Globally trained neural network architecture for image compression

The authors discuss the development of a coding system for image transmission based on block-transform coding and vector quantization. Moreover, a classification of the image blocks is performed in the spatial domain. An architecture incorporating both multilayered perceptron and self-organizing feature map neural networks and a block classification is considered to realize the image coding scheme. A framework is proposed to globally train the whole image coding system. The achieved results confirm the merits of such an image coding scheme. The neural network integration is performed with a single learning phase, allowing faster training and better performance of the image coding system.<>