An Approach to Better Portable Graphics (BPG) Compression with Providing a Desired Quality
Fangfang Li, Sergey Krivenko, Vladimir Lukin
Abstract
Image compression is getting more importance for numerous applications like high resolution cameras, computer vision devices, portable instruments, remote sensing imagers, etc. To be useful for practice, an image compression method has to meet several requirements including high compression rate and visual quality, speed of compression and decompression, etc. A novel perspective approach called Better Portable Graphics (BPG) has been proposed recently. Although it aims at providing a desired visual quality using a parameter Q, accuracy of producing this quality is still worth improving. An approach proposed in this paper employs a two-step procedure based on averaged characteristics of the BPG image compression encoder. Experiments have been carried out for a set of different grayscale images. The experimental results show that the proposed approach is able to quickly and efficiently provide a desired visual quality of BPG compressed images with considerably improved accuracy.