Picture crispening by adaptive digital signal processing

A technique is proposed for enhancing picture edges (sharpening boundaries) which will crispen the appearance of television pictures. Two closely related methods for both recognizing an edge, and introducing the enhancement, are presented. The first method is simpler, while the second causes less distortion. Either method is applied based on a criterion that defines the location of an edge. This enhancement applies to digital video, and could be used in television sets or monitors that use digital signal processing. The intent is to make a crisper looking picture than the transmission data rate would normally allow without producing the unwanted distortions associated with peaking. In its simplest form, an additional sample is inserted between each of the received and digitized samples, thereby doubling the sample rate locally. Rather than interpolating from the received samples to get the values of the locally inserted ones, one pattern out of a set of nine basic patterns is selected that adds to the locally generated sample in such a way as to synthesize faster edge transients. This technique can be applied to both the horizontal and vertical on a television frame.