A hybrid EMPCA-Scott approach for estimating probability distributions in mutual information

Multi-modal Brain MR Image Registration using A Novel Local Binary Descriptor based on Statistical Approach

“…δ denotes the standard deviation of the intensity values of all pixels within the patch, as defined by (10):…”

“…Specifically, the 𝑆𝐿𝐵𝐷 is used to represent the binary pattern of the original fixed image and the transformed moving images. The dissimilarity value between 𝑆𝐿𝐵𝐷 of 𝐼 𝑓 and 𝑆𝐿𝐵𝐷 of 𝑇 𝐼 𝑚 ; 𝜇 is calculated by Α �𝐼 𝑓 ,𝑇 𝐼 𝑚 ; 𝜇� as defined by (10). To find the optimal transformations, 𝑆𝐿𝐵𝐷 minimizes the dissimilarity values with Α �𝐼 𝑓 ,𝑇 𝐼 𝑚 ; 𝜇� .…”

“…Particularly, it has been extensively and successfully used to measure the intensity relationships in image processing [6] [7]. During the last decade, many researchers have enhanced the accuracy of image registration based on MI, such as Tsallis and Renyi's entropies [8], Jensen-Renyi's entropy [9], hybrid EMPCA-Scott approach [10], and self-similarity α-MI (SeSaMI) [11]. However, the notion of MI alone still has a well-known drawback, i.e., it ignores spatial information [12].…”