论文部分内容阅读
设计出一个基于机器视觉中点结构光测量原理和以CMOS图像传感器局部感兴趣窗口(ROI)功能为基础的动态姿态角测量系统。介绍了系统测量原理、系统总体设计、以及软硬件算法的设计及实现。该系统硬件平台包括点结构光光源和姿态敏感器,实现了CMOS图像传感器的ROI功能,大大减少了图像数据量,提高了帧频。同时,采用FPGA实现了并行流水的图像预处理。基于序贯四连通域算法和一阶矩计算的实时光斑质心定位,将硬件处理结果实时传输给软件算法进行数据再处理,减少了软硬件平台间交互的数据量,提高了系统测量速度。软件平台实现了系统的标定和测试,并给出了结果。实验结果表明:该测量系统测量范围大,测角速度快,较好地实现了非接触式的动态测量。
A dynamic attitude measurement system based on the mid-point structure light measurement principle of machine vision and the ROI (local interest window) of CMOS image sensor is designed. The principle of system measurement, the overall system design, and the design and implementation of hardware and software algorithms are introduced. The system hardware platform includes point structure light source and attitude sensor, which realizes the ROI function of CMOS image sensor, greatly reducing the amount of image data and increasing the frame rate. At the same time, using FPGA to achieve parallel flow of image preprocessing. Based on the sequential four-connected domain algorithm and the first-order moment centroid localization, the hardware processing results are transmitted to the software algorithm in real time for data re-processing, which reduces the amount of data exchanged between the software and hardware platforms and improves the system measurement speed. Software platform to achieve the system calibration and testing, and gives the results. The experimental results show that the measurement system has a large measuring range and fast angular velocity, and the non-contact dynamic measurement is well realized.