Litcius/Paper detail

Sequential ISAR Target Classification Based on Hybrid Transformer

Ruihang Xue, Xueru Bai, Xiangyong Cao, Feng Zhou

2022IEEE Transactions on Geoscience and Remote Sensing22 citationsDOI

Abstract

To make full use of the sequential information obtained by continuous inverse synthetic aperture radar (ISAR) imaging, this article proposes a sequential ISAR target classification network based on hybrid transformer (HT). First, a temporal–spatial encoder based on the attention mechanism is designed to extract long-term and global features from sequential images. Meanwhile, a local feature encoder based on the 3-D convolution neural network is designed to extract short-term and local features. Then, the above two features are fused and the classification labels are obtained by a channel encoder–decoder. In 4-satellite target classification experiments, the proposed HT shows high accuracy and robustness to the unknown image scaling, rotation, and combined deformations.

Topics & Concepts

Computer scienceInverse synthetic aperture radarArtificial intelligenceSynthetic aperture radarRobustness (evolution)EncoderPattern recognition (psychology)Feature extractionConvolutional neural networkComputer visionRadar imagingRadarTelecommunicationsOperating systemGeneChemistryBiochemistryAdvanced SAR Imaging TechniquesGeophysical Methods and ApplicationsSynthetic Aperture Radar (SAR) Applications and Techniques