International Journal of Computer Applications |
Foundation of Computer Science (FCS), NY, USA |
Volume 46 - Number 18 |
Year of Publication: 2012 |
Authors: Vikash Kumar, Jitu Sharma, Shahanaz Ayub |
10.5120/7043-9486 |
Vikash Kumar, Jitu Sharma, Shahanaz Ayub . Image Compression using FFN for ROI and SPIHT for background. International Journal of Computer Applications. 46, 18 ( May 2012), 30-35. DOI=10.5120/7043-9486
The objective of this paper is to develop different method for image compression without reducing the resolution for Region of Interest in the images. Mostly medical images for diagnosis purpose have some specific region; this region is known as Region of Interest (ROI). Here we separate our ROI from background and then apply Feed Forward Neural Network (FFN) compression to ROI and background is compressed using Set Partitioning in Hierarchical Tree (SPIHT) Technique. This paper is in the context to understand the FFN technique which provides low compression by maintaining images resolution of ROI whereas SPIHT technique which gives very high compression ratio for the background. Thus the combination of compressed ROI and compressed background will result in an image with highly compressed background without effecting image resolution of ROI.