Nb. Karayiannis et Pi. Pai, FUZZY VECTOR QUANTIZATION ALGORITHMS AND THEIR APPLICATION IN IMAGE COMPRESSION, IEEE transactions on image processing, 4(9), 1995, pp. 1193-1201
This paper presents the development and evaluation of fuzzy vector qua
ntization algorithms, These algorithms are designed to achieve the qua
lity of vector quantizers provided by sophisticated but computationall
y demanding approaches, while capturing the advantages of the frequent
ly used in practice k-means algorithm, such as speed, simplicity, and
conceptual appeal, The uncertainty typically associated with clusterin
g tasks is formulated in this approach by allowing the assignment of e
ach training vector to multiple clusters in the early stages of the it
erative codebook design process, A training vector assignment strategy
is also proposed for the transition from the fuzzy mode, where each t
raining vector can be assigned to multiple clusters, to the crisp mode
, where each training vector can be assigned to only one cluster, Such
a strategy reduces the dependence of the resulting codebook on the ra
ndom initial codebook selection, The resulting algorithms are used in
image compression based on vector quantization, This application provi
des the basis for evaluating the computational efficiency of the propo
sed algorithms and comparing the quality of the resulting codebook des
ign with that provided by competing techniques.