Morphologic weak edge detection by multi-structure and multi-scale in anterior chamber OCT images

In this paper, a new algorithm of weak edge detection is proposed, it is based on the traditional Morphological edge detection algorithm, which has been combined with evidence syncretic fusion algorithm and information entropy. The images are expanded by multi-structure and multi-scale to get more edge images, the weighted coefficients are calculated by the differences between two information entropies of edge images, finally, the edge images are fused by the weighted coefficients. The proposed algorithm can adapt to the features of anterior chamber OCT images very well, it not only has the stronger ability of anti-noise, but also can detect more clearly edge images.