织物疵点的图像信息检测方法  

Image information detection method for fabric defect

作　　者：韩晓军[1] 黄雷[1] 

机构地区：[1]天津工业大学电子与信息工程学院,天津300387

出　　处：《天津工业大学学报》2015年第5期48-51,57,共5页Journal of Tiangong University

基　　金：国家自然科学基金资助项目(61405144)

摘　　要：针对织物破洞、断经、跳花和浆斑这4种类型疵点的特征,提出一种基于图像信息的检测方法.根据织物疵点的几何特性选择结构元素,应用形态学运算突出疵点形状信息,应用灰度积分投影法确定织物疵点的位置信息,然后分析纹理特征,提取织物疵点面积、周长等特征参数.结果表明:本方法克服了经典边缘检测算子只能给出疵点轮廓的弊端,利用图像信息检测进行计算机辅助验布,对于纺织信息化具有一定的理论意义和实用价值.A detecting method based on image information is proposed for four kinds of fabric defects （holes, cracked ends, skips and hard size） detection. The structure elements are selected according to the geometric characteristics of fabric defects, the morphological operation is used to highlight the shape information of fabric defect and grayscale integral projection method is used to determine the location information of fabric defect. Then texture features are analyzed to extract characteristic parameters such as area of fabric defects and perimeter. The result shows that the proposed method overcomes the problem that the classical edge detection operators can only give profile of fabric. Meanwhile, the proposed method uses image information detection for computer-aided cloth in- spection, which has theoretical significance and practical value for the textile information.

关 键 词：织物疵点 图像信息 图像处理 信息检测 形态学运算 纹理特征 特征参数 

分 类 号：TS101.9[轻工技术与工程—纺织工程] TP274[轻工技术与工程—纺织科学与工程]