Low-light Image Enhancement via Multi-scale Attention combined with Fourier Transform
2026-07-27 • Computer Vision and Pattern Recognition
Computer Vision and Pattern Recognition
AI summaryⓘ
The authors developed a new deep learning method called MSFT to improve pictures taken in very dark places. Their approach uses a special way to look at images in frequency form and focuses attention on different image parts at multiple scales. They combine information about brightness and details to make textures clearer and enhance fine features. Tests on several datasets show their method works better than previous ones, giving clearer and more accurate results in low-light images.
low-light image enhancementFourier transformmulti-scale attentionU-shaped architecturefrequency domaintexture restorationsupervised deep learningpeak signal-to-noise ratiostructural similarity indeximage decoding
Authors
Wenbin Du, Jian Long, Zhu Cao
Abstract
Low-light image enhancement (LLIE) aims to improve image quality and clarity in diverse and demanding low-illumination environments. However, existing deep learning-based LLIE methods struggle to accurately capture real-world illumination and restore texture details, largely because their algorithmic strengths remain underutilized. To address these issues, we present a supervised frequency domain deep learning network for LLIE, named multi-scale attention combined with the Fourier transform (MSFT) which adopts a U-shaped, one-stage architecture that infuses guidance from low-light images into the network by channeling it through multi-scale attention. We further fuse the amplitude information from priori channels with that of the low-light image in MSFT's self-created module, and carry out multi-scale guidance along with the network. Subsequently, to better enhance the faint feature, such as fine content and textures, and to better fuse global context confidence in the decoding stage, we separately introduce a multi-shape synergistic attention and a lightweight network that effectively integrate information in high-dimensional space to embed into the superlative feature space channel containing rich texture information. Extensive experiments conducted on LOL, SID, SMID, and SDSD datasets demonstrate that MSFT significantly outperforms state-of-the-art competitors. For example, compared with Retinexformer, our method achieves a peak signal-to-noise ratio of up to 41.76 decibels on the SDSD-outdoor dataset with an increase of 11.92 decibels and a structural similarity index of 0.988 with a 13.80% improvement.