Work

Frequency Domain Information Integrated Network for Low-Light Image Enhancement

Year: 2025

Type: conference-paper

Abstract: Low-light images often suffer from significant noise and detail loss, making it challenging to effectively distinguish signals from noise when processed directly in the spatial domain. To this end, we introduce frequency domain information to better distinguish high-frequency details from low-frequency components, thereby suppressing noise while enhancing details. The proposed method converts sRGB images into raw-RGB images by reversing the Image Signal Processor (ISP) pipeline to avoid unnecessary effects. A frequency information interaction processing unit is then designed to enhance low-frequency textures using high-frequency information and correct high-frequency data through low-frequency information to reduce structural deformation during restoration. The signal-to-noise ratio prior is introduced in the hidden layer to further remove noise. Experimental results show that the proposed method outperforms existing color and structure enhancement methods.

Source: IEEE International Conference on Acoustics Speech and Signal Processing

DOI: 10.1109/icassp49660.2025.10890113

Authors: Na Li, Xi Luo, Dunlu Peng, Zied Bouraoui

Institutions: University of Shanghai for Science and Technology, Centre National de la Recherche Scientifique, Centre de Recherche en Informatique de Lens

Language: English

Topic: Image Enhancement Techniques

Subfield: Computer Vision and Pattern Recognition

Field: Computer Science

Domain: Physical Sciences

Open Access status: green

Funder: Agence Nationale de la Recherche

Award: Event-Centric Reasoning for Interpreting Everyday Narratives