FR2711876A1 - Video signal processor. - Google Patents

Abstract

The invention relates to a video signal processor in which signals representing images are subjected to anti-aliasing before display. The processor includes means for 1) subjecting these signals to a relatively slow first anti-aliasing process to produce relatively high quality images and 2) to subjecting these signals to a relatively fast second anti-aliasing process in order to produce high quality images relatively weak; and means (8) for making a selection between the first and second treatments. This achieves a desired balance between processing speed and image quality.

Description

1 2711876

  The present invention relates to a signal processor

  video like, for example, a video character generator.

  In television broadcast studios, character generators are used to provide character signals which are superimposed on a television signal to display an image on a television receiver. To display high quality text, an anti-aliasing technique is used to digitally filter the signal representing each character before it is used. This may however result in a slower processing speed than that desired in some applications. According to the present invention, there is provided a video signal processor in which the signals representing images are subjected to an anti-aliasing treatment before display, the processor comprising means for 1) subjecting these signals to a first anti-aliasing treatment relatively slow to produce relatively high quality images and 2) subjecting these signals to a relatively fast second anti-aliasing process in order to produce images of relatively lower quality; and means for making a selection between the

first and second treatments.

  The video signal processor is preferably a video character generator, and the signals representing images preferably represent characters or

  other graphic features.

  The aims, characteristics and advantages of the present invention will emerge more clearly on reading the

  following detailed description of an embodiment

  preferred, given by way of nonlimiting example with reference to the single figure of the accompanying drawings, which represents a functional diagram of an example of a generator

  of video characters using the present invention.

  Referring to this single figure, we can see a block diagram of a video character generator of the type manufactured by Aston Electronic Designs Limited, at 123/127 Deepcut Bridge Road, Deepcut, Camberley, Surrey GU16 6SD,

England, and called "MOTIF".

  The images to be displayed are contained in a device with two image memories 1 comprising two image memories A and B with 32 binaries allowing cuts during transmission and effects to be produced between two pages of memory. The image memory device is designed to present the two fields of the image simultaneously to a motion interpolator 2. The latter performs a horizontal interpolation by taking a weighted sum of three adjacent pixels on the current line, and vertical in taking a weighted sum of corresponding pixel values from the current line of the current frame and the lines on either side (in the other frame). By varying the weights, the resulting output places an image on the output page with an accuracy less than the pixel (horizontally) or less than the television line (vertically). In this way, it is possible to scroll or advance legends

  steadily at any speed.

  The output of the motion interpolator 2 is transmitted to a conventional digital / analog converter (DAC) and to an analog processing output stage 3 as well as to

  a conventional digital output processing stage 4.

  A separate 5 to 32 binary state image memory contains a full color state image for

be displayed to the operator.

  The display structure described above is controlled by a display processor 6 implemented using partial chip components. The generator is controlled by a 68030 7 microprocessor as well as by a mathematical coprocessor which processes 16 MB (megabytes) of memory, comprising a front cache memory, a disk memory and means of communication with the operator and

external devices.

  In practice, the central processor of the generator occupies three units of rack space and has a depth of 530 mm. It includes a vertical backplane with a head-to-tail arrangement in which are inserted four double Euro cards (Double Euro Cards) with a depth of 280 mm, at the front, and two other double Euro cards (Double Euro Cards) at the front. 'back. The four front cards include video output; the image memory device 1; the display processor 6; and the central processor. The two rear boards include control circuits and female video / synchronization input and output connectors; parallel and serial interfaces;

  network interface; and keyboard input.

  The electronic unit also houses a power supply, two standard disk drives (a 127 MB hard disk drive and a 3.5 "flexible disk drive), plus an optional 44 MB Syquest disk drive with removable disks , an encoder / synchronization pulse generator (SPG) manipulator and digital serial and parallel outputs conforming to the Recommendations of the International Radiocommunication Advisory Committee (CCIR) 601 and 656. Two "Video Grab" options are also available , one capable of receiving images in VBR or in key plus Y / Pb / Pr, the other in parallel formats

and series 601.

  A keyboard 8 is provided and it is possible to chain up to each other up to ten keyboards using a standard 75 ohm video cable, up to 300 m maximum

central processor.

  Video outputs are provided for both formatting and transmission. The formatting output from a manipulator 9 is provided in VBR, while a transmission can have a configuration either for VBR or for the key plus Y / Pb / Pr. Separate simultaneous synchronization outputs derived from an external synchronization source or the internal synchronization pulse generator are also provided at the same time

  for shaping and transmission.

  Two parallel interfaces are provided as well as two intelligent serial interfaces which, for electronic information studio applications, can be configured for RF232 or RS422 operation. A parallel Centronics interface can be connected to a printer to print directories, etc. A network interface can support up to eight nodes, or up to 64 nodes with external hardware. On the basis of a Computing Network with Auxiliary Resources (ARCNET), the network can allow the transfer between machines of page files, pages, images of logos and

individual characters.

  An optional device formed by a "synchronization" color coder / SPG 10 and a manipulator 11 provides two key outputs (one of which is provided with a protective bypass), two coded outputs, two synchronization outputs mixed broadcast quality (when they are "synchronized", that is to say synchronized with an external reference signal) and a YC (S) output. An optional digital output according to CCIR Recommendations 601 and 656 exists in two versions: parallel only (transmission and key); and parallel and series. The "Video Capture" option exists for inputs from key VBR sources plus Y / Pb / Pr, or for a parallel digital format /

series.

  The characters (or other graphic data) intended to be processed and / or displayed by the system are entered at the keyboard 8 or remotely from the parallel, serial or network interface and loaded into the

  system through an input / output unit 12.

