methods based on convolutional neural networks (CNNs) have been proposed to classify complex RS scenes, but they cannot fully capture the context in RS images because of the lack of long-range dependencies (the dependency relationship between two distant elements). Recently, some researchers fine-tuned the large pretrained vision transformer (ViT) on small RS datasets to extract long-range dependencies effectively in RS scenes. However, it usually takes more time to fine-tune the ViT on account of high computational complexity. The lack of good local feature representation in the ViT limits classification performance improvement. To this end, we propose a lightweight transformer network (LTNet) for RS scene classification. First, a multi-level group convolution (MLGC) module is presented. It enriches the diversity of local features and requires a lower computational cost by co-representing multi-level and multi-group features in a single module. Then, based on the MLGC module, a lightweight transformer block, LightFormer, was designed to capture global dependencies with fewer computing resources. Finally, the LTNet was built using the MLGC and LightFormer. The experiments of fine-tuning the LTNet on four RS scene classification datasets demonstrate that the proposed network achieves a competitive classification performance under less training time. Full article

(This article belongs to the Special Issue Remote Sensing and Machine Learning of Signal and Image Processing)

► Show Figures

Figure 1

19 pages, 4846 KiB

Open AccessArticle

Dual-View Hyperspectral Anomaly Detection via Spatial Consistency and Spectral Unmixing

by Jingyan Zhang, Xiangrong Zhang and Licheng Jiao

Remote Sens. 2023, 15(13), 3330; https://doi.org/10.3390/rs15133330 - 29 Jun 2023

Cited by 2 | Viewed by 1722

Abstract

Anomaly detection is a crucial task for hyperspectral image processing. Most popular methods detect anomalies at the pixel level, while a few algorithms for anomaly detection only utilize subpixel level unmixing technology to extract features without fundamentally analyzing the anomalies. To better detect [...] Read more.

Anomaly detection is a crucial task for hyperspectral image processing. Most popular methods detect anomalies at the pixel level, while a few algorithms for anomaly detection only utilize subpixel level unmixing technology to extract features without fundamentally analyzing the anomalies. To better detect and separate the anomalies from the background, this paper proposes a dual-view hyperspectral anomaly detection method by taking account of the anomaly analysis at both levels mentioned. At the pixel level, the spectral angular distance is adopted to calculate the similarities between the central pixel and its neighbors in order to further mine the spatial consistency for anomaly detection. On the other hand, from the aspect of the subpixel level analysis, it is considered that the difference between the anomaly and the background usually arises from dissimilar endmembers, where the unmixing will be fully implemented. Finally, the detection results of both views are fused to obtain the anomalies. Overall, the proposed algorithm not only interprets and analyzes the anomalies from dual levels, but also fully employs the unmixing for anomaly detection. Additionally, the performance of multiple data sets also confirmed the effectiveness of the proposed algorithm. Full article

(This article belongs to the Special Issue Remote Sensing and Machine Learning of Signal and Image Processing)

► Show Figures

Graphical abstract

18 pages, 6631 KiB

Open AccessArticle

Removing Time Dispersion from Elastic Wave Modeling with the pix2pix Algorithm Based on cGAN

by Teng Xu, Hongyong Yan, Hui Yu and Zhiyong Zhang

Remote Sens. 2023, 15(12), 3120; https://doi.org/10.3390/rs15123120 - 14 Jun 2023

Cited by 6 | Viewed by 1655

Abstract

The finite-difference (FD) method is one of the most commonly used numerical methods for elastic wave modeling. However, due to the difference approximation of the derivative, the time dispersion phenomenon cannot be avoided. This paper proposes the use of pix2pix algorithm based on [...] Read more.

The finite-difference (FD) method is one of the most commonly used numerical methods for elastic wave modeling. However, due to the difference approximation of the derivative, the time dispersion phenomenon cannot be avoided. This paper proposes the use of pix2pix algorithm based on a conditional generative adversarial network (cGAN) for removing time dispersion from elastic FD modeling. Firstly, we analyze the time dispersion of elastic wave FD modeling. Then, we discuss the pix2pix algorithm based on cGAN, improve the loss function of the pix2pix algorithm by introducing a Sobel operator, and analyze the parameter selection of the network model for the pix2pix algorithm. Finally, we verify the feasibility and effectiveness of the pix2pix algorithm in removing time dispersion from elastic wave FD modeling through testing some model simulation data. Full article

