Image compression using NNs
Abstract: It is well known that real-time movie transmission over serial lines makes a tough problem, which can be solved only by huge image compression rates. The classic compression algorithms have serious limitations on that peculiar case because they are not "custom made" for images. We have developed here a way to get 16:1 up to 21:1 compression rates, using neural networks trained with advanced supervised backpropagation algorithms. These techniques, along with successive frames optimization, may open a way to real-time movie transmission across the internet.
By: Razvan Jigorea and Ludovic Pajtek


 A New Compression Technique Using an Artificial Neural Network
Abstract: In this paper, we present a direct solution method based neural network for image compression. The proposed technique includes steps to break down large images into smaller windows and eliminate redundant information. Furthermore, the technique employs a neural network trained by a non-iterative, direct solution method. An error backpropagation algorithm is also used to train the neural network, and both training algorithms are compared. The proposed technique has been implemented in C on the SP2 Supercomputer. A number of experiments have been conducted. The results obtained, such as compression ratio and transfer time of the compressed images are presented in this paper.
By: B. Verma, M. Blumentein and S. Kulkarni.