基于图像纹理特征的炮膛疵病检测方法

摘要

In order to solve the problem of extracting features of flaw images of gun bore, extracting textural features of flaw images is proposed. The characteristics of several kinds of flaws of gun bore are analyzed. The concept of grey level co-occurrence triangular matrix (GLCTM) and four parameters of grey level co-occurrence matrix are applied to extract features of images. The experimental results show that the proposed method can abridge calculation time and reflect grey airspace information of images well, so the proposed method is effective.%为实现炮膛疵病的自动识别,提出提取炮膛疵病图像的纹理特征的方法.对若干类炮膛疵病特点进行分析,引用灰度共生三角阵的概念和灰度共生矩阵的熵、能量、对比度、相关度4个特征参量,提取图像的纹理特征.实验结果表明,该方法是有效的,能缩短计算时间,并充分反映图像的灰度空间信息.