基于边缘信息的工业CT图像分割法  

Image Segmentation of Industrial CT Based on Edge Information

作　　者：邱钊[1] 朱庆生[1] 卢霞[1] 伍星[1] 

机构地区：[1]重庆大学计算机学院软件工程国家重点实验室,重庆400044

出　　处：《计算机工程》2004年第8期159-161,共3页Computer Engineering

摘　　要：由于CT技术各方面的原因，工业CT扫描测量图像质量不是很理想。要获得精确的测量数据，必须进行准确的图像分割。该文采用多尺度二维小波变换模极大值作为分裂合并图像分割的一致性度量，采用四叉树结构作为图像块的初始划分，设置合适的门限值抑制噪声的干扰并提取出图像边缘信息。实验证明，对于皮鞋内嵌的CT测量图像该算法比常规的算法更理想，能够准确分割出皮鞋的内外腔边缘。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.

关 键 词：图像分割 小波变换 边缘检测 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]