Computer Engineering and Applications ›› 2007, Vol. 43 ›› Issue (15): 196-197.

• 工程与应用 • Previous Articles     Next Articles

Parameter optimization based on GA for camera calibration in stereo vision

GAO Lin-long1,HAO Kuang-rong1,DING Yong-sheng1,2   

  1. 1.College of Information Sciences and Technology,Donghua University,Shanghai 200051,China
    2.Engineering Research Center of Digitized Textile and Fashion Technology,Ministry of Education,Shanghai 201620,China
  • Received:1900-01-01 Revised:1900-01-01 Online:2007-05-21 Published:2007-05-21
  • Contact: GAO Lin-long

一种立体视觉的摄像机标定参数遗传优化方法

高霖龙1,郝矿荣1,丁永生1,2   

  1. 1.东华大学 信息科学与技术学院,上海 200051
    2.数字化纺织服装技术教育部工程研究中心,上海 201620
  • 通讯作者: 高霖龙

Abstract: In this paper,the linear equations are built based on linear models firstly.Camera parameters are evaluated by least square algorithm.Next,the nonlinear equations are joined to revise the effects of camera distortions.The GA is used to optimize the camera parameters to improve the precision and dependability of the camera calibration.At last,the experiment indicates that the method has a good effect and practicability.

Key words: camera calibration, genetic algorithm, parameter optimization, stereo vision

摘要: 首先利用线性模型构建线性方程,通过最小二乘法估算出摄像机的内、外参数,然后考虑到摄像机的畸变影响加入非线性畸变方程。利用遗传算法具有良好全局优化的特点,对摄像机参数进行优化,提高了摄像机标定的精度和可靠性。实验结果表明,该方法达到预期目的且具有很好的实用性。

关键词: 摄像机标定, 遗传算法, 参数优化, 立体视觉