织物纹理疵点快速检测方法

根据小波在奇异信号分解中的特点,提出了一种疵点快速检测新方法.首先通过对被检测图像进行小波提升分解,获得相应的子图.然后把高频子图分割成若干子窗口,统计子窗口能量的标准差与均值加权和作为提取的特征.最后把高频子图中的特征均值与正常子图特征均值相比较,判断疵点是否存在.实验结果表明,该检测方法能快速有效的检测疵点.

Fabric texture defect for rapid detection method

According to wavelet decomposition characteristics in singular signals, an innovative method for rapid defect detection is put forward. Firstly, acquired images are decomposed by lifting scheme into sub-images. Secondly, high frequency sub-images are segmented into many sub-windows,in which standard deviation and average arc calculated. Then they are weighted sum as characteristics. Lastly, the characteristic mean is used as basis of discrimination in every high frequency sub-image, which is compared with normal sub-image's characteristic mean to determine whether there exists defect. The experimental result confirms that the proposed method can is feasible in rapid defect detection.