一种基于多模态特征融合的恶劣环境图像增强与复原方法
By employing weighted least squares optimization through degradation perception and structural adaptive gradient fusion, the structural distortion and detail loss problems of multimodal image fusion under harsh environments are solved, thereby improving image clarity and structural fidelity. This approach is applicable to fields such as military reconnaissance, security monitoring, remote sensing, medical imaging, and autonomous driving.
CN122115236BActive Publication Date: 2026-07-17CHENGDU FANCHEN TECH CO LTD
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- CHENGDU FANCHEN TECH CO LTD
- Filing Date
- 2026-04-27
- Publication Date
- 2026-07-17
Smart Images
Figure CN122115236B_ABST
Abstract
本发明提供一种基于多模态特征融合的恶劣环境图像增强与复原方法，属于图像增强处理领域；该方法首先获取多模态图像并计算各模态退化程度图；基于退化程度与梯度幅值构建融合权重，并通过结构张量分析得到局部主导方向与相干性；根据相干性将像素分区，分别采用硬投影、各向同性融合或软投影策略对各模态梯度进行自适应处理；基于融合权重对处理后的梯度加权融合，并结合增强因子g(x)得到融合梯度场；最后构建包含像素保真项与梯度保真项的能量函数，通过加权最小二乘优化重建增强融合图像；本发明通过退化感知机制抑制模糊模态贡献，通过结构自适应投影解决梯度方向冲突，显著提升了恶劣环境下多模态图像融合的清晰度与结构保真度。
Need to check novelty before this filing date? Find Prior Art