Scattering mechanism-guided zero-shot PolSAR target recognition

TY - JOUR

T1 - Scattering mechanism-guided zero-shot PolSAR target recognition

AU - Li, Feng

AU - Yang, Xiaojing

AU - Zhang, Liang

AU - Wang, Yanhua

AU - Han, Yuqi

AU - Zhang, Xin

AU - Li, Yang

N1 - Publisher Copyright: © 2024 International Society for Photogrammetry and Remote Sensing, Inc. (ISPRS)

PY - 2025/2

Y1 - 2025/2

N2 - In response to the challenges posed by the difficulty in obtaining polarimetric synthetic aperture radar (PolSAR) data for certain specific categories of targets, we present a zero-shot target recognition method for PolSAR images. Based on a generative model, the method leverages the unique characteristics of polarimetric SAR images and incorporates two key modules: the scattering characteristics-guided semantic embedding generation module (SE) and the polarization characteristics-guided distributional correction module (DC). The former ensures the stability of synthetic features for unseen classes by controlling scattering characteristics. At the same time, the latter enhances the quality of synthetic features by utilizing polarimetric features, thereby improving the accuracy of zero-shot recognition. The proposed method is evaluated on the GOTCHA dataset to assess its performance in recognizing unseen classes. The experiment results demonstrate that the proposed method achieves SOTA performance in zero-shot PolSAR target recognition (e.g., improving the recognition accuracy of unseen categories by nearly 20%). Our codes are available at https://github.com/chuyihuan/Zero-shot-PolSAR-target-recognition.

AB - In response to the challenges posed by the difficulty in obtaining polarimetric synthetic aperture radar (PolSAR) data for certain specific categories of targets, we present a zero-shot target recognition method for PolSAR images. Based on a generative model, the method leverages the unique characteristics of polarimetric SAR images and incorporates two key modules: the scattering characteristics-guided semantic embedding generation module (SE) and the polarization characteristics-guided distributional correction module (DC). The former ensures the stability of synthetic features for unseen classes by controlling scattering characteristics. At the same time, the latter enhances the quality of synthetic features by utilizing polarimetric features, thereby improving the accuracy of zero-shot recognition. The proposed method is evaluated on the GOTCHA dataset to assess its performance in recognizing unseen classes. The experiment results demonstrate that the proposed method achieves SOTA performance in zero-shot PolSAR target recognition (e.g., improving the recognition accuracy of unseen categories by nearly 20%). Our codes are available at https://github.com/chuyihuan/Zero-shot-PolSAR-target-recognition.

KW - Distributional correction

KW - PolSAR target recognition

KW - Scattering center

KW - Zero-shot learning

UR - https://www.scopus.com/pages/publications/85213867048

U2 - 10.1016/j.isprsjprs.2024.12.022

DO - 10.1016/j.isprsjprs.2024.12.022

M3 - Article

AN - SCOPUS:85213867048

SN - 0924-2716

VL - 220

SP - 428

EP - 439

JO - ISPRS Journal of Photogrammetry and Remote Sensing

JF - ISPRS Journal of Photogrammetry and Remote Sensing

ER -