基于双分支解耦网络及多模态感知的低光照图像增强方法
By employing a dual-branch decoupled network and a multimodal perception approach, the problems of illumination optimization conflict and high-resolution processing difficulties in low-light image enhancement are resolved. This achieves efficient illumination restoration and detail reconstruction under low computational load, thereby improving the visual quality and detail clarity of the image.
CN121961894BActive Publication Date: 2026-07-17JINAN UNIVERSITY
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- JINAN UNIVERSITY
- Filing Date
- 2026-02-10
- Publication Date
- 2026-07-17
Smart Images
Figure CN121961894B_ABST
Abstract
本发明公开了基于双分支解耦网络及多模态感知的低光照图像增强方法，属于计算机视觉和图像处理技术领域。该方法构建了以通道自注意力恢复网络为主干的双分支解耦架构，包含高亮度注意力分支和低亮度注意力分支，分别针对图像中的亮部与暗部区域进行差异化恢复；为实现精准解耦，设计了亮度损失注意力算法，利用自适应掩膜策略引导各分支专注于特定照度区域的特征提取；同时结合CLIP多模态损失，借助文本语义信息引导生成图像向高质量视觉风格逼近；最终通过RCAB超分模块对双分支特征进行深度融合与上采样，输出高分辨率增强图像。本发明解决了低光照图像增强中全局处理导致的光照优化冲突以及高分辨率图像处理算力消耗过大的技术问题。
Need to check novelty before this filing date? Find Prior Art