Optical flow angular rate determination

One application of optical flow analyses is to determine the egomotion of moving bodies based on intensity variations in sequential sensor imagery. Full egomotion determination requires an estimate of the body's translational velocity as well as the angular rates. Such calculations of the full egomotion require solving a system of nonlinear equations. However, if the assumption is made that the translational velocities can be estimated by another sensor (i.e. accelerometers), the resulting equations for the angular rate prediction become linear. This paper presents the set of equations for the determination of the angular rates and range from the spatio-temporal image gradients introduced when employing the optical flow constraint. Results of simulations show typical accuracy for the proposed approach.

[1]  Juan Lopez,et al.  Performance of passive ranging from image flow , 2003, Proceedings 2003 International Conference on Image Processing (Cat. No.03CH37429).

[2]  Shahriar Negahdaripour,et al.  Determining 3-D Motion of Planar Objects from Image Brightness Patterns , 1985, IJCAI.

[3]  David J. Fleet,et al.  Performance of optical flow techniques , 1994, International Journal of Computer Vision.

[4]  Berthold K. P. Horn,et al.  Determining Optical Flow , 1981, Other Conferences.