Abstract

A two-stage coding system for reducing the high data rate of a digital HDTV signal (≈1 Gbit/s) is described. Spatio-temporal subsampling of the interlaced source signal is used in conjunction with an intrafield DPCM of the remaining samples. For the luminance signal two well-known sampling patterns, fieldquincunx and linequincunx, are proposed; both provide a 2:1 subsampling. Prior to subsampling, a digital filter which adapts itself to the movement in the scene, adjusts the three-dimensional spectrum of the television signal to some reduced region supported by the sampling pattern and the properties of the human visual system. Different filter structures are suggested for prefiltering and for interpolation of the missing samples in the receiver. The chrominance signals are horizontally bandlimited before digitization. A modified linequincunx sampling structure is used for subsampling the chrominance signals. Two-dimensional digital prefilters and interpolators are required. Then predictive coding with a fixed quantizer and predictor is applied to the picture elements which remain in the luminance and chrominance paths after subsampling. Sixteen quantization levels are used for the luminance and for each chrominance component to reduce the whole data rate to less than 280 Mbit/s.

Full Text
Paper version not known

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call