Detection of Structural Similarity for Multimodal Microscopic Image Registration

In this paper we propose a novel method to detect the structural similarity in registering color and confocal microscopic images. Our prior work presented the basic idea of detecting the structural similarity of such images, which utilizes the intensity relationships among red-green-blue color channels. The work in this paper will make the detection of structural similarity automatic and adaptive to each individual color microscopic image. The experimental results will demonstrate the effectiveness of the proposed method in detecting the structural similarity of these images and significant improvements in the registration performance.