(This article belongs to the Special Issue Remote Sensing and Machine Learning of Signal and Image Processing)

► Show Figures

Graphical abstract

22 pages, 9929 KiB

Open AccessArticle

High-Precision Segmentation of Buildings with Small Sample Sizes Based on Transfer Learning and Multi-Scale Fusion

by Xiaobin Xu, Haojie Zhang, Yingying Ran and Zhiying Tan

Remote Sens. 2023, 15(9), 2436; https://doi.org/10.3390/rs15092436 - 5 May 2023

Cited by 5 | Viewed by 2496

Abstract

In order to improve the accuracy of the segmentation of buildings with small sample sizes, this paper proposes a building-segmentation network, ResFAUnet, with transfer learning and multi-scale feature fusion. The network is based on AttentionUnet. The backbone of the encoder is replaced by [...] Read more.

In order to improve the accuracy of the segmentation of buildings with small sample sizes, this paper proposes a building-segmentation network, ResFAUnet, with transfer learning and multi-scale feature fusion. The network is based on AttentionUnet. The backbone of the encoder is replaced by the ResNeXt101 network for feature extraction, and the attention mechanism of the skip connection is preserved to fuse the shallow features of the encoding part and the deep features of the decoding part. In the decoder, the feature-pyramid structure is used to fuse the feature maps of different scales. More features can be extracted from limited image samples. The proposed network is compared with current classical semantic segmentation networks, Unet, SuUnet, FCN, and SegNet. The experimental results show that in the dataset selected in this paper, the precision indicators of ResFAUnet are improved by 4.77%, 2.3%, 2.11%, and 1.57%, respectively, compared with the four comparison networks. Full article

(This article belongs to the Special Issue Remote Sensing and Machine Learning of Signal and Image Processing)

► Show Figures

Figure 1

21 pages, 73507 KiB

Open AccessArticle

MFNet: Mutual Feature-Aware Networks for Remote Sensing Change Detection

by Qi Zhang, Yao Lu, Sicheng Shao, Li Shen, Fei Wang and Xuetao Zhang

Remote Sens. 2023, 15(8), 2145; https://doi.org/10.3390/rs15082145 - 19 Apr 2023

Cited by 3 | Viewed by 2245

Abstract

Remote sensing change detection involves detecting pixels that have changed from a bi-temporal image of the same location. Current mainstream change detection models use encoder-decoder structures as well as Siamese networks. However, there are still some challenges with this: (1) Existing change feature [...] Read more.

Remote sensing change detection involves detecting pixels that have changed from a bi-temporal image of the same location. Current mainstream change detection models use encoder-decoder structures as well as Siamese networks. However, there are still some challenges with this: (1) Existing change feature fusion approaches do not take into account the symmetry of change features, which leads to information loss; (2) The encoder is independent of the change detection task, and feature extraction is performed separately for dual-time images, which leads to underutilization of the encoder parameters; (3) There are problems of unbalanced positive and negative samples and bad edge region detection. To solve the above problems, a mutual feature-aware network (MFNet) is proposed in this paper. Three modules are proposed for the purpose: (1) A symmetric change feature fusion module (SCFM), which uses double-branch feature selection without losing feature information and focuses explicitly on focal spatial regions based on cosine similarity to introduce strong a priori information; (2) A mutual feature-aware module (MFAM), which introduces change features in advance at the encoder stage and uses a cross-type attention mechanism for long-range dependence modeling; (3) A loss function for edge regions. After detailed experiments, the F1 scores of MFNet on SYSU-CD and LEVIR-CD were 83.11% and 91.52%, respectively, outperforming several advanced algorithms, demonstrating the effectiveness of the proposed method. Full article

(This article belongs to the Special Issue Remote Sensing and Machine Learning of Signal and Image Processing)

► Show Figures

Figure 1

24 pages, 19626 KiB

Open AccessArticle

An Innovative Approach for Effective Removal of Thin Clouds in Optical Images Using Convolutional Matting Model

by Renzhe Wu, Guoxiang Liu, Jichao Lv, Yin Fu, Xin Bao, Age Shama, Jialun Cai, Baikai Sui, Xiaowen Wang and Rui Zhang

Remote Sens. 2023, 15(8), 2119; https://doi.org/10.3390/rs15082119 - 17 Apr 2023

Cited by 1 | Viewed by 2384

Abstract

Clouds are the major source of clutter in optical remote sensing (RS) images. Approximately 60% of the Earth’s surface is covered by clouds, with the equatorial and Tibetan Plateau regions being the most affected. Although the implementation of techniques for cloud removal can [...] Read more.

