A TECHNIQUE OF IMAGE COMPRESSION BASED ON DISCRETE WAVELET IMAGE DECOMPOSITION AND SELF ORGANIZING MAP
Megha Sharma, Rashmi Kuamri · 2014
Abstract. Image compression is the growing research area for the real world applications which is spreading day by day by the explosive growth of image transmission and storage. This paper presents the algorithm for gray scale image compression using self organizing map (SOM) and discrete wavelet transform (DWT). Self organizing map network is trained with input patterns in the form of vectors which gives code vector (weight matrix) and index values as the output. The discrete wavelet transform is applied on the code vectors and storing only the approximation coefficients (LL) and the index values obtained from the self organizing map. The result obtained shows the better compression ratio as well as better peak signal to noise ratio (PSNR) in comparison with the existing techniques.