K-Coherence Terrain Texture Upscaling Using Gabor Wavelets

Abstract

Terrain texture is ubiquitous visual information that is widely used in geo-experience. It can describe a variety of surface characteristics such as terrain, plants, minerals, fur and skin and it covers the spectrum of textures from completely regular to near-regular to irregular. A near-regular texture deviates geometrically and photo-metrically from a regular congruent tiling. They can be readily observed in man-made and nature environments: buildings, wallpapers, floors, tiles, windows, fabric and honeycomb, animal fur, feathers and so on. This paper focuses on a class of near-regular textures (NRT) that are regular textures under locally smooth geometric and appearance deformations. In this paper, Gabor wavelet filter is introduced into K-Coherence calculation. In this method, the statistics of filtered wavelet are regarded as feature vectors, and the difference in two points is measured by the Euclidean Distance between feature vectors. In this way, the matching process of K-Coherence will be more accurate, which will further result in the improvement on terrain texture synthesis to some degree.