A new image coding scheme based upon image pattern recognition

Author

Li, Hongsong ; Li, Da

Author_Institution

Coll. of Inf. & Commun., Guilin Univ. of Electron. Technol., Guilin, China

Volume

1

fYear

2010

fDate

29-31 Oct. 2010

Firstpage

172

Lastpage

174

Abstract

This paper presents a new image coding scheme based upon image pattern recognition (IPR) and discrete wavelet transforms (DWT). First, the original image is predicted by image pattern recognition, then the predicted error image is encoded by DWT, and the pattern library learning is based upon self-organizing maps (SOM) algorithm. To improve the performance of SOM algorithm, a frequency sensitive self-organizing feature maps (FSOM) algorithm is proposed. Experiment results show that the IPR-FSOM-DWT coding scheme can get better coding performance than standard JPEG2000.

Keywords

discrete wavelet transforms; image coding; image recognition; self-organising feature maps; JPEG2000 standard; discrete wavelet transform; image coding scheme; image pattern recognition; original image; self organizing feature map algorithm; Discrete wavelet transforms; Encoding; Image coding; Image reconstruction; Predictive models; Streaming media; Transform coding; JPEG2000; image coding; pattern recognition; self-organizing map;

fLanguage

English

Publisher

ieee

Conference_Titel

Intelligent Computing and Intelligent Systems (ICIS), 2010 IEEE International Conference on

Conference_Location

Xiamen

Print_ISBN

978-1-4244-6582-8

Type

conf

DOI

10.1109/ICICISYS.2010.5658654

Filename

5658654