Self organizing map and discrete wavelet transform based gray scale image compression

In this paper, the concept of self organizing map (SOM) with the combination of discrete wavelet transform (DWT) is applied for the compression of gray scale image. The index value and code vector are obtained after training the self organizing map and then the discrete wavelet transform is applied on the code vector by keeping only approximation coefficients. The result analysis shows that the compression ratio and peak signal to noise ratio (PSNR) value of the obtained decompressed image is better than the other existing methods.