跳到主要导航 跳到搜索 跳到主要内容

Hybrid Transformer-CNN with Two-Stage Training for SAR Despeckling

  • Zhenyu Guo
  • , Weidong Hu
  • , Shichao Zheng
  • , Yuanyuan Zhang
  • , Ming Zhou
  • , Qian Wang
  • , Ruizhe Zhang
  • Beijing Institute of Technology
  • China Aerospace Science and Technology Corporation
  • Jiaxing Glead Electronics Co., Ltd.

科研成果: 书/报告/会议事项章节会议稿件同行评审

摘要

Coherent speckle noise in Synthetic Aperture Radar (SAR) imagery severely degrades image fidelity, compromising the accuracy and reliability of subsequent advanced interpretation tasks such as target recognition, scene classification, and semantic segmentation. Convolutional neural networks (CNNs) face a significant domain gap due to their reliance on synthetic data, resulting in insufficient generalization capabilities in real-world scenarios. To address this challenge, we propose a novel denoising framework centered on a hybrid network combining hierarchical Transformers and CNNs. This framework bridges the domain gap through a two-stage 'pre-training-fine-tuning' paradigm and achieves end-to-end optimization via a spatiotemporalfrequency domain joint loss function. Experiments on synthetic and real SAR datasets demonstrate that our approach achieves superior performance across multiple key metrics and visual quality, significantly enhancing despeckling capabilities and detail fidelity.

源语言英语
主期刊名CISS 2025 - 6th China International SAR Symposium
出版商Institute of Electrical and Electronics Engineers Inc.
ISBN(电子版)9798319517609
DOI
出版状态已出版 - 2025
已对外发布
活动6th China International SAR Symposium, CISS 2025 - Yiwu, 中国
期限: 25 10月 202527 10月 2025

丛书

姓名CISS 2025 - 6th China International SAR Symposium

会议

会议6th China International SAR Symposium, CISS 2025
国家/地区中国
Yiwu
时期25/10/2527/10/25

学术指纹

探究 'Hybrid Transformer-CNN with Two-Stage Training for SAR Despeckling' 的科研主题。它们共同构成独一无二的学术指纹。

引用此