An investigation towards wavelet based optimization of automatic image registration techniques

TL;DR

This paper explores how wavelet transforms can optimize automatic image registration techniques, particularly enhancing SIFT-based methods for better accuracy in high-resolution remote sensing applications.

Contribution

It investigates the integration of wavelet transforms with existing image registration techniques, demonstrating improved performance in feature extraction and registration accuracy.

Findings

Wavelet-enhanced SIFT yields more accurate key points.

Improved registration performance with high-resolution images.

Enhanced feature detection reduces control point requirements.

Abstract

Image registration is the process of transforming different sets of data into one coordinate system and is required for various remote sensing applications like change detection, image fusion, and other related areas. The effect of increased relief displacement, requirement of more control points, and increased data volume are the challenges associated with the registration of high resolution image data. The objective of this research work is to study the most efficient techniques and to investigate the extent of improvement achievable by enhancing them with Wavelet transform. The SIFT feature based method uses the Eigen value for extracting thousands of key points based on scale invariant features and these feature points when further enhanced by the wavelet transform yields the best results.