【摘要】 在分析织物疵点特点的基础上,提出了一种新的检测方法.首先根据正常织物小波分解后的能量来确定小波基和小波分解层数.接着对测试图像进行小波静态分解,然后提取子图中的子窗口标准差作为提取的特征;在此基础上,通过与正常子图中的特征相比较来判断小疵点是否存在.最后通过提取特征构造的标准差图确定小疵点的位置.实验表明,该检测方法是有效的.更多还原

【Abstract】 On the basis of analysis of the defect characteristicsof fabrics,a new method for small defect detection was presented.Firstly,the normal fabric energy of wavelet decomposition was used to determine the wavelet func-tion and level of wavelet decomposition.The test image was decomposed statically by wavelet,and then stan-dard deviation of sub-window in the sub-image was extracted as the extracted features,which was compared with the normal sub-image features to determine the existence of defects.Finally,the location of defect was determined by the standard deviation image.Experimental results show that this method is effective.更多还原