WO2002039752A2

WO2002039752A2 - High quality color key extraction apparatus

Info

Publication number: WO2002039752A2
Application number: PCT/EP2001/013019
Authority: WO
Inventors: Jens Rennert; Ralph Escherich
Original assignee: Koninklijke Philips Electronics N.V.
Priority date: 2000-11-10
Filing date: 2001-11-06
Publication date: 2002-05-16
Also published as: KR20020067608A; JP2004514338A; WO2002039752A3; TW535419B

Abstract

The present invention provides a very hardware efficient and flexible way of performing color keying and color components evaluation operations. It also provides a way to significantly limit the visibility of artifacts introduced by filtering on images containing key color information. According to the present invention, there is provided a color key extraction apparatus for determining whether an incoming pixel having a plurality of color components matches a predefined key color. The apparatus comprises a plurality of evaluation circuits each configured to receive and evaluate one of the color components of the incoming pixel against a predefined color component intensity range and output a comparison signal; and a matching circuit, operably coupled to each of the evaluation circuits, that is configured to determine whether a combination of the comparison signals meets a predefined condition and output a match signal if the condition is met, whereby the incoming pixel matches the predefined key color.

Description

High quality color key extraction apparatus

The invention generally relates to color key extraction apparatus and more particularly to color key extraction apparatus for high quality applications.

Color keying is widely used and implemented in various video and graphics processors for various purposes, e.g., for superimposing one image on top of another. It is done by placing an unused color as a key color into a video or graphics image, and displaying a substituting color instead of the unused color. Usually key colors are magenta or other rarely used colors. Color keying is typically done in the mixing stage. An incoming pixel is evaluated to determine if it matches a certain color or color range. If so, certain actions, e.g., color key extraction and color substitution with a key color, will be performed. The evaluation of the incoming pixel and color substitution with a key color are usually done on a full 24 or 32 bit color information, as opposed to on the color components of the pixel. Thus, it is not as flexible as software programmers would like it to have. Furthermore, if any conversion or filtering occurs before the key color is extracted and replaced with a substituting color, the key color will be filtered into an active image portion. This will result in very visible artifacts in the final image.

Therefore, there is a need for providing high flexibility in evaluating incoming pixels and for significantly limiting the visibility of artifacts in the final image.

The present invention provides a very hardware efficient and flexible way of performing color keying and color components evaluation operations. It also provides a way to significantly limit the visibility of artifacts introduced by filtering on images containing key color information.

According to the present invention, there is provided a color key extraction apparatus for determining whether an incoming pixel having a plurality of color components matches a predefined key color. The apparatus comprises a plurality of evaluation circuits each configured to receive and evaluate one of the color components of the incoming pixel against a predefined color component intensity range and output a comparison signal; and a matching circuit, operably coupled to each of the evaluation circuits, that is configured to determine whether a combination of the comparison signals meets a predefined condition and output a match signal if the condition is met, whereby the incoming pixel matches the predefined key color.

According to one aspect of the invention, the matching circuit includes a storage element that stores a plurality of bits, the value of each bit indicating whether a particular combination of the comparison signals meets the predefined condition. The predefined condition for a given combination of the comparison signals is met when a bit in the storage element corresponding to the given combination is set. The matching circuit may be implemented with a very low cost, highly flexible programmable raster operation block. According to another aspect of the invention, each of the evaluation circuits comprises: a first circuit configured to receive a predetermined color component of the incoming pixel, mask the color component with a preselected mask value, and output a masked color component value; a second circuit configured to mask a preselected upper boundary value of the predefined color component intensity range with the mask value and output a masked upper boundary value; a third circuit configured to mask a preselected lower boundary value of the color component intensity range with the mask value and output a masked lower boundary value; a first comparator that compares the masked color component value with the masked upper boundary value and outputs a first result; a second comparator that compares the masked color component value with the masked lower boundary value and outputs a second result; and an output circuit configured to receive the first and second results and output a first comparison signal.

According to yet another aspect of the invention, there is further provided a color substitution circuit, operably coupled to the matching circuit, that is configured to replace an incoming pixel with a replacement pixel if the matching circuit outputs a matching signal. By exchanging a key color with a replacement color (e.g., a gray color) before any filtering, the resulting image has a change in the brightness, but not in the color tone.

Other objects and attainments together with a fuller understanding of the invention will become apparent and appreciated by referring to the following description and claims taken in conjunction with the accompanying drawings.

The invention is explained in further detail, and by way of example, with reference to the accompanying drawings wherein: FIG. 1 shows a functional block diagram of an exemplary system suitable for implementing the present invention;

FIG. 2 shows a color key extraction and color substitution circuit according to the present invention; and FIG. 3 shows a ROP (raster operation) table used for programming a programmable raster operation block in FIG. 2.

