The ability to employ neural networks to the task of image compression has been pointed out in research. The pre-dominant approach to image compression is centered around the backpropagation algorithm used to train on overlapping frames of the original picture. Several deficiencies can be identified with this approach. First, no potential time bounds are provided for compressing images. Second, utilizing backpropagation is difficult due to its computational complexity. To overcome these shortcomings we propose a different approach by concentrating on a general class of 3-layer neural networks of 2(N+1) hidden units. It is shown that the class /spl Nscr/* can uniquely represent a large number of images, in fact, the growth of this class is larger than exponential. Instead of training a network, it is automatically constructed. The obtainable compression rates (lossless) exceed 97% for square images of size 256/spl times/256 pixels.<>
Applying constructed neural networks to lossless image compression
Proceedings of 1st International Conference on Image Processing ; 3 ; 948-952 vol.3
1994-01-01
432726 byte
Conference paper
Electronic Resource
English
Applying Constructed Neural Networks to Lossless Image Compression
British Library Conference Proceedings | 1994
|Fast lossless image compression
IEEE | 1996
|Fast Lossless Image Compression
British Library Conference Proceedings | 1996
|Near-lossless image compression techniques
British Library Online Contents | 1998
|Lossless and Near-Lossless Image Compression Based on Multiresolution analysis
British Library Conference Proceedings | 2005
|