Medical image registration based on PCA and M_PSNR

Journal of Complexity in Health Sciences

Xinrui Bi¹  Jincai Chang¹  Jin Jin¹  Miaomiao Jiang²  Jianzhong Cui² 

[1] College of Sciences, North China University of Science and Technology, Tangshan, 063210, China;Department of Neurosurgery, Tangshan Gongren Hospital, Tangshan, 063000, China;

关键词: medical image registration;    pca neural network;    psnr;    mutual information value;   

DOI  :  10.21595/chs.2020.21266

来源: DOAJ

【 摘 要 】

Medical image registration is not only the premise of medical image fusion, but also the focus of medical image processing. It plays an important role in clinical diagnosis and treatment planning. Aiming at the problem of mutual information value in image registration, a PCA neural network and M_PSNR based method for rigid registration of reference and floating images was proposed. In this method, the centroid of the image needs to be figured out through the image matrix, and the rotation Angle and translation of the image registration can be calculated by PCA method to obtain the optimal rotation transformation and the initial registration value. Finally, M_PSNR method improved by PSNR was used for similarity test to increase the speed and efficiency of registration and simplify the calculation process.

【 授权许可】

Unknown