基于点激光和双目视觉的货物尺寸测量的研究

发布时间：2018-05-03 06:30

本文选题：物流货物 + 点激光测距　；参考：《大连理工大学》2016年硕士论文

【摘要】：随着计算机硬件和图像处理技术的快速发展,计算机视觉技术逐渐走向实用化,如在当前的航天遥测、生物医学、智能导航、虚拟现实、农业产品分级、军事侦察、零件测量和物联网等社会生活多个领域中计算机视觉技术都开始体现出自己的价值和重要作用。而计算机视觉领域研究的一个热点——双目立体视觉技术更是受到各个领域中的研究人员的广泛关注。当前物流业在国内快速发展,如何能够快速准确对物流货物和包裹进行非接触测量已成为难点。因此本文以双目立体视觉技术为基础,对相关技术进行了深入的研究和讨论,分析了当前这些技术中存在的问题,设计并实现了一套针对物流货物物体的测量模型,提出了结合点激光源进行规则货物物体三维测量的具体模型和技术方法。本文在深入研究摄像机成像原理及相关理论的基础上,详细分析和阐述了摄像机的各种坐标系及各种坐标系之间的变换关系,并介绍了摄像机标定技术和利用标靶图像进行标定和求解的过程。通过理论分析,并结合物流业的特点,建立了非接触的物体测量模型,还对极线校正的内涵和算法进行深入研究,提出了一种极线约束条件下基于几何投影相关性的图像立体匹配方法。其理念是根据双目立体视觉成像的原理,寻找空间物体点在左右摄像机中的成像点对在几何投影上存在的相关性,并依据几何投影的相关特性,通过对图像的相关处理和引入点激光辅助手段来建立起图像点对的几何投影关联关系,并设计了匹配算法。论文在双目立体视觉技术的基础上,结合点激光测距的相关理论和技术,设计了物流货物物体三维尺寸的测量模型和计算方法,并在算法上考虑了实际应用中的问题。对传统的关键特征点定位和匹配方法进行了修正,避免了对整幅图像里的所有点进行查找,大大提高了计算效率。通过大量实验验证,利用点激光获得深度来定位匹配的图像特征点的方法有效提高了测量精度,为实际应用奠定了基础。同时通过实验验证,本文提出的模型实现了实际中对规则物流货物尺寸测量快速准确的目标。
[Abstract]:With the rapid development of computer hardware and image processing technology, computer vision technology is becoming more and more practical, such as space telemetry, biomedicine, intelligent navigation, virtual reality, agricultural product classification, military reconnaissance, etc. In many fields of social life, such as parts measurement and Internet of things, computer vision technology has begun to embody its own value and important role. Binocular stereo vision technology, a research hotspot in the field of computer vision, has been paid more and more attention by researchers in various fields. With the rapid development of logistics industry in China, how to measure goods and packages quickly and accurately has become a difficult problem. Therefore, based on binocular stereo vision technology, this paper deeply studies and discusses the related technologies, analyzes the problems existing in these technologies, and designs and implements a set of measurement models for logistics goods objects. A concrete model and technical method for 3D measurement of regular cargo object by combining point laser source are presented. Based on the research of camera imaging principle and related theories, this paper analyzes and expounds the transformation relationship between various coordinate systems and coordinate systems in detail. The camera calibration technology and the process of calibrating and solving using target image are introduced. Based on the theoretical analysis and the characteristics of the logistics industry, a non-contact object measurement model is established, and the intension and algorithm of pole line correction are deeply studied. An image stereo matching method based on geometric projection correlation is proposed. According to the principle of binocular stereo vision imaging, the idea is to find the correlation between the imaging points of space objects in the left and right camera and the geometric projection, and according to the related characteristics of the geometric projection. Based on the image correlation processing and the introduction of point laser auxiliary method, the geometric projection correlation relation of image point pair is established, and a matching algorithm is designed. On the basis of binocular stereo vision technology, combining the theory and technology of point laser ranging, the measurement model and calculation method of 3D size of logistics cargo object are designed, and the practical application problems are considered in the algorithm. The traditional key feature point location and matching methods are modified to avoid finding all points in the whole image and greatly improve the computational efficiency. Through a large number of experiments, the method of using point laser to obtain depth to locate the matching image feature points can effectively improve the measurement accuracy and lay a foundation for practical application. At the same time, through experimental verification, the model proposed in this paper achieves the goal of fast and accurate measurement of the size of regular logistics goods in practice.
【学位授予单位】：大连理工大学
【学位级别】：硕士
【学位授予年份】：2016
【分类号】：TP391.41

【相似文献】