Multimodal Medical Image registration using Discrete Wavelet Transform and Gaussian Pyramids

TL;DR

This paper introduces a novel multimodal brain image registration method combining discrete wavelet transform and Gaussian pyramids, demonstrating improved registration quality over existing techniques.

Contribution

The paper presents a new multimodal image registration approach using DWT and Gaussian pyramids, outperforming methods that use only one of these techniques.

Findings

Proposed method achieves higher MI values.

Method shows better correlation coefficients.

Outperforms wavelet-only and pyramid-only methods.

Abstract

In this research paper, authors propose multimodal brain image registration using discrete wavelet transform(DWT) followed by Gaussian pyramids. The reference and target images are decomposed into their LL, LH, HL and LL DWT coefficients and then are processed for image registration using Gaussian pyramids. The image registration is also done using Gaussian pyramids only and wavelets transforms only for comparison. The quality of registration is measured by comparing the maximum MI values used by the three methods and also by comparing their correlation coefficients. Our proposed technique proves to show better results when compared with the other two methods.