摘要 |
<p>In the DPCM predictive coding of frame sequences, the prediction error is determined from past prediction values. In the method according to the invention, a current prediction value (Pn) is determined from a past prediction value (Pn-1) and the difference of two past prediction values (Pn-1, Pn-2) in a normal case. When edges occur, in contrast, only one past prediction value (Pn-1) is utilised. Due to this measure, an oscillation of the prediction error can be prevented. The method according to the invention also allows the sampling frequency to be selected independently of the colour subcarrier, which facilitates the accommodation of the DPCM signals in an SDH container. <IMAGE></p> |