钢丝综质量检测方法研究  

Research of Steel Wire Heald Automatic Detection

作　　者：刘春辉[1] 徐平[1] 许彬[1] 常英杰[1] 

机构地区：[1]杭州电子科技大学生命信息与仪器工程学院,浙江杭州310018

出　　处：《杭州电子科技大学学报（自然科学版）》2015年第5期48-53,共6页Journal of Hangzhou Dianzi University：Natural Sciences

基　　金：国家自然科学基金资助项目(61205200);杭州电子科技大学研究生科研创新基金资助项目(2014KYJJ018)

摘　　要：针对钢丝综产品的质量检测问题进行了研究,设计出一套双面成像装置。用主轴旋转法校正钢丝综图像,采用废次品延时剔除的策略,提出了一种钢丝综分类识别方法。方法首先用5个几何量特征定位感兴趣区域,接着提取灰度直方图和梯度特征以及灰度共生矩阵特征作为分类特征,用带PCA的SVM进行分类识别。通过实验表明,钢丝综分类识别法具有较高的识别能力。The steel wire heald is indispensable component of textile machine, in order to solve the key technology of steel wire heald detection,designed a double-sided imaging device;offered a method for correct the wire heald image bashed on spindle rotation;put forward a kind of waste eliminating strategy;then five geometric features were used to locate the region of interested,next,feature of gray histogram,gradient and gray level co-occurrence matrix features were fused, at last, adapt SVM with PCA to classify and identify the characteristics.In this paper,the experimental results show that the method has higher recognition ability.

关 键 词：钢丝综 感兴趣区域定位 灰度共生矩阵 降维 支持向量机 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]