The learning vector quantization (LVQ) algorithm is widely used in image compression because of its intuitively clear learning process and simple implementation. However, LVQ strongly depends on the initialization of the codebook and often converges to local optimal results. To address the issues, a new two-step LVQ (TsLVQ) algorithm is proposed in the paper. TsLVQ uses a correcting learning stage after LVQ to move the synaptic weight vector away from the incorrectly clustered training vector and towards the correctly clustered training vector. Experimental results show that TsLVQ outperforms kernel-based LVQ (KLVQ) and LVQ in terms of peak signal-to-noise ratio.
BracciniCCocurulloFLavagettoF (1995) A fast algorithm for high quality vector quantization codebook design. Lecture Notes in Computer Science974: 643–648.
3.
ChatterjeeSSreenivasTV (2008) Optimum switched split vector quantization of LSF parameters. Signal Processing88(6): 1528–1538.
4.
CovõesTFHruschkaERGhoshJ (2013) Competitive learning with pairwise constraints. IEEE Transactions on Neural Networks and Learning Systems24(1): 164–169.
5.
DaiQChenSC (2006) Integrating the improved CBP model with kernel SOM. Neurocomputing69: 2208–2216.
6.
FengHHChenCYYeF (2007) Evolutionary fuzzy particle swarm optimization vector quantization learning scheme in image compression. Expert Systems with Applications32: 213–222.
7.
Gómez-RuizJAMuñoz-PerezJLópez-RubioE. (2001) Expansive and competitive neural networks. Lecture Notes in Computer Science2084: 355–362.
8.
Gómez-RuizJAMuñoz-PerezJLópez-RubioE. (2002) Expansive and competitive learning for vector quantization. Neural Processing Letters15(3): 261–273.
9.
HanaCCChenbYNLocCC. (2007) A novel approach for vector quantization using a neural network, mean shift, and principal component analysis-based seed re-initialization. Signal Processing87(5): 799–810.
10.
HorngMH (2009) Honey bee mating optimization vector quantization scheme in image compression. AICI, Lecture Notes in Computer Science5855: 185–194.
KohonenT (1984) Self-organization and Associative Memory. Berlin: Springer-Verlag.
13.
KohonenT (1986) Learning vector quantization for pattern recognition. Technical Report TKK-F-A601, Helsinki Institute of Technology, Finland.
14.
KohonenT (1990) Improved versions of learning vector quantization. Proceedings of the IEEE International Joint Conference on Neural NetworksI: 545–550.
15.
KohonenT (1991) Learning vector quantization for pattern recognition. Report TKK-F-A601. Helsinki University of Technology, Finland.
16.
KrishnaaKMurtybVLNRamakrishnanKR (2001) Vector quantization of excitation gains in speech coding. Signal Processing81(1): 203–209.
17.
LaskarisNAFotopoulosS (2004) A novel training scheme for neural-network based vector quantizers and its application in image compression. Neurocomputing61(1): 421–427.
18.
LindeYBuzoAGrayRM (1980) An algorithm for vector quantizer design. IEEE Transactions on Communications28: 84–94.
19.
LughoferE (2008) Extensions of vector quantization for incremental clustering. Pattern Recognition41(3): 995–1011.
20.
MacQueenJB (1967) Some methods for classification and analysis of multivariate observations. Proceedings of 5th Berkeley Symposium on Mathematical Statistics and Probability. Berkeley, CA: University of California Press, volume 1, pp. 281–297.
21.
MaoJCJainAK (1996) A self-organizing network for hyperellipsoidal clustering (HEC). IEEE Transactions on Neural Networks7(1): 16–29.
22.
Mérida-CasermeiroELópez-RodríguezDGalán-MarínG. (2007) Improved production of competitive learning rules with an additional term for vector quantization. Lecture Notes in Computer Science4431: 461–469.
23.
PalNRBezdekJCTsaoEC (1993) Generalized clustering networks and Kohonen’s self-organizing scheme. IEEE Transactions on Neural Networks4 (4): 549–557.
24.
PatanéGRussoM (2001) The enhanced LBG algorithm. Neural Networks14(9): 1219–1237.
25.
PedryczWHirotacK (2007) Fuzzy vector quantization with the particle swarm optimization: a study in fuzzy granulation–degranulation information processing. Signal Processing87(9): 2061–2074.
26.
ShenFHasegawaO (2006) An adaptive incremental LBG for vector quantization. Neural Networks19(5): 694–704.
27.
UchiyamaTArbibMA (1994) Color image segmentation using competitive learning. IEEE Transactions on Pattern Analysis and Machine Intelligence16(2): 1197–1206.
28.
UedaNNakanoR (1994) A new competitive learning approach based on an equidistortion principle for designing optimal vector quantizers. Neural Networks7(8): 1211–1227.
29.
WangYFengXYHuangYX. (2007) A novel quantum swarm evolutionary algorithm and its applications. Neurocomputing70: 633–640.
30.
XuLKrzyzakAOjaE (1993) Rival penalized competitive learning for clustering analysis, rbf net, and curve detection. IEEE Transactions on Neural Networks4(4): 636–649.
31.
YairEZegerKGershoA (1992) Competitive learning and soft competition for vector quantizer design. IEEE Transactions on Signal Processing40(2): 294–308.