Clouds are the major source of clutter in optical remote sensing (RS) images. Approximately 60% of the Earth’s surface is covered by clouds, with the equatorial and Tibetan Plateau regions being the most affected. Although the implementation of techniques for cloud removal can significantly improve the efficiency of remote sensing imagery, its use is severely restricted due to the poor timeliness of time-series cloud removal techniques and the distortion-prone nature of single-frame cloud removal techniques. To thoroughly remove thin clouds from remote sensing imagery, we propose the Saliency Cloud Matting Convolutional Neural Network (SCM-CNN) from an image fusion perspective. This network can automatically balance multiple loss functions, extract the cloud opacity and cloud top reflectance intensity from cloudy remote sensing images, and recover ground surface information under thin cloud cover through inverse operations. The SCM-CNN was trained on simulated samples and validated on both simulated samples and Sentinel-2 images, achieving average peak signal-to-noise ratios (PSNRs) of 30.04 and 25.32, respectively. Comparative studies demonstrate that the SCM-CNN model is more effective in performing cloud removal on individual remote sensing images, is robust, and can recover ground surface information under thin cloud cover without compromising the original image. The method proposed in this article can be widely promoted in regions with year-round cloud cover, providing data support for geological hazard, vegetation, and frozen area studies, among others. Full article

(This article belongs to the Special Issue Remote Sensing and Machine Learning of Signal and Image Processing)

► Show Figures

Figure 1

26 pages, 61627 KiB

Open AccessArticle

KeyShip: Towards High-Precision Oriented SAR Ship Detection Using Key Points

by Junyao Ge, Yiping Tang, Kaitai Guo, Yang Zheng, Haihong Hu and Jimin Liang

Remote Sens. 2023, 15(8), 2035; https://doi.org/10.3390/rs15082035 - 12 Apr 2023

Cited by 6 | Viewed by 2332

Abstract

Synthetic Aperture Radar (SAR) is an all-weather sensing technology that has proven its effectiveness for ship detection. However, detecting ships accurately with oriented bounding boxes (OBB) on SAR images is challenging due to arbitrary ship orientations and misleading scattering. In this article, we [...] Read more.

Synthetic Aperture Radar (SAR) is an all-weather sensing technology that has proven its effectiveness for ship detection. However, detecting ships accurately with oriented bounding boxes (OBB) on SAR images is challenging due to arbitrary ship orientations and misleading scattering. In this article, we propose a novel anchor-free key-point-based detection method, KeyShip, for detecting orientated SAR ships with high precision. Our approach uses a shape descriptor to model a ship as a combination of three types of key points located at the short-edge centers, long-edge centers, and the target center. These key points are detected separately and clustered based on predicted shape descriptors to construct the final OBB detection results. To address the boundary problem that arises with the shape descriptor representation, we propose a soft training target assignment strategy that facilitates successful shape descriptor training and implicitly learns the shape information of the targets. Our experimental results on three datasets (SSDD, RSDD, and HRSC2016) demonstrate our proposed method’s high performance and robustness. Full article

(This article belongs to the Special Issue Remote Sensing and Machine Learning of Signal and Image Processing)

► Show Figures

Figure 1

24 pages, 16710 KiB

Open AccessArticle

Remote Sensing Image Change Detection Based on Deep Multi-Scale Multi-Attention Siamese Transformer Network

by Mengxuan Zhang, Zhao Liu, Jie Feng, Long Liu and Licheng Jiao

Remote Sens. 2023, 15(3), 842; https://doi.org/10.3390/rs15030842 - 2 Feb 2023

Cited by 41 | Viewed by 6006

Abstract

Change detection is a technique that can observe changes in the surface of the earth dynamically. It is one of the most significant tasks in remote sensing image processing. In the past few years, with the ability of extracting rich deep image features, [...] Read more.

