基于散斑结构光与双目视觉的盾构出渣量实时检测方法 .pdfVIP

基于散斑结构光与双目视觉的盾构出渣量实时检测方法 西安中铁工程装备有限公司 陕西西安 710200 摘要：针对盾构施工过程中出渣量难以实时检测的难题，提出一种基于散斑 结构光与双目视觉的盾构出渣量实时检测方法。首先将散斑结构光投影至渣土表 面以增加特征信息，再采用半全局立体匹配算法快速获取渣土表面点云数据。然 后提出一种基于层次 B 样条的渣土表面三维点云插值算法，解决了渣土表面点云 数据缺失的问题。最后利用采集到的点云数据对渣土表面轮廓进行了三维重构和 渣土体积计算。该检测方法应用于西安市某地铁盾构区间的工程现场试验，结果 表明：测得的渣土体积与实际渣土体积之间的最大相对误差为 4.6%，满足实际施 工的精度要求 。 关键词：双目视觉；三维重构；渣土体积计算；工程现场试验 中图分类号：TP391.4 文献标识码：A Real-time detection method of shield slag discharge based on speckle structured light binocular vision ZHANG Xin, WEI Mingle （Xi’an China Railway Engineering Equipment Co., Ltd, Xi’an 710200, China） Abstract: Aiming at the problem of real-time detection of slag discharge in shield construction, a real-time detection method of slag discharge of shield based on speckle structured light binocular vision is proposed. Firstly, speckle structured light is projected onto the surface of the muck to increase the feature information, and then the semi-global stereo matching algorithm is used to quickly obtain the point cloud data of the muck surface. Then, a 3D point cloud interpolation algorithm based on hierarchical B-spline is proposed to solve the problem of missing point cloud data on the surface of muck. Finally, the 3D reconstruction of the surface contour of muck and the calculation of muck volume are carried out by using the collected point cloud data. This detection method is applied to the field test of a subway shield section in Xi an, and the test results show that the maximum relative error between the muck volume measured and the actual muck volume is 4.6%, which meets the accuracy requirements of actual construction. Key words: speckle structured light; binocular vision; 3D reconstruction; calculation of muck volume; engineering field test 0 引言 在土压平衡盾构施工过程中，出渣量的检测对提高施工精细化程度和施工安 全预警有着重要的作用。盾构机向前掘进时，掌子面前方土体偶尔会存在空洞、 地下河等不良地质条件，而出渣量的多少可以反映掌子面前方土体的异常情况。 为了防止地层变形对盾构机造成损坏，需通过调