布匹瑕疵检测算法研究进展  

Research progress of fabric defect detection

作　　者：程汉权 熊继平[1] 陈经纬 Cheng Hanquan;Xiong Jiping;Chen Jingwei(Zhejiang Normal University,College of Physics and Electronic Information Engineering,Jinhua,Zhejiang 321004,China)

机构地区：[1]浙江师范大学物理与电子信息工程学院,浙江金华321004

出　　处：《计算机时代》2023年第11期16-21,共6页Computer Era

基　　金：金华市公益项目资助(2021-4-116)。

摘　　要：纺织品行业中利用计算机视觉技术对织物瑕疵进行检测,已形成一种趋势。然而由于织物瑕疵种类繁多,形状、大小复杂,加之背景花色、纹路的存在,其检测十分具有挑战性。本文对比了常用的图像处理方法和深度学习算法,总结了目前织物瑕疵检测存在的问题和研究现状,并探讨了织物瑕疵检测的发展趋势,为研究者提供参考。In the textile product industry,the use of computer vision technology for fabric defect detection has been a trend.However,the detection of fabric defects is very challenging due to the wide variety of fabric defects with different shapes and sizes,as well as the complexity of the background patterns,colors and textures.In this paper,the commonly used image processing methods and deep learning algorithms are compared,the current problems and research status of fabric defect detection are summarized,and the development trend of fabric defect detection is explored,which provides reference for researchers.

关 键 词：深度学习 织物瑕疵 图像处理 瑕疵检测 

分 类 号：TP29[自动化与计算机技术—检测技术与自动化装置]