基于改进KCF算法的织物折皱回复检测研究  

Research on fabric wrinkle recovery detection based on an improved KCF algorithm

作　　者：郭栩源 李忠健 王蕾[1] 潘如如[1] 高卫东[1] GUO Xuyuan;LI Zhongjian;WANG Lei;PAN Ruru;GAO Weidong(Key Laboratory of Science&Technology of Eco-Textile,Ministry of Education,Jiangnan University,Wuxi 214122,China;Key Laboratory of Clean Dyeing and Finishing Technology of Zhejiang Province,Shaoxing University,Shaoxing 312000,China)

机构地区：[1]江南大学生态纺织教育部重点实验室,江苏无锡214122 [2]绍兴文理学院浙江省清洁染整技术研究重点实验室,浙江绍兴312000

出　　处：《丝绸》2024年第4期79-86,共8页Journal of Silk

基　　金：纺织之光应用基础研究计划项目(J202109);国家自然科学基金项目(61802152);浙江省基础公益研究计划项目(LGG21F030007);中国博士后科学基金面上资助项目(2020M681736);江南大学研究生科研与实践创新项目(KYCX-23-ZD01,KYCX-23-ZD02)。

摘　　要：织物折皱回复性能是评价织物形态稳定性的关键指标。传统折皱回复角测试方法存在检测过程依赖人工操作、难以量化折皱动态演变等问题。为实现对折皱回复全过程的自动化监测,文章提出一种基于改进核相关滤波算法的动态折皱回复检测方法。该方法使用高速摄像机捕捉织物折皱形变回复过程,应用改进的核相关滤波算法检测追踪折皱顶点的运动角度变化。通过引入多特征融合提高检测鲁棒性,利用Canny边缘检测自适应调整目标区域,减小边界效应。在此基础上提取感兴趣区域骨架,计算折皱顶角度随时间变化信息。结果表明,不同织物折皱角度变化规律与织物组织结构高度相关。最后与标准测试结果建立线性模型,验证所提方法的有效性。文章实现了对织物折皱回复全过程的自动化检测与定量评估,提供了一种更为高效准确的折皱回复性能检测新思路,具有广阔应用前景。Fabric wrinkle recovery performance is an important indicator for evaluating the shape retention of fabrics.It reflects the ability of fabrics to spontaneously recover their original shape after wrinkling during use and processing.Good wrinkle recovery performance enables fabrics to quickly recover after folding,transportation and cutting,thereby improving utility.The currently widely used standard method for evaluating fabric wrinkle recovery performance is the wrinkle recovery angle test.This involves folding and pressing fabric samples and then measuring the angle of recovery for the resulting creases to assess the fabric’s wrinkle recovery capability.However,this method is complex to operate,relies on manual judgment and measurement for testing,yields results vulnerable to subjective effects,and cannot distinguish or quantify the dynamic evolution of wrinkles during different recovery stages.It can only provide an approximate overall evaluation and fails to accurately describe wrinkle recovery patterns in detail.Therefore,there is a need to develop a new testing method to achieve automated monitoring and quantified parameterization analysis of the entire wrinkle recovery process,in order to more efficiently,precisely and comprehensively evaluate the wrinkle recovery performance of different fabric materials and provide a basis for fabric performance optimization.In view of this,this study proposed a fabric wrinkle recovery detection method based on an improved kernel correlation filter algorithm to achieve automated monitoring and key parameter extraction of the entire fabric wrinkle recovery process.As for the method,a high-speed camera was used to collect and record images throughout the dynamic process of the formation and recovery of fabric wrinkles and a set of detection system was designed to process and analyze the image sequence.In the system,a robustness-enhanced improved KCF algorithm was first applied to track and locate fabric wrinkle areas.This algorithm improved adaptation to texture and shape

关 键 词：折皱回复角 特征融合 目标追踪 核相关滤波器 改进KCF 

分 类 号：TS103.7[轻工技术与工程—纺织工程]