Laser & Optoelectronics Progress, Volume. 62, Issue 14, 1439002(2025)

Multiscale Adversarial-Based Reconstruction Method for Occluded Polarized Images

Han Han1,2, Xin Wang1,2、*, Xiankun Pu3, Peifeng Pan1,2, Yao Zha1,2, Yajun Xu1,2, and Jun Gao1,2
Author Affiliations
  • 1School of Computer Science and Information Engineering, Hefei University of Technology, Hefei 230009, Anhui , China
  • 2Laboratory of Image Information Processing, Hefei University of Technology, Hefei 230009, Anhui , China
  • 3School of Automotive and Traffic Engineering, Hubei University of Arts and Science, Xiangyang441053, Hubei , China
  • show less

    To address the challenges of restoring details in heavily occluded areas and enhancing network generalization capabilities in occluded polarized image reconstruction tasks, the research proposes a novel occluded image reconstruction model, PolarReconGAN, based on multiscale adversarial network. The proposed model integrates with polarization array imaging technology, aims to reconstruct the polarization information of occluded targets, thereby improving image quality and detail representation. We design a multiscale feature extraction module that employs a random window slicing method to prevent information loss due to image resizing, and utilizes data augmentation to enhance model generalization. Additionally, a loss function based on discrete wavelet transform is employed to further improve the reconstruction effects of image details. The experimental results demonstrate that the proposed method achieves an average structural similarity index (SSIM) of 0.7720 and an average peak signal-to-noise ratio (PSNR) of 25.2494 dB on a multi-view occluded polarization image dataset, indicating superior performance in occluded image reconstruction.

    Keywords
    Tools

    Get Citation

    Copy Citation Text

    Han Han, Xin Wang, Xiankun Pu, Peifeng Pan, Yao Zha, Yajun Xu, Jun Gao. Multiscale Adversarial-Based Reconstruction Method for Occluded Polarized Images[J]. Laser & Optoelectronics Progress, 2025, 62(14): 1439002

    Download Citation

    EndNote(RIS)BibTexPlain Text
    Save article for my favorites
    Paper Information

    Category: AI for Optics

    Received: Jan. 16, 2025

    Accepted: Mar. 2, 2025

    Published Online: Jul. 2, 2025

    The Author Email: Xin Wang (wangxin@.hfut.edu.cn)

    DOI:10.3788/LOP250525

    CSTR:32186.14.LOP250525

    Topics