基于联盟博弈和极限学习机的棉花异性纤维识别方法  

Recognition method of foreign fibers in cotton based on cooperative game theory and extreme learning machine

作　　者：赵学华[1] 王名镜 刘双印 徐龙琴 刘文娟[1] ZHAO Xuehua1, WANG Mingjing1 , LIU Shuangyin2, XU Longqin2. , LIU Wenjuan1(1. School of Digital Media, Shenzhen Institute of Information Technology, Shenzhen 518172, China; 2. College of Information Science and Technology, Zhongkai University of Agriculture and Engineering, Guangzhou 510225, Chin)

机构地区：[1]深圳信息职业技术学院数字媒体学院,广东深圳518172 [2]仲恺农业工程学院信息科学与技术学院,广东广州510225

出　　处：《仲恺农业工程学院学报》2018年第1期46-52,共7页Journal of Zhongkai University of Agriculture and Engineering

基　　金：国家自然科学基金(61571444;61471133);广东省自然科学基金(2016A030310072);广东省普通高校省级重大科研项目(2016KZDXM001);广东省科技计划(2017A070712019);教育部人文社会科学研究青年基金(17YJCZH261);深圳信息职业技术学院科研培育项目(ZY201718)资助

摘　　要：针对棉花异性纤维(棉花采摘、摊晒、收购、储存、运输及加工过程中混入棉花中的非棉纤维)识别问题,提出了一种基于联盟博弈和极限学习机相融合的棉花异性纤维识别方法,该方法利用基于联盟博弈的特征选择方法确定最优的特征集,随后利用极限学习机进行棉花异性纤维识别并与支持向量机、k近邻法进行了试验比较.试验结果表明,该方法、支持向量机和k近邻法可以实现的准确率分别为90.15%、88.46%和86.30%.相对于另两种方法,该方法具有最高的识别准确率,并使特征集的特征数由75个降为25个.The cotton foreign fiber referred to the non-cotton fiber such as hair, linen, silk, fiber, dyeing silk and plastic film. To improve the cotton foreign fiber recognition accuracy, the feature selection based on cooperative game theory and extreme learning machine was fused together. The optimal feature set was selected and the dataset was rebuilt, and then the extreme learning machine was trained on the rebuilt dataset. In the experiments, the comparisons were made with support vector machine and k - nearest neighbour （kNN）. The experimental results showed that the accuracy of the proposed method, support vector machine and kNN were 90. 15% , 88.46% and 86. 30% , respectively. Compared to the other two methods, the proposed method had the best accuracy among them and the number of features was reduced from 75 to 25.

关 键 词：棉花异性纤维识别 特征选择 极限学习机 

分 类 号：TP271[自动化与计算机技术—检测技术与自动化装置]