An Information-Fusion Edge Preserving Method in Image Filtering

To improve the edge-preserving capability for the noisy image in image filtering, this paper proposed an edge-preserving method based on information fusion. This method firstly defined information sources based on region fuzzy entropy and directional structure feature, and then implemented the triangle module fusion operator for edge detection. Finally it operated a modified hybrid filter operation to preserve image edges in image filtering. Experimental results illustrate that, compared with the reference method; the proposed method can achieve better performance in terms of PSNR, average gradient and cross entropy with better edge detection and preservation.