Registration of multimodal brain images based on optical flow
Mengchao Su, Congxuan Zhang, Zhen Chen, Shaofeng Jiang · 2017
An improved registration method of multimodal brain images combined optical flow and Speeded-Up Robust Features (SURF) is presented in this paper. Firstly, histogram specification is used to transform modal of images before registration, the similar modal of images are beneficial to the following image registration. Secondly, the feature points are extracted by SURF, and then the preliminary registered image could be obtained. Thirdly, with the obtained registered image, optical flow algorithm is used to achieve the accurate results. Finally, several typical multimodal brain images are employed to evaluate the performances of the proposed approach, Demons based method, SURF based method and Mutual Information (MI) based method. Experimental results show that the measure indexes of Mean Squared Error (MSE), Peak Signal to Noise Ratio (PSNR) and correlation coefficient (Rcc) of our method are more competitive compared with the other methods, which indicate the proposed approach has the higher accuracy and better robustness.