基于相对总变差模型与自适应形态学的织物瑕疵检测

为解决目前基于图像处理的织物瑕疵检测算法中，因织物纹理的多样性与瑕疵形状尺寸的不确定性所造成的检测效果差的问题，提出一种基于结构-纹理模型与自适应数学形态学的织物瑕疵检测算法。首先采用相对总变差模型对织物图像进行滤波以去除织物纹理，然后在得到的灰度图像上直接进行基于自适应邻域的灰度形态学运算，形态学算子采用开运算算子，最终得到织物瑕疵的增强图像。采用基于相对总变差模型与自适应形态学相结合的方法与2种已知的Gabor算法进行比对，对4类典型织物瑕疵进行检测实验和分析。结果表明，本文方法能更好地提取出织物瑕疵。

Fabric defect detection based on relative total variation model and adaptive mathematical morphology

Because of the variety of fabric texture and the uncertainty of the shape and size of defects,the existing fabric defect detection methods based on image processing are low in accuracy.In order to solve this problem,a new method of fabric defect detection based on a structure-texture model and the adaptive mathematical morphology was designed.The fabric texture was firstly filtered based on the relative total variation model,then,the gray morphological operation based on adaptive neighborhood was directly performed on the gray level image,which is morphological opening,finally the enhanced image of fabric defects was obtained.The algorithm based on the relative total variation model and the adaptive mathematical morphology as well as the other two known algorithms based on Gabor filter was carried out on 4 types of fabric defects with high frequency,and the results show that the method can more effectively extract the fabric defects.