Color-Texture Image Segmentation by Combining Region and Photometric Invariant Edge Information

摘要

An improved approach for JSEG algorithm is proposed for unsuper-vised color-texture image segmentation. The region and photometric invariant edge information are combined. A novel measure for color-texture homogeneity is defined by weighting the textural homogeneity measure with photometric invariant edge measure. Based on the map whose pixel values are values of the new measure, region growing-merging algorithm used in JSEG is then employed to segment the image. Finally, experiments on a variety of real color images demonstrate performance improvement due to the proposed method.