一种基于改进Canny算法的塑胶手套残次品检测方法  

An Improved Canny Algorithm for Inspection of Defective Plastic Gloves

作　　者：李文文[1] 杨先海[1] 潘广堂 

机构地区：[1]山东理工大学机械工程学院,山东淄博255049

出　　处：《塑料科技》2017年第10期102-105,共4页Plastics Science and Technology

基　　金：山东省自然科学基金(ZR2014EEM046)

摘　　要：为提高塑胶手套质量检测的检测精度和效率,提出了一种Canny结合多阈值的改进边缘检测算法。该改进算法运用直方图拟合曲线提供多阈值,从而对塑胶手套边缘进行检测。实验证明,该方法能够准确识别出塑胶手套中的残次品,并且实现了塑胶手套检测自动化。In order to improve the accuracy and efficiency of quality inspection, an intelligent detection approach for inspection of defective plastic gloves based on the improved Canny algorithm is proposed, which uses the histogram fitting curve to provide multi threshold, and detects the edge of plastic gloves. The experiment proof that this method can accurately identify the defects of plastic gloves and it is easy to realize the automation of plastic glove detection.

关 键 词：塑胶手套 多阈值 边缘检测 灰度直方图 拟合曲线 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]