Abstract: A receiver interpolates finer intensity levels in critical areas of a dither-quantized TV picture that is sent (or recorded) with very few intensity levels. This improves picture rendition and reduces the visibility of stipple dots. One embodiment, for video quantized with multiphase ordered dither, substitutes the mean of several successive frames at all picture elements that are not changing with time. Another embodiment averages spatially on the frame in areas devoid of high-resolution detail. Both versions use a simple test for activity that permits instant pel-to-pel switching between averaged samples and input samples. The two embodiments can be combined.

Abstract: Television systems that transmit dither-quantized samples for psychovisually compressed signalling are improved by substituting novel multi-phase dithers of the invention for previously known multi-phase dithers. The improved dithers comprise ordered-dither frame patterns and typically have sixteen or eight phases and sixteen sizes of dither samples. A first variety, not self-interpolating, merely shifts the frame-pattern structure from frame-to-frame. Methods and means are disclosed for generating novel self-interpolating dithers with sixteen or eight phases, including preferred dithers. Self-interpolating dithers are generally superior and they also facilitate receiver display of more quantum levels than actually transmitted.

Abstract: Dither-quantizing with Cinematic Dithers permits coarse quantization of the Nyquist samples transmitting television luminance or chrominance in analog or digital signals. Means are disclosed for generating improved p-phase Cinematic Dithers corresponding to any desired value of p and to liberal choice in the number of dither sizes. In general, a selected random or ordered basic arrangement of the dither is altered cyclically from frame to frame and recurs every p television frames. The frame-to-frame changes are also tailored to the spatial and temporal filtering characteristics of the human visual system in order that the Nyquist samples can be restricted to very few amplitude levels without corresponding impairment of perceived picture quality.

Abstract: Novel means are disclosed for conditioning a standard color-television composite signal, such as the NISC composite, so that pictorial information is represented according to Nyquist samples which are very coarsely quantized and therefore regenerable. This conditioning does not affect compatibility with a standard receiver or perceptibly reduce the quality of a picture displayed thereon. Carefully designed dither, such as 3-dimensional nasik dither, is employed to preserve picture information and quality despite coarse quantization.An improved receiver also regenerates the received signal so that unwanted noise, distortion and interference are removed. The disclosure exemplifies the teachings of the invention in terms of broadcast and cable television systems and video recording systems. An improved videodisk record is disclosed and means whereby, in a pay-television system, the picture incorporated into a standard composite signal may be key-encrypted according to a private cryptographic key.

Abstract: The chrominance constituent of a color-television picture is transmitted by means of chrominance samples dither-quantized according to a small number of transmittable chrominance values. By use of a look-up table or equivalent analog means, the chrominance vector can be quantized directly; else vector components are quantized. Three-dimensional dither causes the TV picture to be displayed with pointillistically colored micropatterns that change from frame to frame. The effectiveness of psychovisual filtering in time and space domains, which effectively suppresses these micropatterns, is improved by arranging that neutral-colored (e.g. gray) elements shall be generated directly on each frame. This can be done by making zero chrominance a transmittable quantum value, or otherwise as in disclosed examples.

Abstract: Novel means are disclosed for conditioning a standard color-television composite signal, such as the NTSC composite, so that pictorial information is represented according to Nyquist samples which are very coarsely quantized and therefore regenerable. This conditioning does not affect compatibility with a standard receiver or perceptibly reduce the quality of a picture displayed thereon. Carefully designed dither, such as 3-dimensional nasik dither, is employed to preserve picture information and quality despite coarse quantization. An improved receiver also regenerates the received signal so that unwanted noise, distortion and interference are removed. The disclosure exemplifies the teachings of the invention in terms of broadcast and cable television systems and video recording systems. An improved videodisk record is disclosed and means whereby, in a pay-television system, the picture incorporated into a standard composite signal may be key-encrypted according to a private cryptographic key.

Abstract: Novel means are disclosed for conditioning a standard color-television composite signal, such as the NTSC composite, so that pictorial information is represented according to Nyquist samples which are very coarsely quantized and therefore regenerable. This conditioning does not affect compatibility with a standard receiver or perceptibly reduce the quality of a picture displayed thereon. Carefully designed dither, such as 3-dimensional nasik dither, is employed to preserve picture information and quality despite coarse quantization. An improved receiver also regenerates the received signal so that unwanted noise, distortion and interference are removed. The disclosure exemplifies the teachings of the invention in terms of broadcast and cable television systems and video recording systems. An improved videodisk record is disclosed and means whereby, in a pay-television system, the picture incorporated into a standard composite signal may be key-encrypted according to a private cryptographic key.

Abstract: One or more of the video components of prior art are quantized coarsely before being sent to a receiver. To prevent perceptible loss of picture information and to minimize visible artifacts, suitable ordered dither is added before a component is quantized. The transmitted signal is preferably also sampled at the transmitter and resampled and requantized in the receiver for combatting noise and distortion. Some forms of the invention are compatible with prior art receivers. Other forms provide compressed digital signalling and coarsely quantized pulse-amplitude modulation.