基于边缘定向扩散的图像增强方法

针对前向后向扩散方程不能较好的保持流线状结构，而相干增强不能锐化边缘的缺点, 提出一种新的基于边缘定向的张量型前向后向扩散模型.该模型将前向-后向扩散方程引进到张量型扩散方程中，在扩散系数的选取上合并了相干增强扩散与前向-后向扩散的长处，又克服了他们各自的缺点.采用类似相干增强扩散的边缘定向算子实现对边缘的定向，然后根据边缘定向的结果设置扩散张量的特征根，使扩散张量沿边缘方向为正向扩散以增强边缘，而垂直于边缘方向为逆扩散以锐化边缘.理论分析和数值计算表明，该方法具有比相干增强扩散及前向-后向扩散更好的增强效果.

Image Enhancement Based on Edge-directed Diffusion

Since forward and backward diffusion cannot hold flow-like edges, and coherence enhancement diffusion cannot sharpen edges, a new edge-directed nonlinear diffusion equation to enhance image is put forward. The new model introduces forward and backward diffusion to the tensor diffusion, and gets a new diffusivity, which combines the merits of these two methods. The new model adapts an operator like coherence enhancement diffusion for edge orientation, and defines eigenvalues based on edgers orientation such that the new tensor has positive diffusion coefficient in edge direction and negative diffusion coefficient in gradient direction. Both theory analysis and numerical results show that the new model has better enhancement result than coherence enhancing diffusion and forward and backward diffusion.