面向高速视觉检测的精确抓拍安全策略研究  被引量:5

Research on accurate capture security strategy for high-speed visual inspection

在线阅读下载全文

作  者:张堃[1,2] 王震 张培建 华亮[1] 费敏锐[3] Zhang Kunl,;Wang Zhen;Zhang Peijian;Hua Liang;Fei Minrui(School of Electrical Engineering, Nantong University, Nantong 226007, China;Nantong Research Institute for Advanced Communication Technologies, Nantong 226007, China;Shanghai Key Laboratory of Power Station Automation Technology, School of Mechatronic Engineering and Automation, Shanghai University, Shanghai 210053, China)

机构地区:[1]南通大学电气工程学院,南通226007 [2]南通先进通信技术研究院,南通226007 [3]上海大学机电工程与自动化学院上海市电站自动化技术重点实验室,上海210053

出  处:《仪器仪表学报》2018年第2期232-240,共9页Chinese Journal of Scientific Instrument

基  金:国家自然基金重点项目(61633016);江苏省产学研前瞻性项目(BY2016053-11);南通市应用基础研究-工业创新项目(GY12016022);南通大学-南通智能信息技术联合研究中心项目(KFKT2016A08)资助

摘  要:为了解决高速流水线机器视觉检测相机精确抓拍问题,以香烟爆珠为例,提出了两层网络控制的精确抓拍安全策略。两层网络的底层网络实现对目标抓拍,通过线搜索获得感兴趣区域位置,并上送到高层网络进行控制;高层网络主要处理复杂算法为基于柔性专家控制的迭代学习控制和粒子滤波相结合的相机快门控制方法,不断更新香烟爆珠在图像中的位置偏移量,进一步控制相机快门延迟时间,实现精确抓拍高速运动的爆珠的图像。针对上述方法,建立了控制模型,并对模型进行仿真、分析。最后,通过实验平台测试,可以抵消平台运行中的干扰,保证了精确抓拍安全策略的可行性。To precisely capture the object in the scene of high-speed pipeline by machine vision inspection,this paper takes accurate snapshots of cigarette pop bead as an example,and a two-layer network control algorithm to capture the details in the scene is proposed.The bottom structure of the two-layer network captures the object image and obtains the area of perceptive interests through line search.This area is transmitted to the upper layer to achieve the following control. The upper layer network mainly deals with the complex algorithm. The camera shutter control method is the combination of iterative learning control and particle filter based on the flexible expert control. The camera shutter delay time is adjusted by the continuously updated position of cigarette pop bead. In this way,the accurate position of pop bead under the high-speed movement condition can be achieved. Based on the above method,the control model is formulated,simulated and analyzed. Evaluated by the experimental platform,the results show that the interferences can be reduced and the safe strategy of accurate capture is feasible.

关 键 词:高速机器视觉 精确抓拍 两层网络 迭代学习 柔性专家控制 

分 类 号:TP273[自动化与计算机技术—检测技术与自动化装置] TH89[自动化与计算机技术—控制科学与工程]

 

参考文献:

正在载入数据...

 

二级参考文献:

正在载入数据...

 

耦合文献:

正在载入数据...

 

引证文献:

正在载入数据...

 

二级引证文献:

正在载入数据...

 

同被引文献:

正在载入数据...

 

相关期刊文献:

正在载入数据...

相关的主题
相关的作者对象
相关的机构对象