Image Segmentation of Industrial CT Based on Edge Information
Xing Li Wu · Jisuanji gongcheng · 2004
The image of ICT (industrial CT ) is not very satisfaction because of technology of CT. The smart image segmentation is the base of getting smart measurement data .This paper uses the dimension wavelet multi-resolution transform coefficients as the measure of split-and-merge image segmentation, uses four-tree structure as original segment,and sets up suitable valve to restrain noise and pick-up the pictrure edge. The experimentations prove the approach is rational than the routine algorithm in segmenting the image of shoe lumen by CT. It can segment the edge of shoe lumen exactly.