| 摘要 |
Composite video signals are re-formatted into a compressed line-by-line serial format, by sampling the luminance and chrominance components at a predetermined sampling rate (22) and serially arranging them. Each line contains a pair of groups of samples, one of the pair of groups containing luminance samples, the other of the pair of groups containing chrominance samples. The separated components of the original video signals are sampled at first and second frequencies f1 and f2, which are related by the expression f2 = f1/N where (N) is the compression ratio. Both sampled signals are stored in a storage device (28) at first and second clock frequencies f1, f2, and are read from the storage device (28) with a higher frequency third clock signal f3. The luminance signal is compressed by a factor f3/f1 and the chrominance signal is compressed by a factor of N x f3/f1. After further processing, e.g. recording on a media for editing (12), the signals are reconverted (13) by sampling each line at a frequency f3' and combining samples from the luminance group selected at a rate f1' with samples from the chrominance group selected at a rate f2' equal to f1'/N. To optimize the recovery of the chrominance information a "mezzanine" signal (M) is inserted in the gap between the luminance and chrominance components in each re-formatted video line. The mezzanine signal (M) has a magnitude which comprises the median value between black and peak white, and serves as a subsequent reference level for both positive and negative excursions of the chrominance component for each given line. |
