Colour in Context
Research group
Computer Vision Center

Image compression


The aim of this work is to apply perceptual concepts to define a perceptual pre-quantizer (HiSET-CIWaM) in order to improve image compression algorithms. The approach consist in quantiazing multiresolution wavelet transform coefficients using some of the human visual system behaviour properties. Noise is fatal to image compression performance, because it can be both annoying for the observer and consumes excessive bandwidth during image transmission. Perceptual prequantization reduces unpercievable details and thus improve both visual appearance and transmission properties. The comparison between JPEG2000 without and with percetual pre-quantization shows that the latter is not favorable in PSNR, but the recovered image is more compressed at the same or even better visual quality measured with a weighted PSNR. Perceptual criteria were taken from the CIWaM (Chromatic Induction Wavelet Model).


Examples

(Click on images to see them in original resolution)

Original image
JPEG2000, 0.4 bpp

HiSET-CIWaM, 0.4 bpp


Original image
JPEG2000, 0.5 bpp

HiSET-CIWaM, 0.5 bpp



Original image
JPEG2000, 0.22 bpp

HiSET-CIWaM, 0.22 bpp







  2008 Colour in context Group | Computer Vision Center. All rights reserved | Contact webmaster |  Last updated: Monday 11 May 2009     eXTReMe Tracker