Title :
Codebook design algorithm for classified vector quantization based on fuzzy clustering
Author :
Supot, Sookpotharom ; Manas, Sangworasil
Author_Institution :
Dept. of Electr. Eng., Bangkok Univ., Pathumtani, Thailand
Abstract :
Classified vector quantization (CVQ) is used for coding images that achieves good perceptual results while reducing the computational load of the process. In this paper, image is sub-divided into 4×4 pixel blocks (vectors). Each vector is classified into an edge vector and a shade vector. Both edge vectors and shade vectors are used to design the codebooks of CVQ by Fuzzy C-Means (FCM) method. By doing so, the CVQ-FCM method can preserve the edge of image, make good image quality, and reduce the processing time while constructing the codebooks.
Keywords :
fuzzy set theory; image classification; image coding; image segmentation; vector quantisation; Fuzzy C-Means; classified vector quantization; codebook design algorithm; computational load; edge vector; fuzzy clustering; image coding; image quality; image segmentation; pixel blocks; shade vector; Algorithm design and analysis; Clustering algorithms; Degradation; Image coding; Image quality; Information technology; Iterative algorithms; Pixel; Speech coding; Vector quantization;
Conference_Titel :
Industrial Technology, 2002. IEEE ICIT '02. 2002 IEEE International Conference on
Print_ISBN :
0-7803-7657-9
DOI :
10.1109/ICIT.2002.1189260
