基于激光三角法的线结构激光瓷砖铺贴平整度检测技术研究

发布时间：2018-05-01 06:23

本文选题：结构光 + 线激光　；参考：《东华大学》2017年硕士论文

【摘要】：随着我国国民经济迅猛发展,建筑行业对建造速度要求越来越快。而瓷砖铺贴仍局限于手工铺贴的方式,而瓷砖平整度的验收也依赖于人工操作,使得瓷砖的铺贴与验收成为阻碍建筑速度提高的一个重要因素。而瓷砖铺贴平整度的检测是实现瓷砖铺贴自动化与瓷砖平整度自动化验收的重要技术,本文针对该问题,提出了基于线结构激光的瓷砖铺贴平整度检测方法,主要进行如下几个方面的工作:1.根据课题的应用场景,提出了可实现的系统设计方案,完成了支撑结构、激光发生器、相机、滤镜、计算平台等的设计与选型工作,并完成了软件的流程设计、原型研发。2.研究了线结构激光三角检测法的关键技术,包括相机标定技术、图像预处理技术、条纹中心提取技术、物像点转换技术,并确定了相应的技术路线。其中,相机标定技术选用了张正友标定法与PnP标定法;图像预处理中采用最大类间方差法(OSTU)和灰度属性邻域法;条纹中心提取基于自适应模板法;物像点转换则基于相机的成像模型推导了像点到物点的转换关系。3.本文在条纹中心提取中,对可变方向模板法的基础进行改进,通过模板自适应处理,提出了自适应方向模板法。该方法提高了在光条宽度发生变化时提取结果的鲁棒性,同时提升了处理速度。4.设计了实验对检测装置的性能进行评估。通过实验证明了系统方案的可行性与可靠性,并通过试验确定了系统对线结构光入射角度的要求,以及在选用不同大小的模板与高斯参数时对自适应方向模板法提取结构光中心的影响,给出了相应的参数设置建议。
[Abstract]:With the rapid development of our national economy, the construction industry requires faster and faster construction speed. However, tile paving is still limited to manual paving, and the acceptance of tile flatness also depends on manual operation, which makes tile paving and acceptance become an important factor that hinders the improvement of building speed. The detection of tile flatness is an important technology to realize the automation of tile paving and the automatic acceptance of tile flatness. In this paper, a method of detecting tile flatness based on line structure laser is put forward in order to solve this problem. Work in the following areas: 1. According to the application scene of the subject, the design scheme of the system is put forward, and the design and selection of the supporting structure, laser generator, camera, filter and computing platform are completed, and the flow design of the software is completed, and the prototype R & D. 2. 2. The key techniques of linear laser triangulation detection are studied, including camera calibration, image preprocessing, fringes center extraction, image point conversion, and the corresponding technical route is determined. The camera calibration techniques include Zhang Zhengyou calibration method and PnP calibration method, maximum inter-class variance method and gray attribute neighborhood method in image preprocessing, and stripe center extraction based on adaptive template method. On the other hand, based on the camera's imaging model, the conversion relationship between image points and object points is derived. In this paper, the basis of variable direction template method is improved in the extraction of fringe center, and the adaptive direction template method is proposed through adaptive template processing. This method improves the robustness of the extraction results when the width of the optical strip changes, and improves the processing speed of .4. Experiments were designed to evaluate the performance of the device. The feasibility and reliability of the system scheme are proved by experiments, and the requirements of the system for the incident angle of linear structured light are determined by experiments. The influence of different sizes of templates and Gao Si parameters on the extraction of structured light centers by adaptive directional template method is also presented, and the corresponding parameter setting suggestions are given.
【学位授予单位】：东华大学
【学位级别】：硕士
【学位授予年份】：2017
【分类号】：TU767;TN247

【相似文献】