基于双目视觉的废料尺寸测量系统设计与研究  

Design and Experiment Research Device of Waste Size Measurement System Based on Binocular Vision

作　　者：杨亮[1] 王琦 YANG Liang;WANG Qi(School of Mechanical Engineering,Dalian Jiaotong University,Dalian 116028,China)

机构地区：[1]大连交通大学机械工程学院,辽宁大连116028

出　　处：《大连交通大学学报》2024年第5期69-74,共6页Journal of Dalian Jiaotong University

摘　　要：为了克服工业废料尺寸测量时人工测量成本高且效率低的问题,提出了一种双目视觉系统。该系统融合了精密机械结构、双目视觉技术及先进图像分析算法,实现了废料三维尺寸的快速自动检测。在双目视觉测量技术应用上,系统基于平行式双目视觉理论,通过SGBM算法实现立体匹配。与BM算法相比,SGBM算法获得了更优的匹配精度与边缘清晰度,有效提升了测量的准确性,可输出高分辨率的视差图像与点云数据,为废料尺寸的关键参数提取提供了坚实基础。试验结果表明,该系统测量误差较低,不超过2.5 mm,符合要求。A binocular vision system is proposed for industrial waste size measurement tasks,aiming to over⁃come the high cost and low efficiency of manual measurement.This system integrates precision mechanical structure,binocular vision technology,and advanced image analysis algorithms to achieve rapid and automatic detection of three-dimensional dimensions of waste materials.In the application of binocular vision measure⁃ment technology,the system is based on the theory of parallel binocular vision and achieves stereo matching through the SGBM algorithm.Compared with the BM algorithm,it achieves better matching accuracy and edge clarity,effectively improving the accuracy of measurement with high-resolution disparity images output and point cloud data,providing a solid foundation for extracting key parameters of waste size.The experiment re⁃sults show that the measurement error of the system does not exceed 2.5 millimeters.

关 键 词：双目视觉 尺寸测量 三维重建 立体匹配 

分 类 号：X705[环境科学与工程—环境工程] TP391.41[自动化与计算机技术—计算机应用技术]