We present a new color image segmentation method that combined texture measures and the JSEG (J measure based JSEGmentation) algorithm. In particular, two major contributions are set forth in this paper. (1) The two measures defined in JSEG and the Laws texture energy is discussed respectively and then we find that the over-segmentation problem of JSEG could be attributed partly to the absence of color discontinuity in the J measure. (2) A new measure is proposed by integrating the Laws texture energy measures into the J measure, on which our segmentation method is based. The new segmentation method taking account of both textural homogeneity and color discontinuity in local regions can be used to detect proper edges at the boundaries of shadows and highlights. Performance improvement due to the proposed modification was demonstrated on a variety of real color images.
To solve the problem of interposition between trees or trees and between trees and ground in forest images which would bring on error-matching and be unable to construct a full 3D-network, a new approach for segmentation of forest images was proposed. The proposed color divergence was defined over the index class map by quantized image, which was a good indicator of whether that area was in region center or near region boundaries. Using this measure, image texture was analyzed by multi-resolution. Then the initial over-segmented regions were merged according to Laws texture energy measure. Experimental demonstrated that the segmentation results of forest images on the proposed approach hold favorable consistency in terms of human perception. The classification accuracy was 80%. The recognized trees and ground can offer dependable data for image matching and 3D modeling.