Change detection is a technique that can observe changes in the surface of the earth dynamically. It is one of the most significant tasks in remote sensing image processing. In the past few years, with the ability of extracting rich deep image features, the deep learning techniques have gained popularity in the field of change detection. In order to obtain obvious image change information, the attention mechanism is added in the decoder and output stage in many deep learning-based methods. Many of these approaches neglect to upgrade the ability of the encoders and the feature extractors to extract the representational features. To resolve this problem, this study proposes a deep multi-scale multi-attention siamese transformer network. A special contextual attention module combining a convolution and self-attention module is introduced into the siamese feature extractor to enhance the global representation ability. A lightly efficient channel attention block is added in the siamese feature extractor to obtain the information interaction among different channels. Furthermore, a multi-scale feature fusion module is proposed to fuse the features from different stages of the siamese feature extractor, and it can detect objects of different sizes and irregularities. To increase the accuracy of the proposed approach, the transformer module is utilized to model the long-range context in two-phase images. The experimental results on the LEVIR-CD and the CCD datasets show the effectiveness of the proposed network. Full article

(This article belongs to the Special Issue Remote Sensing and Machine Learning of Signal and Image Processing)

► Show Figures

Figure 1

18 pages, 6287 KiB

Open AccessArticle

SSML: Spectral-Spatial Mutual-Learning-Based Framework for Hyperspectral Pansharpening

by Xianlin Peng, Yihao Fu, Shenglin Peng, Kai Ma, Lu Liu and Jun Wang

Remote Sens. 2022, 14(18), 4682; https://doi.org/10.3390/rs14184682 - 19 Sep 2022

Viewed by 2553

Abstract

This paper considers problems associated with the large size of the hyperspectral pansharpening network and difficulties associated with learning its spatial-spectral features. We propose a deep mutual-learning-based framework (SSML) for spectral-spatial information mining and hyperspectral pansharpening. In this framework, a deep mutual-learning mechanism [...] Read more.

This paper considers problems associated with the large size of the hyperspectral pansharpening network and difficulties associated with learning its spatial-spectral features. We propose a deep mutual-learning-based framework (SSML) for spectral-spatial information mining and hyperspectral pansharpening. In this framework, a deep mutual-learning mechanism is introduced to learn spatial and spectral features from each other through information transmission, which achieves better fusion results without entering too many parameters. The proposed SSML framework consists of two separate networks for learning spectral and spatial features of HSIs and panchromatic images (PANs). A hybrid loss function containing constrained spectral and spatial information is designed to enforce mutual learning between the two networks. In addition, a mutual-learning strategy is used to balance the spectral and spatial feature learning to improve the performance of the SSML path compared to the original. Extensive experimental results demonstrated the effectiveness of the mutual-learning mechanism and the proposed hybrid loss function for hyperspectral pan-sharpening. Furthermore, a typical deep-learning method was used to confirm the proposed framework’s capacity for generalization. Ideal performance was observed in all cases. Moreover, multiple experiments analysing the parameters used showed that the proposed method achieved better fusion results without adding too many parameters. Thus, the proposed SSML represents a promising framework for hyperspectral pansharpening. Full article

(This article belongs to the Special Issue Remote Sensing and Machine Learning of Signal and Image Processing)

► Show Figures

Figure 1

24 pages, 10039 KiB

Open AccessArticle

Hyperspectral Image Classification Based on a Least Square Bias Constraint Additional Empirical Risk Minimization Nonparallel Support Vector Machine

by Guangxin Liu, Liguo Wang and Danfeng Liu

Remote Sens. 2022, 14(17), 4263; https://doi.org/10.3390/rs14174263 - 29 Aug 2022

Cited by 5 | Viewed by 2176

Abstract

Hyperspectral image classification technology is important for the application of hyperspectral technology. Support vector machines (SVMs) work well in supervised classifications of hyperspectral images; however, they still have some shortcomings, and their use of a parallel decision plane makes it difficult to conform [...] Read more.

Hyperspectral image classification technology is important for the application of hyperspectral technology. Support vector machines (SVMs) work well in supervised classifications of hyperspectral images; however, they still have some shortcomings, and their use of a parallel decision plane makes it difficult to conform to real hyperspectral data distribution. The improved nonparallel support vector machine based on SVMs, i.e., the bias constraint additional empirical risk minimization nonparallel support vector machine (BC-AERM-NSVM), has improved classification accuracy compared its predecessor. However, BC-AERM-NSVMs have a more complicated solution problem than SVMs, and if the dataset is too large, the training speed is significantly reduced. To solve this problem, this paper proposes a least squares algorithm, i.e., the least square bias constraint additional empirical risk minimization nonparallel support vector machine (LS-BC-AERM-NSVM). The dual problem of the LS-BC-AERM-NSVM is an unconstrained convex quadratic programming problem, so its solution speed is greatly improved. Experiments on hyperspectral image data demonstrate that the LS-BC-AERM-NSVM displays a vast improvement in terms of solution speed compared with the BC-AERM-NSVM and achieves good classification accuracy. Full article

