A self-organizing neural network performing learning vector quantization (LVQ) is proposed in this paper to compress image data in still pictures. The advantages of our model are its low training time complexity, high utilization of neurons, robust clustering capability, and simple computation; further, a VLSI implementation is highly feasible. By unsupervised learning, our LVQ neural model finds near-optimal clustering from image data and builds a compression codebook in the synaptic weights. The compression results are competitive comparing with the currently popular transform codings such as JPEG and wavelet methods. The neural codebook trained by a few pictures can be used to compression other pictures efficiently. Special image types such as the fingerprints exhibit this property in our experiments. Other experiments involve some filtering effects and techniques to enhance the neural codebook learning to yield higher picture quality.<>
Fingerprint image compression by a natural clustering neural network
Proceedings of 1st International Conference on Image Processing ; 2 ; 341-345 vol.2
1994-01-01
755788 byte
Conference paper
Electronic Resource
English
Fingerprint Image Compression by a Natural Clustering Neural Network
British Library Conference Proceedings | 1994
|Image compression with a hierarchical neural network
IEEE | 1996
|Size-Adaptive Neural Network for Image Compression
British Library Conference Proceedings | 1994
|Image Compression with a Hierarchical Neural Network
Online Contents | 1996
|Size-adaptive neural network for image compression
IEEE | 1994
|