Parallel Bacterial Foraging Optimization for Video Compression

Parallel Bacterial Foraging Optimization (PBFO), a new evolutionary soft computing tool, has been proposed for real time video compression. The convergence of BFO is very slow and its performance heavily degraded for real time processing. In BFO, best optimum value is updated after each fitness function evaluations but in PBFO best optimum value is updated after fitness function evaluations all bacteria. The PBFO is used to reduce computational time of motion estimation in video compression. In the proposed technique, the parallel computation of fitness function ensures the convergence of the optimum solution is very fast. The adaptive step size with prediction, zero motion vector and Von Neumann neighborhood topology implemented in PBFO ensures the best matching block in current frame computationally very fast. The chemotaxis and swimming step can be varied to save computational time and to enhance peak signal to noise ratio. The presented method saves computational time up to 96.59 % when compared with other published methods. The proposed method is new and computationally efficient and suitable for any real time applications.