Multi resolution transform and artificial neural network based digital image coder

Image processing involves changing the nature of an image in order to either 1. Improve its pictorial information for human interpretation, 2. Render it more suitable for autonomous machine perception. Compressing images is essential for efficient image processing, since a good quality image consumes a large storage space. There are various methods to perform image compression. Fourier analysis and Wavelet analysis can help us to compress an image significantly. Wavelet transform which is more efficient is used along with SOM to compress an image. Artificial neural networks have been trained to perform complex functions in various fields of application including pattern recognition, identification, classification, compression, speech, vision and control systems. Here we have used self organising map, a method of unsupervised learning, in order to perform vector quantisation. Encoding and decoding is done during the stages of compression and decompression respectively. Here the type of encoding used is run length encoding.