Locally adaptive template sizes for matching repeat images of mass movements

This paper presents an algorithm for locally adaptive template sizes in normalized cross-correlation (NCC) based image matching for measuring surface displacement of mass movements. After adaptively identifying candidate templates based on the image signal-to-noise ratio (SNR), the algorithm iteratively looks for the size at which the maximum cross-correlation coefficient attains a local peak and the matching position gets fixed. The algorithm is tested on Landsat7 ETM+ and radar intensity (Radarsat-2) image pairs of glacier flow. It is evaluated in comparison with globally (image-wide) fixed template sizes ranging from 11 to 101 pixels based on the performance of the matching. The adaptive algorithm results in more reliable displacement estimates as it matches the pixels more accurately.