A hierarchical receptive network oriented to target recognition in SAR images. (June 2022)

• Record Type: Journal Article
• Authors: Dong, Ganggang; Liu, Hongwei
• Is Part Of: Pattern recognition. Volume 126(2022)
• Publication Date: 2022-06
• Subjects: Target recognition -- Deep learning -- Knowledge -- Hierarchical receptive -- SAR Image; Pattern perception -- Periodicals; Perception des structures -- Périodiques; Patroonherkenning; 006.4
• DOI: 10.1016/j.patcog.2022.108558
• Languages: English
• ISSNs: 0031-3203

Abstract

Highlights: This paper develops a hierarchical receptive architecture for target recognition in SAR images. The proposed architecture is composed of the signal-wise receptive and the patch-wise receptive. The empirical knowledge and the learned representation are combined for target recognition. Multiple comparative studies are performed to demonstrate the advantage of proposed strategy. Abstract: The recent years have witnessed a resurgence on neural network. Many functional layers are stacked hierarchically to learn the high-level representations. Yet the large album of radar image with label information are scarce. The fitting power of deep architectures are therefore limited. Additionally, the coherent imaging mechanism inevitably produce many speckles. They are with the statistical specificity of multiplicative noise, and hence make the image interpretation difficult. To solve the problems, this paper presents a new hierarchical receptive neural network. A signal-wise receptive module is first built by a family of delicate convolutional filters, with which the empirical features and knowledge are encoded. The receptive features are further refined in a patch-wise receptive unit, where some convolutional blocks are configured sequentially. The refined representations are finally used to make the inference. Multiple comparative studies are performed to demonstrate the advantage of proposed strategy.