High-resolution wavelet-fractal compressed optical coherence tomography images.

Abstract

Three-dimensional (3D) optical coherence tomography (OCT) images could assist specialists in the diagnosis of a disease in a tissue by providing morphological information from it. Since the size of such images is usually extremely large, an appropriate image compression method can help in the storage and transmission of these images. Fractal image compression provides very high compression ratios, and discrete wavelet transform (DWT) retains frequency and spatial information in the signal. In order to combine these two techniques, fractal coding has to be performed in the wavelet domain. In this work, we propose a three-dimensional extension version of the wavelet-fractal coding algorithm. The use of 3D fractal approximation to encode 3D wavelet coefficients allows us to exploit inter- and intra-redundancy of the image sequences. The compression results of several OCT images using the 3D wavelet-fractal algorithm are evaluated qualitatively and quantitatively and are compared with the results of the two-dimensional approach. The numerical results illustrate the superior performance of 3D wavelet-fractal algorithm in terms of compression ratio.