基于机器视觉的织物疵点检测技术探讨  

Discussion on the Detection Technology of Fabric Defects Based on Machine Vision

作　　者：纪惠军[1] 张小栋[2] 杨建民[1] 

机构地区：[1]陕西工业职业技术学院,陕西咸阳712000 [2]西安交通大学机械工程学院,西安710049

出　　处：《现代纺织技术》2011年第2期11-15,共5页Advanced Textile Technology

摘　　要：针对传统人工视觉检测技术的缺陷,建立一种基于机器视觉的织物疵点检测技术方案。在深入讨论检测系统硬件设计的基础上,重点讨论了织物疵点检测流程、获取织物特征的拟合方法、疵点特征提取流程、织物疵点分类与织物等级评定等。同时,讨论了织物疵点自动检测系统实际应用时应注意的几个关键问题。为快速、准确、有效的检测织疵,提升检测技术水平和加强产品质量控制,具有十分重要的现实意义。A program of testing fabric defects based on machine vision was established to overcome the weak points in the traditional artificial vision test.On the basis of discussion about the hardware design of the testing system,the paper emphasized on the testing process of fabric defects,the fitting methods of gaining fabric features,the extraction process of defects features,the classification of fabric defects and the grade evaluation of fabric,and so on.At the same time,the paper discussed several key problems which should be paid attention to in practical application of the automatic detection system of fabric defects.The paper had important practical significance to make the defects detection faster,more accurate and more effective,to improve the detection technology level and to enhance the quality control of products.

关 键 词：机器视觉 疵点检测 特征提取 等级评定 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]