METHOD FOR COMPRESSING IMAGES BY SELF-ORGANIZATION OF A NEURONAL NETWORK

摘要

The image compression method comprises: treating a current block of pixels moved in the image to be compressed using a network of N neurons, from 1 to N, arranged in the same plane, each neuron having an equal number of inputs the dimension of the blocks extracted from the image, and such that, after a learning phase during which the network is self-organized, so that neurons represent the most probable input values, each neuron receives at the same moment as inputs the pixels of the current block and outputs a signal, or potential, dj, which is positive and decreases as the state of the corresponding neuron j is close to the state of the inputs ei, to code the current pixel block by a code associated with neuron j dj whose potential is minimal. The invention applies in particular to the compression of television pictures.