Title :
Sample weighting when training self-organizing maps for image compression
Author_Institution :
Neural Networks Res. Centre, Helsinki Univ. of Technol., Espoo, Finland
fDate :
31 Aug-2 Sep 1995
Abstract :
Image compression is an essential task for image storage and transmission applications. Vector quantization is often used when high compression rates are needed. Self-organizing map (SOM) algorithm can be used to generate codebooks for vector quantization. Previously it has been demonstrated that using the special property of the SOM algorithm that the codebook entries are ordered one can use prediction coding of codewords to make the compression more effective. In this paper it is shown that training the SOM algorithm by using different weighting for sample blocks having different statistical characteristics one can further increase the compression efficiency
Keywords :
data compression; image coding; learning (artificial intelligence); self-organising feature maps; statistical analysis; vector quantisation; codebook generation; compression efficiency; high compression rates; image compression; image storage; image transmission; sample weighting; self-organizing map training; statistical characteristics; vector quantization; Algorithm design and analysis; Clustering algorithms; Decoding; Image coding; Image storage; Iterative algorithms; Neural networks; Pixel; Self organizing feature maps; Vector quantization;
Conference_Titel :
Neural Networks for Signal Processing [1995] V. Proceedings of the 1995 IEEE Workshop
Conference_Location :
Cambridge, MA
Print_ISBN :
0-7803-2739-X
DOI :
10.1109/NNSP.1995.514908
