Method and apparatus approximately correcting errors in a digital video signal
US4485399A · kind A · utility
Assignee
Inventors
Key dates
| Filing date | Mar 12, 1982 |
| Grant date | Nov 27, 1984 |
| Priority date | — |
| Expiry date | Mar 12, 2002 |
Classification
- Technology area (CPC H)Electricity
- CPC primaryH04N19/895
- WIPO fieldAudio-visual technology
- WIPO sectorElectrical engineering
Abstract
Picture points represented by a digital signal that is found to be in error are set to zero (black) value. Incoming 8-bit signals are twice delayed by one line delays, and 8-bit shift registers are provided for the undelayed, once delayed and twice delayed signals, each shift register providing three places for successive picture points. Averages are continuously formed horizontally, vertically and on both diagonals about every picture point, and when one value which is averaged is faulty, the divider at the output of the addition stage has its divisor changed from 2 to 1. The averages themselves are averaged to eliminate averages that are unusable because both inputs are faulty. The color television signal must be separately coded for the color and luminance information, and essentially the same scheme may be used for covering over errors in the color information as in the luminance information.
Source: USPTO / EPO open patent data. Objective bibliographic and citation counts.