Feature extraction of cable television picture impairments

Abstract

The properties of the two-dimensional Fourier spectra of the most significant impairments occurring in cable television transmission systems are studied. These are snow noise, composite triple beats, picture carrier beats impairments, and power line interference. Features of these impairments are found. These features are to be used in an automatic television picture impairment detection and classification system. It is shown that the impaired television pictures have different spectrum properties which enable design and implementation of a simple automatic television noise detection and classification system.<>