炭素制品X射线图像缺陷的自动提取与分割

Automatical defect extraction and segmentation in X-ray images of carbon produce

针对炭素制品X射线检测图像的特点,对其缺陷提取技术进行了研究。首先设计了目标边界提取和基于小波变换的图像增强算法,实现了原始图像中目标区域的增强及其背景的去除。在此基础上,提出对不同的缺陷类型,可分别通过两条途径来实现:一是采用小波变换提取缺陷边缘,二是采用数学形态学结合迭代阈值法提取缺陷区域。实验结果表明,两者均较好地实现了缺陷的自动提取与分割,为缺陷特征参数的提取与选择奠定了良好的基础。

Regarding the characteristic of X-ray detection images of carbon produce, defect extraction techniques were studied with target boundary extraction algorithm and image enhancement algorithm based on wavelet, background removal and enhancement of object region were implemented successfully. Based on this, there had two ways to recognize different type of defect： firstly, wavelet transforms was introduced to extract defect edge, secondly, mathematical morphology linking iteration threshold was adopted to extract defect area. The experimental resuhs indicate that both of methods can achieve defect extraction and segmentation automatically, which will lay a good foundation for flaw feature parameter extraction.