基于小波变换的物体轮廓缺陷图像检测研究

Research on Object Contour Defect Detection Using the Imaging Wavelet Transformation

提出将二维物体的轮廓图像坐标信息转换成基于质心距离的一维向量信息来进行检测.并设计了基于双正交对称小波的去噪算法.实验结果表明,这种算法可以有效地去除噪声信号和缓变信号,保留缺陷的突变信息,是一种有效和实用的轮廓缺陷检测算法.

The two-dimensional contour imaging coordinate information is transferred to one-dimensional vector of the centroid-based distance is proposed in the paper.The de-noising algorithm based on symmetric bi-orthogonal wavelet is designed.The experimental results show that the defect detect result of this algorithm can effectively remove the noise signal and the gentle change signal with the deficient mutant information retained.It is an effective and practical contour defect detection algorithm.