基于混合自适应小波基的织物疵点检测算法

为了改进具有复杂纹理织物的疵点检测效果,提出了一种基于混合自适应小波基的织物疵点检测算法,采用各自优化的自适应小波基实现对不同层织物图像的分解变换。首先对正常图像和其经一层小波分解后的低频子图像优化得到混合自适应小波基,然后用该小波基将织物疵点图像进行二层小波分解,最后采用阈值法对径向子图像进行分割得到检测结果。实验结果证明,本文提出的算法能有效实现疵点检测,具有较好的疵点分割和定位结果。

Fabric Defect Detection Based on Hybrid Self-adaptiveWavelet Basis

In order to improve the detection effect for the fabric with complex texture, proposed a fabric defect detection algorithm based on hybrid self-adaptive wavelets, and adopted different adaptive wavelet basis to decompose each layer of the fabric image. Firstly, the hybrid self-adaptive wavelet basis is generated from the original image and 1-L wavelet decomposed image; then the generated self-adaptive wavelet is used to implement 2-L wavelet decomposition; finally, the threshold segmentation method split the radial sub image into defect region and background region. Experimental results demonstrate that the proposed algorithm can effectively detect defect with better result of defect segmentation and location.