  The character generator displays the characters (or graphical data) in an anti-aliasing form to suppress visual aliasing, such as "stair steps" along the diagonal lines. This is done by controlling the intensity of the pixels which define the characters more precisely than by simply "on" or "off". The intensity of each pixel can be appropriately selected from

  256 intensity levels available.

  The quality of the perceived character can be judged mainly using two criteria: the "regularity" of all the lines and curves of the character (for example, absence of "stair steps"), and the visual stability of the character ( for example, no interlacing flicker). Therefore, the generator can perform anti-aliasing between individual scan lines in order to avoid visual flicker which may occur due to the interlaced regeneration process through

  to which a television picture is produced.

  The method by which anti-aliasing is performed is the main factor that determines the quality of the characters output. However, as previously described, higher degrees of anti-aliasing

  require longer processing times.

  One of the standard anti-aliasing techniques consists in restoring the desired character according to a size greater than the required size (for example twice greater than the final size), but in a non-anti-aliased manner. We then reduce this non-anti-aliased reference character to the required size by passing a two-dimensional filter over it. In accordance with the example of the present invention, the processor 68030 7 changes the characteristics of the filter which is to be used by the character generator, as well as the reduction ratio required from the reference height to the required height, so that different character qualities and, therefore, different processing speeds can be chosen by the user. The user can for example choose between high quality characters with a relatively slow processing speed and lower quality characters with a relatively slow processing speed

quick.

  To do this, the generator must have at least two treatment techniques. Two processing techniques requiring different processing speeds and producing characters of different qualities will

  now be described as an example.

  According to a relatively slow first technique, a two-dimensional Gauss filter is used to reduce the image and produce a high quality character. Assuming a reduction rate in both height and width between the reference character and the anti-aliasing character, a two-dimensional Gauss filter having a dimension of N pixels (in width) over M lines (in height) is used to produce a weighted sum over an area of the reference character. The value of this weighted sum at a particular point is used to give the intensity of the corresponding pixel in the antialiasing character. The filter is applied every R pixels and every R lines through the reference character to produce the dimension of

anti-aliasing required.

  More precisely, the weighted sum W for a pixel at a point (x, y) of the anti-aliasing character is calculated using the following equation:

M N

2 2

W) = F (x. + Iy + j) M (h + iR.y + J)

M N

2 2

  in which: F (X Y) represents the weight of the filter at an offset (x, y) (note that (0, 0) corresponds to the center of the filter); and M (X y) represents the pixel value, or its intensity, in the reference character at the location level (x, y). M (^ y) can only have a value of 0 or 1, which

  indicates that the pixel is "off" or "on".

  Typically, the following constraints will apply: N, M 2 R to ensure that all the reference pixels are sampled; and M 2 N to help reduce the interlacing flicker. The weight F of the Gaussian filter at x pixels and y lines from the center of the filter can be calculated from the following equation: F (x) = ke "42 in which: k is a scale constant used to control the total weight of the filter; e is the natural logarithmic base; and d is the distance, scaled by a constant c, from the center of the filter to the position at x pixels and y lines from the center of the filter, given by: d = c x- + y2 It is possible to obtain different qualities of character and different processing speeds by

  varying the constants R, M, N, k and c above.

  A second relatively faster technique (which typically provides lower quality) is a variation of the first technique and uses a flat filter and not a Gauss filter. In this case, all filter weights have the same constant value f, i.e.: F (xAy) f In this technique, the values corresponding to M and N will typically be lower to the

  equivalent values obtained with the first technique.

  The operator can make, for example, using the keyboard 8 a choice between these techniques or, optionally between other techniques to achieve a desired balance

  between processing speed and quality of characters.

Claims (10)

  1. Video signal processor in which signals representing images are subjected to anti-aliasing before display, characterized in that it comprises means for 1) subjecting these signals to a first relatively slow anti-aliasing treatment in order to produce images intended to be displayed of relatively high quality and 2) subjecting these signals to a second relatively fast anti-aliasing process in order to produce images intended to be displayed of relatively low quality; and means (8) for making a selection between the first and second treatments.   2. Processor according to claim 1, characterized in that the first anti-aliasing processing comprises the steps which consist in receiving the data which define the images with a resolution higher than that with which they are to be displayed, and in processing the data received to produce the images intended to be displayed.   3. Processor according to claim 2, characterized in that the first anti-aliasing treatment comprises the step which consists in applying a first filter to the data received. 4. Processor according to claim 3, characterized in that the first anti-aliasing treatment comprises the step which consists in determining the intensity of a pixel of an image intended to be displayed by calculating the sum W of the products of 1 ) the intensity of each corresponding location of the data received and 2) the value of a first function of   filter suitable for this location.   5. Processor according to claim 4, characterized in that the first filter function is a function of Gaussian filter.   6. Processor according to any one of the claims   previous, characterized in that the second anti-aliasing processing comprises the steps which consist in receiving the data which define the images with a resolution higher than that with which they are to be displayed, and in processing the data received to produce the   images to be displayed.   7. Processor according to claim 6, characterized in that the second anti-aliasing treatment comprises the step   which consists in applying a second filter to the reviewed data.   8. Processor according to claim 7, characterized in that the second anti-aliasing treatment comprises the step which consists in determining the intensity of a pixel of an image to be displayed by calculating the sum W of the products of 1) l intensity of each corresponding location of the received data and 2) the value of a filter function appropriate to this location.   9. Processor according to claim 8, characterized in that the filter function has a constant value for all the slots.   10. Processor according to any one of the claims   above, characterized in that it consists of a video character generator, and in that the signals which represent the images include representations of   characters or other graphic characteristics.