An FFT-Based Full-Search Block-Matching Algorithm With SSD Criterion
Zhen Li, Atsushi Uemura, Hitoshi Kiya · Hokkaido University Collection of Scholarly and Academic Papers (Hokkaido University) · 2009
An FFT-based full-search block-matching algorithm is described that uses the sum of squared difference (SSD) criterion. The proposed method focuses on the relationship between the circular cross-correlation and the SSD criterion. Because FFT is used to calculate the cross-correlation between signals of different sizes, the processing speed of block matching is greatly increased. If the macroblock is composed of real signals, two blocks can be matched at the same time. In a simulation of motion estimation, the proposed method achieved the same performance as a direct SSD full search, but with a processing speed 8 to 700 times higher.