基于光照曝光联合建模的物理引导低光图像增强方法
By using joint modeling of illumination and exposure and a dual-condition adaptive enhancement module, the problems of local overexposure and insufficient brightening of dark areas in low-light image enhancement are solved, achieving uniform restoration of image brightness and color consistency, which conforms to the physiological perception habits of the human eye.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- COLLEGE OF SCI & TECH NINGBO UNIV
- Filing Date
- 2026-04-30
- Publication Date
- 2026-07-17
AI Technical Summary
Existing low-light image enhancement techniques sever the physical connection between local illumination and global exposure, resulting in overexposure in local areas or insufficient brightening of dark areas. Furthermore, the lack of spatial prior guidance leads to color distortion and structural blurring, which fails to conform to the physiological perception habits of the human eye.
By jointly modeling illumination and exposure, a local illumination map is generated and the global exposure factor is derived serially. The dual-condition adaptive enhancement module is combined to perform feature modulation and spatial guidance, and an overall loss function including exposure consistency loss is constructed for network optimization.
It achieves uniform restoration of brightness across the entire image, suppresses noise amplification, maintains consistency in image color and structure, and outputs an enhanced image that conforms to the visual brightness distribution of the human eye.
Smart Images
Figure CN122115254B_ABST