摘要
在自动化车间中,视觉AGV系统的性能与标识符的检测速度和准确率密切相关。圆形工位作为系统常用的标识符在实际的工厂环境中经常受到污染导致脏污残缺现象,造成系统性能降低。针对这个问题,提出了一种改进的Hough变换算法。该算法利用边界跟踪方法提取完整工位轮廓,接着使用改进的Hough变换对工位轮廓像元进行两次筛选确定最终检测结果。编程实验证明该算法能在20 ms时间内准确地识别脏污残缺圆形工位,达到了在线检测要求。
In automation workshop,the speed and accuracy of identifiers detection affect the performance of visual AGV system.In order to resolve the problem of performance reduce caused by identifiers＇ dirty and incomplete,an improved Hough transform algorithm was presented.It used the method of boundary tracing to extract the full profile,and then,the pixels of profile were filtered twice to reach the detect result by improved HT.Programming experiments show that the algorithm can accurately identify dirty and incomplete circular station within 20ms,to on-line testing requirements.
出处
《化工自动化及仪表》
CAS
北大核心
2010年第9期46-49,共4页
Control and Instruments in Chemical Industry
基金
河北省自然科学基金资助项目(F2008000860)
