Journal of South China University of Technology (Natural Science Edition) ›› 2011, Vol. 39 ›› Issue (5): 44-48,54.doi: 10.3969/j.issn.1000-565X.2011.05.008

• Electronics, Communication & Automation Technology • Previous Articles     Next Articles

Optical Flow Field Estimation Based on Illumination Radiation Model

Liao Bin1,2  Du Ming-hui1  Hu Jin-long3   

  1. 1.School of Automation Science and Engineering,South China University of Technology,Guangzhou 510640,Guangdong,China;2.Department of Electronic Science,Huizhou University,Huizhou 516007,Guangdong,China
  • Received:2010-09-17 Revised:2010-11-29 Online:2011-05-25 Published:2011-04-01
  • Contact: 廖彬(1980-),女,博士生,讲师,主要从事运动检测、数字视频分析研究 E-mail:liaobin_lb@sohu.com
  • About author:廖彬(1980-),女,博士生,讲师,主要从事运动检测、数字视频分析研究
  • Supported by:

    国家自然科学基金重点项目(U0735004);国家科技支撑计划项目(2008BAH37B08)

Abstract:

Proposed in this paper is an illumination radiation-based optical flow field model for the motion detectionof various objects in different environments. In the investigation,first,the parameters of the general dynamic imagemodel are explained and amended in the Clifford algebra region. Then,according to the results deduced in the Cliffordalgebra region and obtained from experiments,a new optical flow field model is proposed based on the principleof movement vision-generating principle and the illumination radiation theory,to which a gradient field is appliedfor the purpose of improving the accuracy. Finally,a total variation algorithm is presented for the proposed model. Experimental results show that the proposed model can achieve accurate,continuous and consistent optical flowfields for the purpose of being generally used in different illumination conditions.

Key words: optical flow field model, Clifford algebra, illumination radiation, total variation method