Vector quantization of images using fractal dimensions

Abstract

In conventional vector quantization (VQ), for example, generalized Lloyd algorithm (GLA), an image is divided into blocks that are all the same size. This uniform division could be redundant. Furthermore, it could not attain both a high compression rate and high quality of encoded image. We propose a new method of VQ in which the block size to divide an image is determined by a local fractal dimension (LDF). Computational experiments were carried out to show the effectiveness of the method. Results of experiments showed that a compression rate of the proposed method is higher than that by the GLA under the condition that PSNR (Peak Signal-to-Noise Ratio) is more than 35.0 dB. Therefore, the proposed method is useful for practical image compression.