SAR Image Ship Object Generation and Classification With Improved Residual Conditional Generative Adversarial Network.

Machine learning driven object detection and classification within non-visible imagery has an important role in many fields such as night vision, all-weather surveillance and aviation security. ... Furthermore, we show that the generation of interpolated mixed class (non-visible domain) image examples via our novel Conditional CycleGAN Mixup Augmentation (C2GMA) methodology can lead to a significant ... Generative Adversarial Networks (GAN) [10] have significantly impacted data augmentation within DNN training. ...

arXiv:2005.02436v2 fatcat:e2sscf25kbcvtft64jxzmlj22q

Multiple Versions

However, it requires class labels to generate adversarial attacks and suffers significant accuracy dropping on testing data. ... Specifically, we utilize a contrastive learning framework to train a robust DNN with unlabeled data, which aims to maximize the similarity of representations between a random augmentation of a SAR image ... Residual networks such as ResNet18, ResNet101, and DenseNet201 behave well in the classification of black-box adversarial examples. ...

doi:10.3390/rs13204158 fatcat:hummhc2ro5fjnpav4oxm3wxzom

DOAJ Szczepanski

This work is amply introduced by providing general elements of ML-based RSP and by stating the motivations behind them. ... This paper then concludes with a series of open questions and proposed research directions, in order to indicate current gaps and potential future solutions and trends. ... A novel adversarial AE was proposed to improve the orientation generalization ability for SAR-ATR tasks in [400] , which learned a code-image-code cyclic network by adversarial training for the purpose ...

arXiv:2009.13702v1 fatcat:m6am73324zdwba736sn3vmph3i

generative adversarial network (GAN), we designed an image enhancement module driven by object characteristics, which improves the quality of the ship target in the images while augmenting the training ... Experiments were carried out on two types of ship datasets, optical RSI and Synthetic Aperture Radar (SAR) images. ... The generative adversarial network (GAN) [30] is often used to generate sample images with different styles and texture characteristics as a complement to the original dataset. ...

doi:10.3390/rs13071327 fatcat:phazoirsb5cfbajmwnbldx25lu

DOAJ Szczepanski

Acknowledgments The author would like to thank the anonymous reviewers for their insightful comments and suggestions, which definitely made the work more technically sound. ... In [163] , a fully convolutional network (FCC) with 20 layers were proposed for ship detection in SAR images collected by Gaofen-3 and TerraSAR-X. ... Another feasible alternative is data augmentation with generative adversarial network (GAN). ...

doi:10.1109/access.2021.3119561 fatcat:4mwnmgkedbfjfbfrfg6gruvf4m

DOAJ

This paper comprehensively benchmarks several advanced deep learning models for SAR ATR with multiple distinct SAR imagery datasets Specifically we train and test five SAR image classifiers based on Residual ... Synthetic Aperture Radar SAR Automatic Target Recognition ATR is a key technique of remote-sensing image recognition which can be supported by deep neural networks The existing works of SAR ATR mostly ... 10] SynthWakeSAR [11] SAR device Airborne-based Ground-based Airborne-based Target object Military vehicles Ceramic objects Ship wakes Data type Images Raw radar data Images Training size 27, 000 5, 147 ...

arXiv:2312.06940v1 fatcat:qdyatvqksnevfgicmnbtlgkvlq

Open Access

Finally, we validate the effectiveness of our method on three publicly available SAR ship detection datasets, SAR-Ship-Dataset, high-resolution SAR images dataset (HRSID), and SAR ship detection dataset ... With the development of deep learning, synthetic aperture radar (SAR) ship detection and recognition based on deep learning have gained widespread application and advancement. ... Introduction SAR is an active microwave imaging sensor, which can obtain high-resolution radar images under low visibility weather conditions, and it is widely used in the field of ship monitoring (Yang ...

doi:10.3389/fnbot.2024.1293992 pmid:38298467 pmcid:PMC10824852 fatcat:dfmlgmtatndkpczdspbmaw2v6e

DOAJ

The whole structure can realise the fast detection and precise positioning of ships with an arbitrary orientation. Finally, the authors conduct detailed experiments on the SAR ship image dataset. ... In order to realise the fast detection of ships in synthetic aperture radar (SAR) images, a detection method combining visual saliency and a cascade convolutional neural network (CNN) is proposed. ... In future research, the generative adversarial networks will be used to overcome the lack of training data, and the cascade networks will be optimised to further improve the detection accuracy and speed ...

doi:10.1049/iet-rsn.2020.0113 fatcat:2vnykuzwznhp3itxwzfvcnp3ra

DOAJ

The research status of SAR image target detection based on CNN is summarized and compared in detail with traditional algorithms. ... Aiming at the poor robustness and low detection accuracy of traditional detection algorithms, SAR image target detection based on the Convolutional Neural Network (CNN) is reviewed in this paper. ... The author would also like to thank the anonymous reviewers for their helpful suggestions and comments to improve the article. ...

