| 摘要 |
In a high-quality component video digital transmission system (FIGS. 1 and 2), transmitter DPCM encoders (110-112) and corresponding receiver DPCM decoders (210-212) include respective adaptive quantizers (402-502 and 802-902) in which a set of prediction error-representative values is selected by the predicted value of only the one sample currently being encoded or decoded. At each encoder and decoder, the predicted values are quantized by a quantizer selector (406-506 and 807-907) into groups. One group is selected by the predicted value of the current sample. An adaptive quantizer set is then selected by the selected group. The selected set's prediction error-representative values span a range of error values bounded by the maximum and minimum error values possible for any predicted value of the selecting group. Furthermore, the predicted value of each chrominance sample is forced at each chrominance encoder and decoder by a respective limiter (505, 905) to lie within bounds determined by the correspoding luminance sample value (FIGS. 6 and 7). Improvements in image quality thus obtained permit further transmission rate reduction, by using interleaved subsamplers (107-109). At the transmitter, an interpolation selector (113) selects interpolation methods for reconstructing sample values discarded during subsampling on the basis of only luminance signal samples. At the receiver, both luminance and chrominance interpolators (207-209) apply the same method to reconstruct the value of a luminance and a corresponding chrominance discarded sample. These techniques together enable high-quality component video transmissions to proceed over a 45 Mbps link (200).
|
