光学学报, 2013, 33 (3): 0312004, 网络出版: 2013-02-06   

基于机器视觉的大幅面陶瓷地砖尺寸测量研究

Research on Dimension Measurement of Large Size Ceramic Floor Tiles Based on Machine Vision
作者单位
1 佛山科学技术学院机电系, 广东 佛山 528000
2 华南理工大学机械与汽车工程学院, 广东 广州 510640
3 东莞科视自动化科技有限公司, 广东 东莞 523000
摘要
提出一种基于机器视觉的单相机大幅面陶瓷地砖尺寸高精度测量方法。设计一种由多个反射镜组成反射光路的光学系统。该光学系统采用单个相机采集陶瓷地砖4个角的图像,而不是整幅图像,避免了采用多个相机采集测量图像导致不同步的问题,提高了测量精度。提出一种基于边界搜索拟合和动态补偿的瓷砖测量算法。在该算法中,考虑实际生产中瓷砖的偏移和旋转,利用采集的瓷砖4个角图像,通过对图像进行边界搜索和测量尺寸动态补偿,获取陶瓷地砖高精度的对角线和边长尺寸。设计的系统用于测量600 mm×600 mm及800 mm×800 mm瓷砖的对角线和边长尺寸时,其标准差均小于0.02 mm,重复性精度接近0.06 mm。实验结果表明,该测量系统能够在高档陶瓷地砖生产企业得到广泛的应用。
Abstract
A dimensional measurement method with high accuracy for large size ceramic floor tiles based on the machine vision using single camera is presented. An optical system is designed, whose reflective light path consists of multiple mirrors. The images of four corners of the ceramic floor tiles are acquired by the single camera rather than the whole image, thus the asynchronous problem caused by multiple cameras acquiring the measured images is solved. Then the accuracy of measurement is improved. An algorithm for the tiles measurement is proposed based on boundary searching fitting and dynamic compensation. In such an algorithm, considering shift and rotation of the tiles in practical production, the dimension of diagonals and sidelines with high accuracy can be obtained by using the images of the four corners, which boundary searching of the measured images and dynamic compensation of measured dimension is performed. The measured standard deviation is less than 0.02 mm, and the repeatability approaches 0.06 mm when measuring dimension of diagonals and sidelines of 600 mm×600 mm and 800 mm×800 mm tiles. The experimental results show that the measurement system can be widely used in ceramic floor tiles manufacturers with high quality.

卢清华, 许重川, 王华, 张宪民, 范彦斌. 基于机器视觉的大幅面陶瓷地砖尺寸测量研究[J]. 光学学报, 2013, 33(3): 0312004. Lu Qinghua, Xu Chongchun, Wang Hua, Zhang Xianming, Fan Yanbin. Research on Dimension Measurement of Large Size Ceramic Floor Tiles Based on Machine Vision[J]. Acta Optica Sinica, 2013, 33(3): 0312004.

本文已被 7 篇论文引用
被引统计数据来源于中国光学期刊网
引用该论文: TXT   |   EndNote

相关论文

加载中...

关于本站 Cookie 的使用提示

中国光学期刊网使用基于 cookie 的技术来更好地为您提供各项服务,点击此处了解我们的隐私策略。 如您需继续使用本网站,请您授权我们使用本地 cookie 来保存部分信息。
全站搜索
您最值得信赖的光电行业旗舰网络服务平台!