Fuzzy vector quantization algorithms

This paper presents the development of efficient algorithms employing fuzzy logic for codebook design. These algorithms achieve the quality of vector quantizers provided by computationally demanding approaches, while capturing the advantages of the k-means algorithm, such as speed, simplicity, and conceptual appeal. The development of these algorithms is based on effective strategies for the transition from soft to crisp decisions during the clustering process. The uncertainty associated with training vector assignment is quantitatively measured by various families of membership functions, including those used in fuzzy k-means algorithms. The application of the proposed algorithms in image compression based on vector quantisation provides the basis for evaluating their computational efficiency and comparing the quality of the resulting codebook design with that provided by competing techniques