基于视觉的姿态测量方法设计
首发时间:2017-12-27
摘要:提出了一种运用视觉来进行姿态测量的方法,设计了一种视觉算法通过固定的矩形合作物来获得相机姿态。对本文所使用的Kinect相机进行了标定得到了相机的内参数和畸变参数。然后对相机采集到的靶标图像进行灰度化、二值化、中值滤波去噪等预处理,然后利用ORB对预处理后的图像进行特征提取,对提取到的角点用K-means进行角点聚类,得到矩形到的四个顶点,最后利用PNP算法对相机姿态进行解算得到了相机的姿态。在三轴旋转平台上进行了实验验证,实验表明本文提出到的方法能够估计相机的姿态,为系统的姿态测量提供了一种新的方法。
For information in English, please click here
Design of attitude measurement method based on vision
Abstract:A kind of attitude measurement method based on vision is put forward by using cooperationof the rectangular target fixed on the blackboard in this paper. The internal and distortion parameters of the camera are obtained by calibrating the Kinect camera. The image acquired by the camera, after aseries of preprocess of grayscale, binarization, median filtering denoising, and the using ORB algorithm to extract the feature points of the rectangular region. The four corner points of the rectangular can be extracted by the K-Means clustering algorithm, and then the camera coordinates in the world coordinate system can be solved by PNP algorithm. Experimental verification on the three axis rotating platformshows that this method can estimate the attitude of the camera, and provides a new method for the attitude measurement of the system.
Keywords: Computer vision ORB algorithm PNP problem Attitude measurement
基金:
引用
No.****
同行评议
勘误表
基于视觉的姿态测量方法设计
评论
全部评论0/1000