A Fast Context -Based Fractal Mobile Video Compression with GA and PSO
- DOI
- 10.2991/ictcs-14.2014.29How to use a DOI?
- Keywords
- fractal video compression; image segmentation; GA; PSO
- Abstract
Mobile video compression is an important method to translate multimedia from server to client, especially mobile video compress with fractal model is a good way to solution the unbalance computer capacity. Because computer capacity of servers are so powerful that complete encode of fractal video compress for server-end, comparing the mobile phone is so poor that decode of fractal video compress for client-end. In this paper, basing on the segment, we present GA to improve the quality of video compress through adjust the fixed parameter to suitable value in fractal compress algorithm. For improving speed, the paper describes PSO to speed up the fractal video compress. PSO method could find the best domain when we face huge samples to search. Compare the GA method, as it does not need the crossover and mutation, the speed of convergence is clear faster when we use PSO method to find the best G. Another issue is that we couldn’t ensure that the new crossover and mutation must follow the field the original Domains. If that, we should add more samples to domains. Experiments shows that the PSO could better match the requirement of searching the best domain comparing GA method on the premise of guaranteeing the decoded image’s quality.
- Copyright
- © 2014, the Authors. Published by Atlantis Press.
- Open Access
- This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).
Cite this article
TY - CONF AU - Junshe Wan AU - Linru You PY - 2014/05 DA - 2014/05 TI - A Fast Context -Based Fractal Mobile Video Compression with GA and PSO BT - Proceedings of the 2nd International Conference on Teaching and Computational Science PB - Atlantis Press SP - 112 EP - 115 SN - 1951-6851 UR - https://doi.org/10.2991/ictcs-14.2014.29 DO - 10.2991/ictcs-14.2014.29 ID - Wan2014/05 ER -