摘要
提出一种流动拍照检测电视背面工艺的系统方法,该系统硬件包括定位机构和检测箱,其中检测箱包括镜头、相机、光源以及工控机。系统硬件结合深度学习图像算法,从而实现对电视背面工艺中的螺钉、卡扣、胶带进行精确检测。该系统平台采用流动拍照,相较静止拍照来说,节省了每台电视流板时间,因此检测速度更快。实验数据结果显示:该系统平台对电视背面卡扣、螺钉、胶带的检测精度可达95%以上。
This paper proposes a system platform for mobile photography to detect the backside of television.The hardware of the system includes a positioning mechanism and a detection box,wherein the detection box includes a lens,a camera,a light source and an industrial computer.The system hardware combines the deep learning image algorithm to realize the accurate detection of the screws,buckles and tapes on the back of the television.The system platform adopts mobile photography.Compared to still taking pictures,it saves the streaming time of each television, so the detection speed is faster.According to the experimental data,the system platform can detect more than 95% of the buckles,screws and tapes on the back of the television.
出处
《工业控制计算机》
2022年第5期49-50,共2页
Industrial Control Computer
关键词
流动拍照
深度学习
图像算法
mobile photography
deep learning
image algorithms
