A picture multiplex analysis hybrid device comprising a device for forming the picture of an object in a given plane, a plurality of photodetectors forming a mosaic thereof, and a plurality of optical devices sending the light from a sector of the picture in its plane to a respective photodetector. The number of optical devices is equal to that of the photodetectors, wherein the light beams received in the elementary areas forming a picture sector are modulated by a moving encoding grid comprising opaque portions corresponding to a binary 0 and transparent portions corresponding to a binary 1, the grid modulating each beam by a binary, periodic and pseudo-random function, the beams emanating from the elementary areas being modulated by identical parts of the grid.

A method and associated apparatus is provided for converting halftone and screened films (12A and 12B) for use with electromechanical gravure engravers (22), comprising the steps of positioning an array (26) of photosensitive elements adjacent at least one film (12A) having an image thereon in the form of transparent and opaque picture elements (pixels) (30 and 32) having predetermined dimensions, affecting relative movement between the array (26) and the film (12A) for sampling a plurality of pixels of the film by the array (26), assigning a predetermined number of photosensitive elements to a predetermined number of photosensitive elements to a predetermined width of pixels, effectively creating a first sample area (34) bounded by the predetermined width of pixels and a pre-established number of samples, effectively creating a second sample area (36) smaller than the first sample area (34) and bounded by a plurality of pixels, the center of the second sample area (36) being coincident with that of the first sample area (34), counting the photosensitive elements which provide signals indicating transparent pixel areas within the first sample area (34), counting the photosensitive elements which provide signals indicating transparent pixel areas within the second sample area (36), adding the count of the photosensitive elements of the second sample area (36) to the count of the photosensitive elements of the second sample area (36) minus the count of the first sample area (34) to provide a signal representing the percentage dot area (PDA) for use with an electromechanical gravure engraver (22).

The invention provides for deciding whether a recording dot is to be printed for the reproduction of scan dots. The scan dots being considered are compiled in an input switching network into groups, for which an arithmetic processor determines the tonal value sum. The arithmetic processor emits a signal, if the tonal value sum exceeds a certain figure. The recording dot to be printed is determined by the presence of the signal in an output switching network. This process is performed successively for increasingly larger areas of the original, that is, for increasingly larger quantities of scan dots.

A method and apparatus for minimizing errors in digital processing of electrical signals whereby a digital input signal is divided into a digital useful signal having a lower resolution than the input signal and into a digital error signal having lower resolution than the input signal. A deviation of the useful signal is reproduced from the input signal and a check is performed to see whether the sum error of a plurality of successive useful signal values crosses a prescribed upper or lower threshold. When a threshold is crossed, an error value of the useful signal which comes closest to this threshold is identified and the useful signal and the error signal of this signal value are corrected such that the sum error no longer crosses the threshold.

The invention provides for deciding whether a recording dot is to be printed for the reproduction of scan dots. The scan dots being considered are complied in an input switching network into groups, for which an arithmetic processor determines the tonal value sum. The arithmetic processor emits a signal, if the tonal value sum exceeds a certain figure. The recording dot to be printed is determined by the presence of the signal in an output switching network. This process is performed successively for increasingly larger areas of the original, that is, for increasingly larger quantities of scan dots.