基于BRDPSO算法的织物表面瑕疵检测  

Surface Defect Detection of Fabric Based on BRDPSO

作　　者：张家玮 李岳阳 罗海驰 ZHANG Jiawei;LI Yueyang;LUO Haichi(Jiangsu Provincial Engineering Laboratory of Pattern Recognition and Computational Intelligence,Jiangnan University,Wuxi 214122;Key Laboratory of Advanced Process Control for Light Industry,Ministry of Education,Jiangnan University,Wuxi 214122)

机构地区：[1]江南大学江苏省模式识别与计算智能工程实验室,无锡214122 [2]江南大学轻工过程先进控制教育部重点实验室,无锡214122

出　　处：《计算机与数字工程》2022年第5期1119-1125,共7页Computer & Digital Engineering

基　　金：国家自然科学基金项目(编号:51405198)资助。

摘　　要：针对织物表面瑕疵检测准确率和效率都偏低的问题,提出一种基于二进制随机漂移粒子群算法(BRDPSO)的同步特征选择与参数优化方法。该算法在原始特征集上进行特征选择,同时优化随机森林分类器(RF)的参数,并以随机森林分类准确率指导BRDPSO算法的搜索。最后用最优参数构建的RF对挑选出的特征子集进行织物表面瑕疵检测。结果表明,同步特征选择与参数优化的BRDPSO算法可以更有效地提高织物表面瑕疵检测准确率和效率,并与已提出的优化算法进行比较,其检测效果更优。Aiming at the problem of low accuracy and efficiency of fabric surface defect detection,a synchronous feature selection and parameter optimization method based on binary random drift particle swarm algorithm(BRDPSO)is proposed.BRDPSO is used to select features on the original feature set and simultaneously optimize the parameters of the RF classifier.The fitness function is constructed by the classification accuracy of random forest and the number of selected features.Finally,random forest classifier based on the optimal parameters is used to detect the surface defects of the fabric.The results show that BRDPSO algorithm based on synchronous feature selection and parameter optimization can improve the accuracy and efficiency of fabric surface defect detection more effectively,and detection results are better compared with the proposed optimization algorithm.

关 键 词：织物表面瑕疵检测 二进制随机漂移粒子群算法 特征选择 参数优化 随机森林 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]