基于Halcon的水平仪气泡自动分类系统研究  

Research on level bubble automatic classification system based on Halcon

作　　者：项新建[1] 黄佩 郑永平 王文丽 XIANG Xinjian;HUANG Pei;ZHENG Yongping;WANG Wenli(School of Automation and Electrical Engineering,Zhejiang University of Science and Technology,Hangzhou 310023,Zhejiang,China)

机构地区：[1]浙江科技学院自动化与电气工程学院,杭州310023

出　　处：《浙江科技学院学报》2019年第3期180-186,共7页Journal of Zhejiang University of Science and Technology

基　　金：浙江省教育厅一般科研项目(Y201018333)

摘　　要：为了实现生产过程中水平仪气泡实时快速分类而研发机器视觉水平仪气泡自动分类系统,这是水平仪生产机器换人的关键。研究了基于Halcon的图像处理与识别技术,结合Qt语言设计了自动分类系统软件。介绍了自动分类系统的硬件构成、工作原理,以及图像获取和识别水平仪气泡特定参数的方法。分析了自动分类过程中图像滤波、边缘检测、轮廓提取等图像处理技术,并进行了实验验证。结果表明,使用基于Halcon的机器视觉图像处理技术,能使水平仪的气泡分类速度更快、更准确,单张图像平均检测时间≤786 ms,长度检测平均误差≤0.02 mm,满足水平仪生产机器换人的需要。The level bubble automatic classification system based on machine vision, which is developed for the sake of real-time rapid classification of level bubbles in the production process, is the key to machine substitution for level production. The study designed the software of automatic classification system with Qt language, on the basis of Halcon image processing and recognition technology. It introduced hardware composition, working principle, image acquisition, and method of identifying the specific parameters of the level bubble. It analyzed the image processing techniques such as image filtering, edge detection and contour extraction in the automatic classification process, and implemented experiments for verification. The results show that using Halcon-based machine vision and image processing technology can accelerate level bubble classification and make it more accurate. The average detection time of a single image is not more than 786 ms, and the average error of length detection is not more than 0.02 mm, which can meet the needs of machine substitution for level production.

关 键 词：HALCON 自动分类系统 图像处理 QT 

分 类 号：TP391.41[自动化与计算机技术—计算机应用技术]