JISE


  [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] [15] [16] [17]


Journal of Information Science and Engineering, Vol. 34 No. 1, pp. 81-102


High-Quality Codebook Generation of Vector Quantization Using the HT-ABC-LBG Algorithm


SHU-CHIEN HUANG
Department of Computer Science
National Pingtung University
Pingtung City, 900 Taiwan
E-mail: schuang@mail.nptu.edu.tw 


    This paper proposes a vector quantization (VQ) codebook generation method for image data compression using a combined scheme of Hotelling transform (HT), the artificial bee colony (ABC) algorithm and the Linde-Buzo-Gray (LBG) algorithm. First, the grayscale image is divided into a set of non-overlapping image blocks. Each block is represented by an input vector, and these input vectors are then sorted by Hotelling transform. Second, the ABC algorithm is employed to select some of the sorted vectors to form an initial codebook. Third, this codebook serves as the input of the LBG algorithm to compute vector quantization codebook. The experimental results show that the proposed HT-ABC-LBG algorithm outperforms the FF-LBG algorithm in terms of the quality of the decompressed image and the computation time.


Keywords: vector quantization, image compression, codebook generation, Hotelling transform, artificial bee colony algorithm

  Retrieve PDF document (JISE_201801_06.pdf)