A hybrid discrete wavelet transform with neural network back propagation approach for efficient medical image compression

Digital image compression technology is of vital importance for the very fast transmission and applicable processing of digital image information on the internet. The main basic aim of image compression is greatly to reduce the number of the image pixel elements with the main intension not to affect the original quality of the original image that is to be compressed. It is usually done by removing the redundancy present in the image. Objective of this paper is to compare few promising compression techniques such as DWT coding, Back Propagation Neural Network (BPNN), new hybrid techniques for compression. DWT improves the quality of compressed image. Back-propagation algorithm can be extensively used as a learning algorithm in Artificial Neural Networks. BPNN comes under Feed-Forward Neural Network Architecture. This type of architecture can be used in approximation of all the problems, which is having high precision. Error correction learning rule is particularly used by this Neural Network. This is very efficient algorithm for Image Compression, which works with the architecture of Artificial Neural Network (ANN). There are different types of parameters, which includes Compression ratio (CR), Peak signal to noise ratio (PSNR), Bits per pixel (BPP) and Mean Square error (MSE). The quality of any compressed image can be assessed using a set of parameters. Then the performance Analysis of different images is carried out (on application of three different algorithms) in this paper. The resulst clearly explais that hybrid Image Compression using hybrid DWT-BP(Discrete Wavelet Transform-Back Propagation) provides better CR and PSNR.