SEPARATION OF VIDEO SIGNAL STRUCTURES TO INCREASE EFFICIENCY OF FACSIMILE TRANSMISSION

The efficiency of transmitting image signals along communications channels depends, largely, on the degree of matching of the statistical characteristics of a specific form of image with potentialities of the code used. Maximum matching occurs by adapting the code to the video signal characteristics. Especially effective is the use of adaptation in the case of transmission of images with sharply differing characteristics, for example newspaper sheets containing text and illustrations of a raster structure (Eq. (1)). However, in this case, a problem arises of reliable separation of video signals of text and illustrations. Moreover, depending on the goal set - to increase the efficiency of compression of the video signal volume or to correct errors - requirements placed on the accuracy of defining the text-illustration boundary may differ. In the first case, an error in defining the boundary leads only to a certain decrease in the coefficient of compression, while in the second, to a degradation in the quality of image reproduction.