摘要
为了满足布匹瑕疵检测的准确性和实时性要求,通过对几种纹理分析方法的讨论,提出了一种智能化的瑕疵检测方法.该方法将图像等分成局部窗口,利用局部标准差和平均值对图像进行初检测,确定瑕疵存在的区域.然后采用Gabor滤波器,在多分辨率和多通道上分别对图像有瑕疵的区域进行滤波,建立金字塔图像表示结构,并对滤波后的多幅图像进行融合和重建,把瑕疵从布匹背景中提取出来,从而完成瑕疵的检测.实验证明,该方法检测速度快,准确率高,能够满足检测要求.
To meet the requirements of accuracy and real-time in fabric defect detection,this article proposed an intelligent method through several kinds of texture analysis method discussion.This method divides the image into several windows,and uses local mean and standard deviation,which is the image early detection,to identify the regions of defects.Then Gabor filters are used in the multi-channel and multi-resolution of the regions and a pyramid structure formed.Finally,the images are fused and reconstructed,thus detection can be defected.Experiments showed that the method is fast,accurate and can meet the detection requirements.
出处
《河南工程学院学报（自然科学版）》
2010年第4期11-14,共4页
Journal of Henan University of Engineering：Natural Science Edition
基金
河南工程学院青年科研基金项目(Y09029)
