Abstract
In this paper, a new multi stage vector quantization with energy clustered training set is proposed for color image coding. The input image is applied with orthogonal polynomials based transformation and the energy clustered transformed training vectors are obtained with reduced dimension. The stage-by-stage codebook for vector quantization is constructed from the proposed transformed training vectors so as to reduce computational complexity. This method also generates a single codebook for all the three color components, utilizing the inter-correlation property of individual color planes and interactions among the color planes due to the proposed transformation. As a result, the color image encoding time is only slightly higher than that of gray scale image coding time and in contrast to the existing color image coding techniques, whose time is thrice greater than that of gray scale image coding. The experimental results reveal that only 35 % and 10 % of transform coefficients are sufficient for smaller and larger blocks respectively, for the reconstruction of images with good quality. The proposed multi stage vector quantization technique is faster when compared to existing techniques and yields better trade-off between image quality and block size for encoding.
Similar content being viewed by others
References
Adams MD, Kossentini F (2000) Reversible integer-to-integer wavelet transforms for image compression: performance evaluation and analysis. IEEE Trans Image Process 9:1010–1024
Annadurai S, Sundaresan M (2009) Wavelet based enhanced color image compression relying on sub-band vector quantization. ICGST-GVIP J 9:9–16
Barlaud M, Sole P, Gaidon T, Antoni M, Mathieu P (1994) Pyramidal lattice vector quantization for multiscale image coding. IEEE Trans Image Process 3:367–381
Canta GR, Poggi G (1998) Kronecker-product gain-shape vector quantization for multispectral and hyperspectral image coding. IEEE Trans Image Process 7:668–678
Chang C-C, Li Y-C, Yeh J-B (2006) Fast codebook search algorithms based on tree-structured vector quantization. Pattern Recognit Lett 27:1077–1086
Courant R, Hilbert D (1975) Methods of mathematical physics, 1st edn. Wiley Eastern, New Delhi
Equitz WH (1989) A new vector quantization clustering algorithm. IEEE Trans Accoust Speech Signal Process 37:1568–1575
Esakkirajan S, Veerakumar T, Senthil Murugan V, Sudhakar R (2006) Image compression using contourlet transform and multi stage vector quantization. GVIP J 6:19–28
Fischer TR (1986) A pyramid vector quantizer. IEEE Trans Inf Theory 32:568–583
Flanagant JK, Morrell DR (1989) Vector quantization codebook generation using simulated annealing. IEEE Int Conf on Acoust Speech and Signal Process 3:1759–1762
Gersho A, Gray RM (1992) Vector quantization and signal compression. Kluwer Academic Press/Springer
Goldberg M (1986) Image compression using adaptive vector quantization. IEEE Trans Commun 34:180–187
Hsieh C-H (1992) DCT-based codebook design for vector quantization of images. IEEE Trans Circuits Syst Video Technol 2:401–409
Hsieh C-H, Shao W-Y, Jing M-H (2000) Image compression based on multistage vector quantization. J Visual Comm and Image Represent 11:374–384
Huang H-C, Pan J-S, Zhe-Ming L, Sun S-H, Hang H-M (2001) Vector quantization based on genetic simulated annealing. Signal Process 81:1513–1527
Kim JW, Lee SU (1992) A transform domain classified vector quantizer for image coding. IEEE Trans Circuits Syst Video Technol 2:3–14
Krishnamoorthy R, Kannan N (2009) Codebook generation for vector quantization on orthogonal polynomials based transform coding. Int J Signal Process 5:67–73
Krishnamoorthy R, Punidha R. FITVQSPC: Fast and improved transformed vector quantization using static pattern clustering. Int Conf on Signal Process Image Process and Pattern Recog (2010) 146–155
Krishnamoorthy R, Punidha R. FVQEOPT: Fast vector quantization encoding with orthogonal polynomials transform. Int Conf on Machine Vis. (2010) 92–96
Lai JZC, Liaw Y-C (2009) A novel encoding algorithm for vector quantization using transformed codebook. Pattern Recognit 42:3065–3070
Lai JZC, Liaw Y-C, Liu J (2008) A fast VQ codebook generation algorithm using codeword displacement. Pattern Recognit 41:315–319
Li RY, Kim J, Al-Shamakhi N (2002) Image compression using transformed vector quantization. Image and Vision Comput 20:37–45
Linde Y, Buzo A, Gray RM (1980) An algorithm for vector quantizer design. IEEE Trans Commun 28:84–94
Liou R-J, Wu J (2008) Quadtree image compression using sub-band DCT features and kohonen neural networks. IEEE Int Conf on Audio Lang and Image Process. 252–256
Wang M, Ma H-P, Zhou C-Q, Yang B (2007) An improved multi-stage vector quantization for image coding. IEEE 3rd Int. Conf. on Intel. Inf. Hiding and Multimed Signal Process. 415–420
Nasrabadi NM, Feng Y (1990) Image compression using address vector quantization. IEEE Trans Comm 38:2166–2173
Pan JS, McInnes FR, Jack MA (1995) VQ codebook design using genetic algorithms. Electronics Lett 31:1418–1419
Salleh MFM, Soraghan J (2007) A new multistage lattice vector quantization with adaptive subband thresholding for image compression. EURASIP J Appl Signal Process 1:1–11
Shen E, Hasegawa O (2006) An adaptive incremental LBG for vector quantization. Neural Netw 19:694–704
Shen G, Zeng B, Liou M-L (2003) Adaptive vector quantization with codebook updating based on locality and history. IEEE Trans Image Process 12:283–295
Sun H, Lam K-Y, Chung S-L, Dong W, Ming G, Sun J (2005) Efficient vector quantization using genetic algorithm. Neural Comput Appl 14:203–211
Swilem A (2010) A Fast vector quantization encoding algorithm based on projection pyramid with hadamard transformation. Image Vision Comput 28:1637–1644
Tsai W, Lee C-Y (2009) A fast VQ codebook generation algorithm via pattern reduction. Pattern Recognit Lett 30:653–660
Yang S-B (2008) Constrained-storage multistage vector quantization based on genetic algorithms. Pattern Recognit 41:689–700
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Krishnamoorthy, R., Punidha, R. Low bit-rate multi stage vector quantization based on energy clustered training set. Multimed Tools Appl 70, 2293–2308 (2014). https://doi.org/10.1007/s11042-012-1244-4
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11042-012-1244-4