A new algorithm for motion estimation from straight-line optical flow based on ant colony algorithm

This paper presents a new linear algorithm for estimating 3D motion from 2D image sequence. The main contribution of the algorithm contains two parts: the establishment of straight-line optical flow mathematical model and the design of ant colony algorithm based on the model. With the method, we just need to match a minimum number of two straight-lines from 2D images over three frames to gain the straight-line optical flow parameters. By introducing the straight-line optical flow parameters into the straight-line optical flow model, we can obtain the 3D motion estimation by solving linear equations. Several experiments and results comparison show that our algorithm has excellent robustness and less estimation error and less time consumption.