基于互相关函数的织物纬档检测

Detection of fabric weft bar based on cross-correlation function

在纺织品的生产中,由于各种不利因素的影响,往往会产生纬档,从而降低织物的品质。在图像的检测中应用随机过程的互相关性,对织物纬档的检测进行了研究,分析了互相关函数值与2幅图像相似性的关系。把模板图像与目标图像中的子图分别看作2个互不相关的随机过程,互相关函数是描述2个随机过程之间的内在联系,其互相关函数的值愈大图像愈相似,否则愈不相似。经验证,实验结果表明是有效的,并可进一步实现纬档定级的数字化。

Due to the effect of disadvantage factors, weft bar are often produced in textile products. The quality of fabric will be debased. A research on detection of weft bar was conducted using cross correlation of random process during image detecting. The relation of the cross correlation value with the similarity of the two images was analyzed. Taking the template image and the sub- image of target image as non-correlation random process, the cross correlation function is described as the inner-relation of two random processes. The bigger the value of the cross correlation function is, the more similar the two images, or vice versa. And verification experiments were implemented. The results indicated that the method is workable and the digitalized grading of the weft bar can be accomplished.