| 摘要 |
<p>A method, for use in an image signal decoding system, for concealing errors in an image signal transmitted in a block-by-block compressed form, the decoding system including a decoding circuit for decoding the transmitted image signal into a decoded image signal on a block-by-block basis and detecting an erroneous block which contains an error, the erroneous block having a predetermined number of surrounding blocks without containing an error therein, each of the surrounding blocks having a multiplicity of pixels therein, the method comprises the steps of: classifying each of the surrounding blocks into a monotonous, an edge or a texture class, the monotonous class representing a block without containing an edge therein, the edge class representing a block containing one edge therein, and the texture class representing a block containing more than one edge therein; classifying the erroneous block as the edge class, if only one pair of the surrounding blocks positioned opposite each other with respect to the erroneous block is of the edge class and has an identical edge direction transversing the erroneous block; and, otherwise, as either the monotonous or the texture class depending on the number of the surrounding blocks having either the monotonous or the texture class; and rectifying the erroneous block based on the classification thereof. <IMAGE></p> |
