摘要
在FIRA MiroSot足球机器人系统中,足球机器人图像处理系统通过视频采集设备捕获有效图像信息并处理,为决策系统提供重要信息来源。通常情况下高速摄像头实时采集比赛场中的图像,存在采集图像速率低、畸变矫正效果不够明显等问题。因此提出了基于单目视觉的足球机器人图像处理系统的畸变矫正方法,即通过两步法进行摄像头标定修正,用最小二乘法计算出畸变系数拟合误差。实验证明,该方法有效提高了系统的整体性能。
The soccer robot image processing system captured the effective image information through the video capture device and processed it,which provided important information sources for the decision-making system. Usually,when the high-speed camera collected the images in the competition field in real time,there were some problems such as low image acquisition rate,distortion correction effect was not obvious. It presented a distortion correction method of soccer robot image processing system based on monocular vision in this paper. The camera calibration was corrected by two steps,and the distortion coefficient was calculated by the least square method,fitting the error. The experimental results showed that the method can effectively improve the overall performance of the system.
引文
[1]廖华丽,王廷旗,傅湘国. MiroSot足球机器人系统的构成分析[J].河海大学常州分校学报,2007,21(2):32-35.
[2]倪天伟,林金珠. Robo Cup3D仿真中双足机器人路径规划的最短切线路径算法[J].蚌埠学院学报,2017,6(3):1-3.
[3]叶峰,王敏,陈剑东,等.共面点的摄像机非线性畸变校正[J].光学精密工程,2015,23(10):2962-2970.
[4] TSAI R Y. A versatile camera calibration technique for high accuracy 3D machine vision metrology using off-theshelf TV cameras and lenses[J]. IEEE Journal of Robotics Automation,1987,3(4):323-344.
[5]征宇,黄诗捷,罗川,等.基于共面条件的摄像机非线性畸变自校正[J].光学学报,2012,32(1):0115002-6.
[6]杨必武,郭晓松.摄像机镜头非线性畸变校正方法综述[J].中国图象图形学报,2005,10(3):269-274.
[7]陶彬彬.基于单目视觉的服务机器人关键技术研究[D].南京:东南大学,2010.
[8]JIANG Dazhi,YU Qian,WANG Binyang,et al. Research and overview of imaging nonlinear distortion in computer vision[J]. Computer Engineering,2001,27(12):108-110.
[9]郁春潮.广角镜头的摄像机非线性标定技术[J].长江大学学报(自科版),2006,3(2):25-27.
[10]倪天伟,江红,林金珠.基于改进人工势场法的移动机器人避障路径规划算法[J].常州大学学报(自然科学版),2016,28(5):74-77.