Abstract: High-amplitude-resolution graphics for raster-scanned television systems are generated beginning with relatively low-amplitude-resolution sampled data from image memory affording spatial resolution approximately the same as that of the television signal, generally at least as large as that afforded by the number of lines per field and by the video bandwidth used in the television system. The sample-data raster scan used in the television system is converted to the coordinate system used for memory addressing. The whole-number portions of the memory address are used to access addressable sample data at points surrounding the point specified by the complete address including whole-number and fractional portions; and the fractional portions of the address are used to perform an interpolation between or among the accessed sample data.

Abstract: A frame comb filter system includes circuitry for reducing motion induced artifacts in reproduced images. Frame comb filtered chrominance signal is line comb filtered to extract motion induced luminance cross components therein. The luminance cross components are added to frame comb filtered luminance signal to reduce motion induced luminance signal errors. The corrected luminance signal is then subtracted from composite video signal to produce the chrominance component of composite video.

Abstract: Descriptions of phantom raster scan are generated with high spatial resolution in angular coordinates and in coordinates, such as radial coordinates, orthogonal to the angular coordinates. The more significant bits of these coordinates are used to generate read addresses for a relatively low spatial resolution image memory supplying, in parallel, four data at a time descriptive of spatially adjacent points in image space. The less significant bits of one of these coordinates are used in calculating first and second linear interpolations between pairs of these data, and the less significant bits of the other of these coordinates are used in calculating a third linear interpolation between the results of the earlier linear interpolations to obtain successive samples of a video signal descriptive of the image stored in image memory. This avoids the need for large-capacity ROM in making the interpolations among image memory data to avoid rastering effects in the display.

Abstract: Improved accumulation apparatus for generating phantom raster in skewed Cartesian coordinates uses two, rather than four, accumulators. Arrangements for accommodating non-square pixels are described; as well as arrangements for accommodating the display image being compressed, expanded, or changed in aspect ratio.

Abstract: The storage locations of a memory system map a field of sample points. Apparatus affording parallel access to a plurality of storage locations describing an array of the sample points in an access window, which window can be shifted to any selected region of field of sample points responsive to orthogonal address coordinates of one of the sample points in the array, is described.

Abstract: The process of linearly interpolating between data taken from adjacent storage locations in a memory non-integrally addressed has the initial subtraction and final addition steps eliminated from it, and the partial products to be summed in the intervening multiplication process are modified to compensate for the elimination of these steps. This modification is done with a relatively fast select-one-of-four process, reducing the time for linear interpolation to that required for a simple multiplication--or, in the case of a single-bit multiplier, to the time for selection of the modified multiplicand to be used as linear interpolation result.

Abstract: Phantom raster scanning of an image memory, in which the image is stored in compacted form, is made possible by introducing offsets in strictly spatial coordinates, rather than in time coordinates parallelling a particular spatial coordinate system.

Abstract: Graphics for television display which emphasize circles, ellipses, circular or elliptical arcs, and the radii of circles or ellipses are conveniently stored in image memories addressable in polar coordinates. Television images which are to be rotated are conveniently stored in such memories, so the angular coordinate of the image memory addresses can be incremented or decremented to rotate the image with hardware as simple as a single adder. To read image memories addressed in polar coordinates, the Cartesian coordinates descriptive of television raster display are scan-converted to polar coordinates. This scan conversion takes place in real time at video sampling rates. The generation of the angular coordinate of the polar coordinates is facilitated by dividing one Cartesian coordinate by another by differentially combining their logarithms as found from read-only-memory table look-up, then finding an arc tangent from the logarithm of their difference as found from further read-only-memory table look-up.

Abstract: Orthogonal Cartesian coordinates descriptive of the raster scanning of a television display screen are generated at a programmable skew angle to the Cartesian coordinates of an image stored in memory. Each of these generated coordinates has a number of more significant bits used to generate the column and row addresses for that memory, four spatially adjacent storage location of which are read from for developing each sample of video signal describing a rotated television image. Each of these generated coordinates has a number of less significant bits affording resolution higher than that with which the memory can be addressed, which are used for calculation of each video sample by two-dimensional interpolation between the read-outs from the four locations in memory. To avoid aliasing as the image is rotated, the digital video to be stored in memory is processed using two-dimensional spatial-frequency low-pass filters to reduce resolution in directions other than horizontal and vertical.

Abstract: The calculation of r.sup.2 =x.sup.2 +y.sup.2, where x and y are the Cartesian coordinates of a raster-scanned space, is carried forward by accumulation of 2x+1 and 2y+1 terms. A read-only memory can then be used for table look-up of r, the radial coordinate of raster scan in polar coordinates.