Error concealment in digital television signals

• Main Authors: WILKINSON, JAMES HEDLEY, IVE, JOHN GOODWIN SPENCER, BORELAND, PETER CHARLES
• Format: Patent
• Language: English; French; German
• Published: 05.03.1986
• Edition: 4
• Subjects: ELECTRIC COMMUNICATION TECHNIQUE; ELECTRICITY; INFORMATION STORAGE; INFORMATION STORAGE BASED ON RELATIVE MOVEMENT BETWEEN RECORDCARRIER AND TRANSDUCER; PHYSICS; PICTORIAL COMMUNICATION, e.g. TELEVISION

Apparatus for concealing errors in a digital television signal, comprises delay elements (11, 12 and 15 to 20) for deriving a current sample value (EO) and a first group of the sample values (S to S8) spatially closest to the current sample value (EO) in the same frame of the television signal, delay elements (10, 13, 14 and 21 to 26) for deriving from the preceding frame of the television signal a delayed sample value EOD) corresponding in position to the current sample value (EO) and a second group of sample values (S1 D to S8D) respectively corresponding in position to the sample values (S1 to S8), means (48) for checking the current sample value (EO) for error, means (48) for checking the delayed sample value (EOD) for error, subtractors (35 to 41 and 44 to 47) for checking pairs of sample values (S to S8 and S1 D to S8D) and corresponding in position in the current and preceding frames of the television signals for errors and for deriving a measure of the difference in the levels of the sample values (S1 to S8 and S1D to S8D) in each pair, and for combining the results of the checks to provide an output signal which is a measure of the number of said errors and of the sum of said differences, a threshold device (48) for checking said output signal against a predetermined threshold level corresponding to an acceptably small number of errors and a relatively low degree of movement in the part of a television picture corresponding to the groups of sample values (S to S8 and S1 D to S8D), and a selector (49) operative, if the sample value (EO) is in error, the delayed sample value (EOD) is not in error, and said output signal is below said predetermined level, for replacing the current sample value (EO) by the delayed sample value (EOD) for supply to an output (50), and, in all other cases, for supplying the current sample value (EO) to the output (50).