摘要
针对人工检测瓶盖表面黑点易出现漏检的问题,设计了基于Canny算子和SVM的瓶盖缺陷检测系统,给出系统总体架构,阐述系统软硬件设计思路,重点研究Canny算子与SVM相结合的瓶盖缺陷识别技术。实验结果表明,相对于传统人工瓶盖黑点的缺陷检测,设计的系统提高了缺陷识别的准确率。
In order to sovle the problem of missed detection of traditional methods in the detection of black spots on bottle caps, a bottle cap defect detection system based on Canny operator and SVM is designed. While presenting the overall architecture of the system, the design ideas of both hardware and software were elaborated. The bottle cap defect identification technology combined with Canny operator and SVM algorithm was mainly studied. The experimental results show that the designed system improves the accuracy of defect recognition compared with the conventional manual inspection.
引文
[1]王宇,吴智恒,陈启愉,等.基于机器视觉的瓶盖封装检测系统[J].自动化与信息工程,2016,37(2):42-44.
[2]龙智帆.基于机器视觉的瓶盖缺陷检测算法设计及可重构方案的研究[D].杭州:杭州电子科技大学,2011.
[3]童季刚,廖菲,罗良传.一种机器视觉的瓶罐缺陷检测系统设计[J].机电工程技术,2016,45(8):28-31.
[4]宋超,秦永彬,许道云.基于K-means与SVM的鸡蛋特征检测[J].计算机与数字工程,2017,45(2):382-386.
[5]李晓宇,张新峰,沈兰荪.支持向量机(SVM)的研究进展[J].测控技术,2006,25(5):7-12.
[6]王雒瑶,高炜欣,王欣.一种基于SVM及LE降维的X射线焊缝缺陷分类算法研究[J].西安石油大学学报(自然科学版),2017,32(5):96-101.