摘要
为提高识别织物疵点的准确率,提出基于离散粒子群算法(BPSO)进行织物疵点特征选择的方法。首先收集织物疵点图像并进行预处理,提取常用的纹理特征构成候选特征。然后采用BPSO对这些候选特征进行选择,得到优选特征和冗余特征。最后分别在这3类特征下训练支持向量机并进行织物疵点识别测试。结果表明,优选特征的疵点识别准确率大大高于另外2类特征,验证了这种方法是有效的。
To improve the accuracy of fabric defect recognition,a texture feature selection method for defeat recognition was proposed based on binary partial swarm optimization（BPSO）.Firstly defect images were collected and preprocessed,and the texture features to form candidate features were extracted.Then BPSO was applied to select optimal features and redundant features from the candidate features.Finally,the support vector machine（SVM） was trained with these three features to classify defects,respectively,and trial test of fabric defeats recognition was conducted with the SVM.The experiment resultes show that the classification accuracy of optimal features is much higher than those of the other two features,demonstrating that the method is feasible and effective for feature selection of fabric defects.
出处
《纺织学报》
EI
CAS
CSCD
北大核心
2011年第11期53-57,共5页
Journal of Textile Research
基金
国家"863"高技术研究发展计划资助项目(2007AA01Z187)
国家自然科学基金资助项目(60775029)
浙江省自然科学基金资助项目(Y1100036)
浙江省教育厅科研计划资助项目(Y201016929)
浙江省高校优秀青年教师资助计划项目(2010)
关键词
离散粒子群算法
织物疵点
特征选择
支持向量机
binary partial swarm optimization
fabric defect
feature selection
support vector machine
