摘要
采用机器视觉技术,应用Halcon软件研究了对工件进行识别和定位的问题。在建立工件模板的基础上,采用一种改进的模板匹配与神经网络相结合的算法对工件特征进行提取,并与工件模板进行对比,然后采用MLP神经网络分类器识别出工件的种类。实验结果表明,工件的识别率和识别速度有了明显提高,可以达到较好的识别定位效果。
This paper uses machine vision technology and Halcon software to study the problem of identifying and locating workpieces.An improved method of template matching and neural network is proposed.The template is established for the workpiece,the image processing algorithm is used to extract the feature of the workpiece,and compared with the workpiece template.Then the MLP neural network classifier is used to identify the type of the workpiece.The experimental results show that the recognition rate and recognition speed of the workpiece are obviously improved,and the recognition and positioning effect can be achieved.
作者
孙宇
许钢
SUN Yu;XU Gang(Anhui University of Engineering, Wuhu 241000 China)
出处
《新余学院学报》
2018年第6期39-44,共6页
Journal of Xinyu University
基金
安徽工程大学检测技术与节能装置安徽省重点实验室开放基金资助项目"基于视觉场景外观建模的SLAM回环检测研究"(2017070503B026)
关键词
HALCON
模板匹配
神经网络
工件识别
工件定位
Halcon
template matching
neural network
workpiece recognition
workpiece localization
