Absrracf-The two-channel system previously reported has been implemented in hardware using system parameters appropriate to consumer television. The basic system divides the spectrum into lowand high-frequency spatial components. The lows are co&ely sampled and finely quantized and the highs finely sampled and coarsely quantized using a companded, randomized quantizer. The purpose of this experiment was to demonstrate the potential of the system for low-cost practical application, to study the effect of the character of the randomizing noise, and to ascertain that there were no deleterious effects due to interlace, motion, or input noise. Theoretical noise calculations were qualitatively confirmed.
INTRODUCTIONM INIMUM noise visibility coding (MNVC) is an image data compression system which exploits visual phenomena t o reduce the visibility of quantizing noise. For a given data rate, MNVC produces images superior to PCM. It is a nonstatistical system using a constant data rate, performs well in the presence of channel errors, and is simple and inexpensive to implement. As such, MNVC is a candidate for use in c o n s u m e r products as an alternative to DPCM.The principal departure of this system from DPCM [ l J is that a separate channel is used to transmit the dc and lowfrequency components of the signal. The filters associated with the high-frequency component can thus be nonrecursive and, as a result, the quantizing error can be made isotropic. Since the quantizer is not inside the feedback loop of the filter, its tapering can be designed for the optimum distribution Manuscript