Weakly Contrastive Learning via Batch Instance Discrimination and Feature Clustering for Small Sample SAR ATR

Zhai, Yikui; Zhou, Wenlve; Sun, Bing; Li, Jingwen; Ke, Qirui; Ying, Zilu; Gan, Junying; Mai, Chaoyun; Labati, Ruggero Donida; Piuri, Vincenzo; Scotti, Fabio

doi:10.1109/tgrs.2021.3066195

Cited by 11 publications

(18 citation statements)

References 41 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…IGAN achieves semi-supervised generation and recognition simultaneously [51]. DNN2(PoseSy) means the recognition performance with the augmentation of pose synthesis, Multiscale [52] employs randomly rotating and flipping, Weakly [53] employs randomly rotating in the recognition process.…”

Section: Comparison With Other Augmentation Methodsmentioning

confidence: 99%

SAR Target Image Generation Method Using Azimuth-Controllable Generative Adversarial Network

Wang

Pei

Liu

et al. 2022

IEEE J. Sel. Top. Appl. Earth Observations Remote Sensing

View full text Add to dashboard Cite

Sufficient synthetic aperture radar (SAR) target images are very important for the development of researches. However, available SAR target images are often limited in practice, which hinders the progress of SAR application. In this paper, we propose an azimuth-controllable generative adversarial network to generate precise SAR target images with an intermediate azimuth between two given SAR images' azimuths. This network mainly contains three parts: generator, discriminator, and predictor. Through the proposed specific network structure, the generator can extract and fuse the optimal target features from two input SAR target images to generate SAR target image. Then a similarity discriminator and an azimuth predictor are designed. The similarity discriminator can differentiate the generated SAR target images from the real SAR images to ensure the accuracy of the generated, while the azimuth predictor measures the difference of azimuth between the generated and the desired to ensure the azimuth controllability of the generated. Therefore, the proposed network can generate precise SAR images, and their azimuths can be controlled well by the inputs of the deep network, which can generate the target images in different azimuths to solve the small sample problem to some degree and benefit the researches of SAR images. Extensive experimental results show the superiority of the proposed method in azimuth controllability and accuracy of SAR target image generation.

show abstract

Section: Comparison With Other Augmentation Methodsmentioning

confidence: 99%

SAR Target Image Generation Method Using Azimuth-Controllable Generative Adversarial Network

Wang

Pei

Liu

et al. 2022

IEEE J. Sel. Top. Appl. Earth Observations Remote Sensing

View full text Add to dashboard Cite

show abstract

“…This aimed to solve the problem of overfitting in SAR ATR by combining strong supervision and weak supervision [25]. Based on the instance-level contrastive loss, Zhai et al proposed batch instance discrimination and feature clustering (BIDFC), which can adjust the embedding distance in the feature space [26]. A contrastive domain adaption methodology was used by Bi et al to reduce the disparity in distribution between a simulated and actual sparse SAR dataset [27].…”

Section: Related Workmentioning

confidence: 99%

Locality Preserving Property Constrained Contrastive Learning for Object Classification in SAR Imagery

et al. 2023

View full text Add to dashboard Cite

Robust unsupervised feature learning is a critical yet tough task for synthetic aperture radar (SAR) automatic target recognition (ATR) with limited labeled data. The developing contrastive self-supervised learning (CSL) method, which learns informative representations by solving an instance discrimination task, provides a novel method for learning discriminative features from unlabeled SAR images. However, the instance-level contrastive loss can magnify the differences between samples belonging to the same class in the latent feature space. Therefore, CSL can dispel these targets from the same class and affect the downstream classification tasks. In order to address this problem, this paper proposes a novel framework called locality preserving property constrained contrastive learning (LPPCL), which not only learns informative representations of data but also preserves the local similarity property in the latent feature space. In LPPCL, the traditional InfoNCE loss of the CSL models is reformulated in a cross-entropy form where the local similarity of the original data is embedded as pseudo labels. Furthermore, the traditional two-branch CSL architecture is extended to a multi-branch structure, improving the robustness of models trained with limited batch sizes and samples. Finally, the self-attentive pooling module is used to replace the global average pooling layer that is commonly used in most of the standard encoders, which provides an adaptive method for retaining information that benefits downstream tasks during the pooling procedure and significantly improves the performance of the model. Validation and ablation experiments using MSTAR datasets found that the proposed framework outperformed the classic CSL method and achieved state-of-the-art (SOTA) results.

show abstract

“…For instance, contrastive learning has been used, for example, in the hyperspectral image (HSI) classification to solve the small-sample problem of HSIs. Meanwhile, it has been adopted in synthetic aperture radar image classification to overcome insufficient labeled data [45], [46]. However, few methods apply contrastive learning to remote sensing image segmentation.…”

Section: Contrastive Learningmentioning

confidence: 99%

BSSNet: Building Subclass Segmentation From Satellite Images Using Boundary Guidance and Contrastive Learning

Xie

Jiang

et al. 2022

IEEE J. Sel. Top. Appl. Earth Observations Remote Sensing

View full text Add to dashboard Cite

Building subclass segmentation, aimed at predicting classes of buildings (high-rise zone, low-rise zone, single highrise, and single low-rise) from satellite images, is beneficial in numerous applications, including human geography, urban planning, and humanitarian aid. However, problems such as complex scenes and similar characteristics of different building categories make it difficult for general models to balance the accuracy of localization and classification in building subclass segmentation. Therefore, this paper proposes a novel network for building subclass segmentation called building subclass segmentation network (BSSNet), which uses two subnetworks to divide and conquer the problem. The first network guides the building locations through binary building segmentation, called localization network. The spatial gradient fusion module in the localization network improves the binary segmentation result by supervising the spatial gradient map of prediction. The second network is a classification network, which predicts building subclasses. Intermediate features of the second network are optimized by contrastive learning loss to improve feature consistency. Finally, predictions of the two networks are combined to obtain the final result. The experimental results demonstrate that our BSSNet can perform significant improvements on the Hainan dataset we produced and the xBD dataset. In particular, the BSSNet achieves the best performance compared to current methods on the Hainan dataset.

show abstract

Weakly Contrastive Learning via Batch Instance Discrimination and Feature Clustering for Small Sample SAR ATR

Cited by 11 publications

References 41 publications

SAR Target Image Generation Method Using Azimuth-Controllable Generative Adversarial Network

SAR Target Image Generation Method Using Azimuth-Controllable Generative Adversarial Network

Locality Preserving Property Constrained Contrastive Learning for Object Classification in SAR Imagery

BSSNet: Building Subclass Segmentation From Satellite Images Using Boundary Guidance and Contrastive Learning

Contact Info

Product

Resources

About