摘要
针对自动扶梯运行速度的检测,提出一种基于机器视觉的自动扶梯梯级测速方法。首先,对机器视觉系统采集得到的梯级运行图像进行一系列的预处理,通过设置阈值实现目标颜色的识别;然后,采用直线扫描法对颜色提取后的不连通区域进行填充,对得到的二值图像进行边缘检测与轮廓识别,返回轮廓面积最大的图像;最后,提取每张图像中黄色区域右边界直线起始端点的横坐标,通过差分法计算得到梯级的速度。结果表明,该方法实现了非接触测量并且能够得到较为准确的检测结果。
In order to detect the running speed of the escalator, a method for the escalator step speed measurement based on machine vision is proposed. Firstly, perform a series of pre-processing on the run-level images of the escalator acquired by the machine vision system and identify the target color by setting thresholds. Then, fill out the disconnected areas after color extraction by linear scanning method, edge detection and contour recognition are used on the obtained binary image, return the image with the largest contour area. Finally, extract the abscissa of the starting point of the right boundary line of the yellow area in each image and the difference method is used to calculate the speed of the escalator step. The results show that the method achieves non-contact measurement and can obtain more accurate detection results.
作者
王桔
张斌
应征
李建
马锦程
WANG Ju;ZHANG Bin;YING Zheng;LI Jian;MA Jincheng(College of Metrology and Measurement Engineering,China Jiliang University,Hangzhou 310018,China;Zhejiang Provincial Special Equipment Inspection and Research Institute,Zhejiang Key Laboratory of Special Equipment Safety Testing Technology,Hangzhou 310020,China)
出处
《中国测试》
CAS
北大核心
2019年第8期44-49,共6页
China Measurement & Test
基金
浙江省自然科学基金资助项目（LY17E050015）
2017年度浙江省质监科研计划项目（20170229）
关键词
机器视觉
速度
差分法
自动扶梯
machine vision
speed
difference method
escalator
