Quantization Effects in Pyramidal Adaptive Approximation Image Coding Techniques

R. Montúfar-Chaveznava, F. García-Ugalde, and B. Pšenička (Mexico)


Image coding, and adaptive approximation techniques


At present, adaptive approximation techniques have become very popular; in consequence, many of them have been developed, such as matching pursuit, basis pursuit and high resolution pursuit. In this work we propose an image coding model based on matching pursuit, which we named pyramidal matching pursuit. The pyramidal matching pursuit coder expands an image over a redundant dictionary, which is selected according to a best basis criterion from a set of bases. Next, the coefficients corresponding to the most important image structures are selected from the image expansion. Matching pursuit performs the selection by a similarity measure. Selected coefficients are quantized just when they are chosen in order to minimize error propagation along the process. The set of selected coefficients corresponds to the image decomposition or a new representation with a reduced number of elements. A simple reconstruction algorithm recovers the original image with a high visual quality. As a consequence of the work carried out, an image coding model is presented, which is as integral as possible, and produces very good visual quality results, with a high compression rate.

Important Links:

Go Back