Local Linear Reconstruction Based Medical Image Registration in DWT Domain

Non rigid image registration assumes a paramount part over medical imaging framework. In the complicated conditions, it is quiet a testing errand. Sometimes the dataset is perverted by spatially varying intensity contortion. To resolve this, issue many techniques based on LLR have been designed. In this paper, the core idea behind the proposed technique is to reduce the reconstruction time by considering the LLR in transform domain. The proposed technique uses DWT to split the dataset image into sub bands. Low frequency sub band image is reconstructed by reference image, followed by combining all subband images to generate an enhanced image by using inverse DWT. The quantitative (peak signal-to-noise ratio, root mean square error, mean square error and normalized cross correlation) and visual results show the superiority of the proposed technique over the conventional LLR technique and the execution time is reduced by 19%.