基于小波变换与局部熵的织物疵点检测方法

Fabric Defect Detection Method Based on Wavelet Transform and Local Entropy

针对传统人工目测以及现有疵点检测方法检测精度与效率不高的问题,结合基于小波变换的织物疵点检测方法和基于局部熵的织物疵点检测方法各自的特点,提出了一种基于小波变换与局部熵的织物疵点检测方法。该方法对正常织物与待测织物图像进行一层小波变换,采用大小相同且不完全重叠的局部窗口提取正常织物与待测织物图像的经向细节子图像局部熵与纬向细节子图像局部熵,计算正常织物子图像与待测织物子图像局部熵的差值绝对值,根据阈值判断是否存在疵点并直接识别常见疵点类型。仿真结果表明,该方法的检测精度比传统的基于局部熵的布匹疵点检测方法更高。

Combining with the advantage of wavelet transform and local entropy, a method of fabric defect detection based on wavelet transform and local entropy is proposed for solving the problem of low detection precision and efficiency caused by the traditional eyeballing and the existed fabric defect detection. The method conducts a layer of wavelet transform to both normal fabric images and the images to be tested, uses local windows with the same size and incompletes overlap to extract their local entropy of warp and weft detail subimages, then calculates the absolute difference of local entropy of the two kinds of fabric subimages, at last estimates whether the defect is exist according to the judgement threshold and recognizes the common defect type directly. The simulation results show that the method has higher detection precision and efficiency than the traditional one based on the fabric defect detection.