基于小波变换和阈值分割的织物疵点边缘检测

Edge Detection of Fabric Defects Based on Wavelet Transform and Threshold Segmentation Algorithm

疵点边缘检测是提取疵点形状特征的基础,是疵点识别的重要步骤。文章提出了基于离散平稳小波变换和最佳阈值分割算法的织物疵点边缘检测方法。首先对织物疵点图像去除背景,然后对其进行离散小波变换和拉普拉斯算子增强,并对增强后的疵点图像采用最佳阈值分割和形态学运算相结合的方法,最后对织物疵点进行边缘检测。经比较发现,所采用的方法优于经典的边缘检测方法,对织物疵点边缘检测更为有效。

The edge detection of fabric defects is the base of geometrical features extraction and the essential process of the fabric defects identification, This paper proposed a method for fabric defects edge detection based on discrete stationary wavelet transform （DSWT） and optimal threshold segmentation algorithm （OTSA）. Firstly, the background of fabric defects picture was removed, then it was executed through DSWT and enhanced by the Laplacian operator. Finally, the edge detection was carded out with both OTSA and morphological operation. By contrast, this method is better than the classic ones, and is effective to fabric defect edge detection.