English

$\ell_1$DecNet+: A new architecture framework by $\ell_1$ decomposition and iteration unfolding for sparse feature segmentation

Image and Video Processing 2025-07-01 v2 Computer Vision and Pattern Recognition Optimization and Control

Abstract

1\ell_1 based sparse regularization plays a central role in compressive sensing and image processing. In this paper, we propose 1\ell_1DecNet, as an unfolded network derived from a variational decomposition model incorporating 1\ell_1 related sparse regularization and solved by scaled alternating direction method of multipliers (ADMM). 1\ell_1DecNet effectively decomposes an input image into a sparse feature and a learned dense feature, and thus helps the subsequent sparse feature related operations. Based on this, we develop 1\ell_1DecNet+, a learnable architecture framework consisting of our 1\ell_1DecNet and a segmentation module which operates over extracted sparse features instead of original images. This architecture combines well the benefits of mathematical modeling and data-driven approaches. To our best knowledge, this is the first study to incorporate mathematical image prior into feature extraction in segmentation network structures. Moreover, our 1\ell_1DecNet+ framework can be easily extended to 3D case. We evaluate the effectiveness of 1\ell_1DecNet+ on two commonly encountered sparse segmentation tasks: retinal vessel segmentation in medical image processing and pavement crack detection in industrial abnormality identification. Experimental results on different datasets demonstrate that, our 1\ell_1DecNet+ architecture with various lightweight segmentation modules can achieve equal or better performance than their enlarged versions respectively. This leads to especially practical advantages on resource-limited devices.

Keywords

Cite

@article{arxiv.2203.02690,
  title  = {$\ell_1$DecNet+: A new architecture framework by $\ell_1$ decomposition and iteration unfolding for sparse feature segmentation},
  author = {Yumeng Ren and Yiming Gao and Chunlin Wu and Xue-cheng Tai},
  journal= {arXiv preprint arXiv:2203.02690},
  year   = {2025}
}

Comments

12 pages, 7 figures