Abstract | ||
---|---|---|
An image compression technique is proposed in which a multilayer perceptron (MLP) predictor takes advantage of the topological properties of the Kohonen algorithm. The Kohonen algorithm creates a code-book which is used for vector quantization of the source image. Then, an MLP is trained to predict references to code-book, allowing further compression. Even with difficult images, the result is a reduction of 15% to 20% of the bit rate compared with classical vector quantization techniques, for the same quality of decoded images |
Year | DOI | Venue |
---|---|---|
1993 | 10.1109/ICNN.1993.298645 | San Francisco, CA |
Keywords | Field | DocType |
feedforward neural nets,image coding,image processing,topology,vector quantisation,kohonen algorithm,code-book,image compression,multilayer perceptron,topological maps,vector quantization,transform coding,decoding,bandwidth,code book,backpropagation | Topology,Pattern recognition,Learning vector quantization,Image processing,Bit rate,Self-organizing map,Multilayer perceptron,Vector quantization,Artificial intelligence,Fractal transform,Image compression,Mathematics | Conference |
Citations | PageRank | References |
2 | 0.68 | 2 |
Authors | ||
2 |
Name | Order | Citations | PageRank |
---|---|---|---|
Gilles Burel | 1 | 297 | 113.35 |
Catros, J.-Y. | 2 | 2 | 0.68 |