TEXTURE SEGMENTATION USING 2-D GABOR ELEMENTARY-FUNCTIONS

Many texture-segmentation schemes use an elaborate bank of filters to decompose a textured image into a joint space/spatial-frequency repres entation. Although these schemes show promise, and although some analy tical work work has been done, the relationship between texture differ ences and the filter configurations required to distinguish them remai n largely unknown. This paper examines the issue of designing individu al filters. Using a 2-D texture model, we show analytically that apply ing a properly configured bandpass filter to a textured image produces distinct output discontinuities at texture boundaries; the analysis i s based on Gabor elementary functions, but it is the bandpass nature o f the filter that is essential. Depending on the type of texture diffe rence, these discontinuities form one of four characteristic signature s: a step, ridge, valley, or a step change in average local output var iation. Accompanying experimental evidence indicates that these signat ures are useful for segmenting an image. The analysis indicates those texture characteristics that are responsible for each signature type. Detailed criteria are provided for designing filters that can produce quality output signatures. We also illustrate occasions when asymmetri c filters are beneficial, an issue not previously addressed.