FPGA implementation of four-step genetic search algorithm

Genetic algorithm (GA) has been applied to block matching algorithm (BMA) and demonstrates positively its capability in BMA. Four-step genetic search (4GS) has been proposed. The mean square error (MSE) performance of 4GS is close to FS. The computational cost of 4GS is close to the well known three-step search (3SS). Realization of 4GS can be applied in video encoding hardware. Practical implementation issues of 4GS using FPGA are discussed. Since FPGA is a reconfigurable device, the configuration of 4GS module can be changed as frame size is changed.

[1]  Angus K. M. Wu,et al.  Four-step genetic search for block motion estimation , 1998, Proceedings of the 1998 IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP '98 (Cat. No.98CH36181).

[2]  Karl M. Guttag,et al.  A single-chip multiprocessor for multimedia: the MVP , 1992, IEEE Computer Graphics and Applications.

[3]  D. Michael Miller,et al.  The analysis of one-dimensional linear cellular automata and their aliasing properties , 1990, IEEE Trans. Comput. Aided Des. Integr. Circuits Syst..

[4]  Bing Zeng,et al.  A new three-step search algorithm for block motion estimation , 1994, IEEE Trans. Circuits Syst. Video Technol..

[5]  Ming Lei Liou,et al.  Genetic motion search algorithm for video compression , 1993, IEEE Trans. Circuits Syst. Video Technol..

[6]  Lai-Man Po,et al.  A novel four-step search algorithm for fast block motion estimation , 1996, IEEE Trans. Circuits Syst. Video Technol..

[7]  Stephen Wolfram,et al.  Universality and complexity in cellular automata , 1983 .