Comparative study for image registration techniques of remote sensing images

Abstract

Image registration determines the relative orientation between two images. As there are different techniques for image registration, it is important to compare these techniques to identify the advantages and disadvantages of each one. In this paper, a comparison between a fast Fourier transform (FFT)-based technique, a contour-based technique, a wavelet-based technique, a Harris–Pulse Coupled Neural Network (PCNN)-based technique and Harris–Moment-based technique is presented. The algorithms were tested on Landsat Thematic Mapper (TM) and SPOT remote sensing images and its performance were compared using the Root Mean Square Error (RMSE). It has been concluded that the order of techniques with less RMSE is the PCNN, the moment, the contour, the wavelet and the FFT-based techniques, respectively. Whereas the order of techniques with the less running time is the contour, the wavelet, the moment, the FFT and the PCNN-based techniques, respectively. And finally the technique that detects the more control points in both images is the wavelet.