成像跟踪系统中摄像机标定算法研究及DSP实现-信号与信息处理专业论文.docxVIP

摘要 摘 要 摄像机标定技术是完成实时成像跟踪系统的重要基础，建立摄像机成像模型 和求解摄像机参数是摄像机标定的两个主要研究内容。本文在深入研究角点检测 算法、摄像机标定算法的基础上，提出了一种新的基于菱形模板的摄像机标定算 法，并搭建了基于 TMS320DM642 的摄像机标定硬件平台，完成了算法的 DSP 实现。本文的主要研究内容如下： (1) 对摄像机标定的成像理论包括摄像机成像线性模型、非线性模型和摄像机 参数进行了研究。 (2) 对角点检测算法的研究现状进行了分析，并重点研究了基于 Harris 的亚像 素角点检测算法，完成了算法的 MATLAB 仿真和 C 语言的程序实现。 (3) 在深入研究基于平面模板的摄像机标定算法基础上，提出了基于菱形模板 的摄像机标定算法，完成了算法的 C 语言实现，并通过分析两种方法的 MATLAB 仿真实验结果，总结了算法的优缺点。 (4) 建立了摄像机标定的硬件平台，完成了角点检测算法和摄像机标定算法的 DSP 实现，为成像跟踪系统的搭建打下了基础。 通过仿真计算和测试，表明本文方法是可行的，能基本满足实时系统的要求。 关键词：成像跟踪，摄像机模型，角点检测，摄像机标定，DSP 实现 I ABSTRACT ABSTRACT Camera calibration technology is the basis of real-time imaging tracking system, including the establishment of the camera imaging model, and solving the camera parameters. This dissertation presents a method of camera calibration using rhombus template, which is based on the research of the corner detection algorithm and the camera calibration algorithm, and builds a hardware platform based on TMS320DM642, complete the realization of the DSP of the corner detection algorithm and camera calibration algorithm; The main contents are as follows: Research the camera calibration theory, including the linear model, nonlinear model and camera parameters. Research the current sisution of the corner detection algorithm, focus on sub-pixel corner detection algorithm based on Harris, and completes the algorithm MATLAB simulation and the C language program. Present a method of camera calibration using rhombus template, which is based on the study of planar template-based camera calibration algorithm theory, complete the algorithm in C language, and through the analysis of the MATLAB simulation results of the two methods, obtain the advantages and disadvantages of the algorithm. Establish the hardware platform of calibration, complete the realization of the DSP of the corner detection algorithm and camera calibration algorithm; establish the foundation for the imaging tracking system. Though the simulation and test, the results show that the algor