Image Registration Based on Fuzzy Similarity
Zhentai Lu, Minghui Zhang, Qianjin Feng, Pengcheng Shi, Wufan Chen · 2007
In this paper, a novel image attribute, fuzzy edge field (FEF) is defined and used to strengthen the similarity between images. As the feature of images, FEF could be extracted automatically without any manual interaction. A new registration algorithm based on fuzzy edge similarity (FES) was presented. Generalized fuzzy operator (GFO) is utilized to detect edge. Gaussian function is explored to expand the edge to the neighborhood region. A fuzzy similarity measure is chosen as the registration function. We evaluate the effectiveness of the EMP-MI approach by applying it to the simulated and real brain image data (CT, MR, PET, and SPECT). Experimental results indicate that the function is less sensitive to low sampling resolution and noise, do not contain incorrect global maxima that are sometimes found in the mutual information (MI) function, and interpolation-induced local minima can be reduced.