(This article belongs to the Special Issue Remote Sensing and Machine Learning of Signal and Image Processing)

► Show Figures

Figure 1

Other

Jump to: Research

16 pages, 24028 KiB

Open AccessTechnical Note

Recalibrating Features and Regression for Oriented Object Detection

by Weining Chen, Shicheng Miao, Guangxing Wang and Gong Cheng

Remote Sens. 2023, 15(8), 2134; https://doi.org/10.3390/rs15082134 - 18 Apr 2023

Cited by 8 | Viewed by 2194

Abstract

The objects in remote sensing images are normally densely packed, arbitrarily oriented, and surrounded by complex backgrounds. Great efforts have been devoted to developing oriented object detection models to accommodate such data characteristics. We argue that an effective detection model hinges on three [...] Read more.

The objects in remote sensing images are normally densely packed, arbitrarily oriented, and surrounded by complex backgrounds. Great efforts have been devoted to developing oriented object detection models to accommodate such data characteristics. We argue that an effective detection model hinges on three aspects: feature enhancement, feature decoupling for classification and localization, and an appropriate bounding box regression scheme. In this article, we instantiate the three aspects on top of the classical Faster R-CNN, with three novel components proposed. First, we propose a weighted fusion and refinement (WFR) module, which adaptively weighs multi-level features and leverages the attention mechanism to refine the fused features. Second, we decouple the RoI (region of interest) features for the subsequent classification and localization via a lightweight affine transformation-based feature decoupling (ATFD) module. Third, we propose a post-classification regression (PCR) module for generating the desired quadrilateral bounding boxes. Specifically, PCR predicts the precise vertex location on each side of a predicted horizontal box, by simply learning the following: (i) classify the discretized regression range of the vertex, and (ii) revise the vertex location with an offset. We conduct extensive experiments on the DOTA, DIOR-R, and HRSC2016 datasets to evaluate our method. Full article

(This article belongs to the Special Issue Remote Sensing and Machine Learning of Signal and Image Processing)

► Show Figures

Figure 1

15 pages, 5504 KiB

Open AccessTechnical Note

A Method of SAR Image Automatic Target Recognition Based on Convolution Auto-Encode and Support Vector Machine

by Yang Deng and Yunkai Deng

Remote Sens. 2022, 14(21), 5559; https://doi.org/10.3390/rs14215559 - 4 Nov 2022

Cited by 7 | Viewed by 2501

Abstract

In this paper, a method of Synthetic Aperture Radar (SAR) image Automatic Target Recognition (ATR) based on Convolution Auto-encode (CAE) and Support Vector Machine (SVM) is proposed. Using SVM replaces the traditional softmax as the classifier of the CAE model to classify the [...] Read more.

In this paper, a method of Synthetic Aperture Radar (SAR) image Automatic Target Recognition (ATR) based on Convolution Auto-encode (CAE) and Support Vector Machine (SVM) is proposed. Using SVM replaces the traditional softmax as the classifier of the CAE model to classify the feature vectors extracted by the CAE model, which solves the problem that the softmax classifier is less effective in the nonlinear case. Since the SVM can only solve the binary classification problem, and in order to realize the classification of the class objectives, the SVM were designed to achieve the classification of the input samples. After unsupervised training for CAE, the coding layer is connected with SVM to form a classification network. CAE can extract the features of the data by an unsupervised method, and the nonlinear classification advantage of SVM can classify the features extracted by CAE and improve the accuracy of the object recognition. At the same time, the high-accuracy identification of key targets is required in some special cases. A new initialization method is proposed, which initializes the network parameters by pretraining the key targets and changes the weights of different targets in the loss function to obtain better feature extraction, so it can ensure good multitarget recognition ability while realizing the high recognition accuracy of the key targets. Full article

(This article belongs to the Special Issue Remote Sensing and Machine Learning of Signal and Image Processing)

► Show Figures

Graphical abstract

Journal Menu

Journal Browser

Remote Sensing and Machine Learning of Signal and Image Processing

Share This Special Issue

Special Issue Editors

Special Issue Information

Keywords

Benefits of Publishing in a Special Issue

Published Papers (15 papers)

Research

Other

Further Information

Guidelines

MDPI Initiatives

Follow MDPI