文献类型：期刊文献

中文题名：基于支持向量机的钢板表面缺陷检测

英文题名：Steel Plate Surface Defect Recognition Based on Support Vector Machine

作者：郭慧[1];徐威[1];刘亚菲[1]

机构：[1]华东理工大学机械与动力工程学院,上海200237

年份：2018

卷号：44

期号：4

起止页码：635

中文期刊名：东华大学学报（自然科学版）

外文期刊名：Journal of Donghua University(Natural Science)

收录：CSTPCD;;北大核心:【北大核心2017】；CSCD:【CSCD_E2017_2018】；

语种：中文

中文关键词：支持向量机;钢板表面缺陷检测;图像处理

外文关键词：support vector machine;steel plate surface defects detection;image processing

摘要：针对钢板缺陷的传统检测方法存在速度慢、工作量大的问题。采用机器视觉的方法,通过采集钢板表面图像信息,由计算机算法处理得到缺陷的特征样本,使用支持向量机提升分类的速度和准确度。试验结果表明,径向基核函数支持向量机方法对钢板表面各种缺陷的准确识别率达到90%及以上,为钢板表面缺陷检测技术提供了很好的支持。
Aiming at the problem of slow speed and large workload of the traditional detectionmethod for steel plate defects, the method of machine vision was adopted to acquire the image information of the surface of the steel plate, and the feature samples of the deby a computer algorithm. The speed and accuracy of the classification are improved by using a support vector machine. The experimental results show that the recognition accuracy of various defects on thesurface of the steel plate reaches 90% and above with the radial basis kernel function support vector machine method,which provides a good support for surface defect detection technology of steel plate.