doi:10.3390/rs14246240 fatcat:lpmezu3isbhtbbq2yz46mbkhze

DOAJ Szczepanski

They have been widely used in various applications such as optical and SAR image scene classification, object detection and recognition, semantic segmentation, and change detection. ... We perform a detailed analysis of several representative adversarial attacks on SAR image recognition models and provide an example of adversarial robustness evaluation. ... Adversarial examples for CNN-based SAR image classification: An experience study[J]. ...

doi:10.12000/jr21048 doaj:25eedd36124c4f2cb5656f7119f1cfa4 fatcat:3ib3m5cjxfeypbjow5223qxghi

DOAJ

Finally, an in-depth discussion on limitations, open challenges, considerations of oil spill classification systems using remote sensing, and state-of-the-art ML algorithms are highlighted along with conclusions ... Necessary preprocessing and preparation of data for developing classification models are then highlighted. ... generative adversarial network (GAN). ...

doi:10.3390/rs12203338 fatcat:awufdmqg4bhgpi2cmsxy5b52pa

DOAJ Szczepanski

The public SAR ship dataset (SSDD) and China Gaofen-3 satellite SAR image are used to validate the proposed method. ... Convolution neural network (CNN)-based detectors have shown great performance on ship detections of synthetic aperture radar (SAR) images. ... Recently, a new Perceptual Generative Adversarial Net-work (Perceptual GAN) model was proposed to improve the detection of small objects through narrowing the representation differences of the small objects ...

doi:10.3390/s20092547 pmid:32365747 fatcat:jpmwb6upp5a6rb72jncvkm4sgu

DOAJ

Convolutional neural networks (CNNs) and transformers have boosted the rapid growth of object detection in synthetic aperture radar (SAR) images. ... arbitrary-oriented SAR ship detection. ... In comparison to existing object detection methods, our method exhibits improvements over both general detectors and SAR ship detectors. ...

doi:10.3390/rs16030483 fatcat:7gvm45c22za5zdymx56sbpmrke

DOAJ Szczepanski

., +, JSTARS 2020 5264-5271 Impact of Satellite Sounding Data on Virtual Visible Imagery Generation Using Conditional Generative Adversarial Network. ... ., +, JSTARS 2020 3701-3710 Impact of Satellite Sounding Data on Virtual Visible Imagery Generation Using Conditional Generative Adversarial Network. ...

doi:10.1109/jstars.2021.3050695 fatcat:ycd5qt66xrgqfewcr6ygsqcl2y

DOAJ

With this effort, we hope to stimulate more research in this interesting yet under-exploited research field and to pave the way for use of deep learning in big SAR data processing workflows. ... of deep learning applied to SAR in depth, summarize available benchmarks, and recommend some important future research directions. ... [173] first employed a conditional generative adversarial network (cGAN) to generate artificial SAR-like images from optical images, then matched them with real SAR images. ...

arXiv:2006.10027v2 fatcat:s3tiroz4qve6nbhavtz77fbis4

Multiple Versions

Data Augmentation via Mixed Class Interpolation using Cycle-Consistent Generative Adversarial Networks Applied to Cross-Domain Imagery [article]

Preserved Fulltext

Adversarial Self-Supervised Learning for Robust SAR Target Recognition

Preserved Fulltext

A Comprehensive Survey of Machine Learning Applied to Radar Signal Processing [article]

Preserved Fulltext

Image Enhancement Driven by Object Characteristics and Dense Feature Reuse Network for Ship Target Detection in Remote Sensing Imagery

Preserved Fulltext

Deep-Learning for Radar: A Survey

Preserved Fulltext

Benchmarking Deep Learning Classifiers for SAR Automatic Target Recognition [article]

Preserved Fulltext

Context-aware SAR image ship detection and recognition network

Preserved Fulltext

Fast ship detection combining visual saliency and a cascade CNN in SAR images

Preserved Fulltext

A Survey of SAR Image Target Detection Based on Convolutional Neural Networks

Preserved Fulltext

Adversarial Robustness of Deep Convolutional Neural Network-based Image Recognition Models: A Review

Preserved Fulltext

Sensors, Features, and Machine Learning for Oil Spill Detection and Monitoring: A Review

Preserved Fulltext

A Novel Detector Based on Convolution Neural Networks for Multiscale SAR Ship Detection in Complex Background

Preserved Fulltext

LPST-Det: Local-Perception-Enhanced Swin Transformer for SAR Ship Detection

Preserved Fulltext

2020 Index IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing Vol. 13

Preserved Fulltext

Deep Learning Meets SAR [article]

Preserved Fulltext