Design of vector quantizer for image compression using self-organizing feature map and surface fitting

被引:35
作者
Laha, A [1 ]
Pal, NR
Chanda, B
机构
[1] Indian Inst Management, Kolkata 700027, W Bengal, India
[2] Indian Stat Inst, Electron & Commun Sci Unit, Kolkata 700108, W Bengal, India
关键词
cubic surface fitting; generic codebook; image compression; self-organizing feature map; vector quantization; Q;
D O I
10.1109/TIP.2004.833107
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
We propose a new scheme of designing a vector quantizer for image compression. First, a set of codevectors is generated using the self-organizing feature map algorithm. Then, the set of blocks associated with each code vector is modeled by a cubic surface for better perceptual fidelity of the reconstructed images. Mean-removed vectors from a set of training images is used for the construction of a generic codebook. Further, Huffman coding of the indices generated by the encoder and the difference-coded mean values of the blocks are used to achieve better compression ratio. We proposed two indices for quantitative assessment of the psychovisual quality (blocking effect) of the reconstructed image. Our experiments on several training and test images demonstrate that the proposed scheme can produce reconstructed images of good quality while achieving compression at low bit rates.
引用
收藏
页码:1291 / 1303
页数:13
相关论文
共 26 条
[1]   Image compression by self-organized Kohonen map [J].
Amerijckx, C ;
Verleysen, M ;
Thissen, P ;
Legat, JD .
IEEE TRANSACTIONS ON NEURAL NETWORKS, 1998, 9 (03) :503-507
[2]  
[Anonymous], P IEEE INT C NEURAL
[3]  
[Anonymous], Pattern Recognition With Fuzzy Objective Function Algorithms
[4]  
BAKER RL, 1983, P INT PICT COD S, P105
[5]   Segmentation based compression for graylevel images [J].
Biswas, S .
PATTERN RECOGNITION, 2003, 36 (07) :1501-1517
[6]   GRADIENT ALGORITHMS FOR DESIGNING PREDICTIVE VECTOR QUANTIZERS [J].
CHANG, PC ;
GRAY, RM .
IEEE TRANSACTIONS ON ACOUSTICS SPEECH AND SIGNAL PROCESSING, 1986, 34 (04) :679-690
[7]  
COSSMAN PC, IN PRESS IEEE T IMAG
[8]  
Gersho A., 1992, VECTOR QUANTIZATION
[9]  
Gonzalez R.C., 1992, DIGITAL IMAGE PROCES
[10]   Combining fractal image compression and vector quantization [J].
Hamzaoui, R ;
Saupe, D .
IEEE TRANSACTIONS ON IMAGE PROCESSING, 2000, 9 (02) :197-208