Motion estimation algorithm for periodic pattern objects based on spectral image analysis

In this paper, we propose a motion estimation algorithm for periodic pattern objects that conventional block matching based motion estimation algorithms cannot give reliable results. The proposed algorithm is based on spectral image analysis and statistical object motion calculation. The proposed algorithm doesn´t significantly increase computational complexity because it uses Fast Fourier Transform and statistical methods rather than complex object segmentation and pixel matching calculations. Experimental results show effectiveness of proposed scheme with significantly reduced defects in image processing results.