Throughout the drawings, the same reference numerals indicate similar or corresponding features or functions.

FIG. 1 shows a functional block diagram of an exemplary system suitable for implementing the present invention. As illustrated in FIG. 1, a digital video/graphics source 10 supplies digital video/graphics signals to a memory buffer 16. The digital data are stored in memory buffer 16 in a standard format. Upon command, a video/graphics data processor 20 receives the digital data from memory buffer 16 for processing. Processor 20 transmits the processed data to a video/graphics encoder 26 in a predetermined format. Encoder 26 then encodes the data and transmits them to a display device 30 for displaying.

FIG. 2 shows a color key extraction and color substitution circuit used in a graphics/ video pipeline in processor 20 according to the present invention. In FIG. 2, each incoming pixel has three color components (RGB or YUV) with 8 bits per component. These color components are evaluated by evaluation circuits 37, 137 and 237, respectively, on their color component intensities. Thus, circuit 37 is used for determining whether an incoming color component R/Y is within a predefined color component intensity range in a spectrum between the darkest R/Y and the lightest R/Y. Similarly, evaluation circuit 137 is used for determining whether an incoming color component G/U is within a predefined color component intensity range in a spectrum between the darkest G/U and the lightest G/U. Finally, evaluation circuit 237 is used for determining whether an incoming color component B/V is within a predefined color component intensity range in a spectrum between the darkest B/V and the lightest B/V. In evaluation circuit 37, the color component R/Y is masked bit by bit by a mask value in a ColorKeyAnd register 38 via an AND gate 40. This mask value is also used to mask bit by bit the values stored in ColorKey registers 42 and 48 via AND gates 44 and 50, respectively. The values in registers 42 and 48 are predefined upper and lower boundary values for establishing the predefined color component intensity range for evaluating the color component R/Y. As will be understood by those skilled in the art, the mask value is determined based on the applications, e.g., whether to key on a specific color or a color range, whether to pre-process the image, and the picture contents. Similarly, the upper and lower boundary values are also application dependent. The masked value at the output of AND gate 40 is evaluated against the masked upper and lower boundary values via comparators 46 and 52 to determine whether the masked color component is within the masked predefined color intensity.

If the masked value is less than or equal to the masked upper boundary value, comparator 46 will output a first indication signal (e.g., logic 1) to an AND gate 60. In a similar way, if the masked value is greater than or equal to masked lower boundary value, comparator 52 will provide a second indication signal (e.g., logic 1) to AND gate 60. If it is determined that the masked value is within the masked predefined color component intensity range, i.e., when both the first and second indication signals are provided to AND gate 60, AND gate 60 will output a first positive comparison signal (e.g., logic 1) for the color component R Y. Otherwise, AND gate 60 will output a first negative comparison signal (e.g., logic 0).

Evaluation circuits 137 and 237 operate in the same manner as evaluation circuit 37. If the color components G/U and B/V are each within their masked predefined color component intensity range, circuits 137 and 237 will output second and third positive comparison signals (e.g., logic 1), respectively. Otherwise, evaluation circuits 137 and 237 will output second and third negative comparison signals (e.g., logic 0), respectively.

It should be noted that in a default mode, the mask value is all 1 's, so that the mask operation is not effectively performed and the upper and lower boundary values define the color component intensity range. The comparison signals (Kl, K2, and K3) from evaluation circuits 37, 137 and

237 are provided to a programmable raster operation block 64. Depending on the particular combination of the comparison signals (positive or negative) and whether a specific bit value in a ColorKeyROP register 68 is set, block 64 may output a match signal (e.g., logic 1) to indicate that there is a match between the incoming color and the predefined key color. The detailed operation of block 64 is described below in connection with FIG. 3.

FIG. 3 is a ROP (raster operation) table used for programming block 64. For a given combination of Kl, K2 and K3, if the ROP bit assigned to the combination (i.e., appearing next to the combination) is set (logic 1), it indicates a match and causes block 64 to output a match signal. If the ROP bit is reset (logic 0), it indicates no match. According to the invention, one or more ROP bits may be set, depending on the applications. This scheme provides the flexibility of allowing one to manipulate the sensitivity of one or more color components. For example, one may be interested in a match of only two color components (e.g., R Y and G/U) for a particular incoming color and would not care about the third component (e.g., B/V). In such a case, ROP bit 6 and 7 will be set to indicate a match on the combinations "110" and "111" for Kl, K2 and K3 in comparison with the key color.

With reference to FIG. 2, after it is determined that there is a match between the incoming color and the key color, color substitution may be optionally performed by enabling an AND gate 72 via an enabling signal from a ColorKeyReplaceEn register 76. For example, if subsequent filtering is to be done, color substitution will be performed. To perform color substitution, a logic 1 will be output by AND gate 72 to activate multiplexers 86, 186, and 286 and select the replacement values stored in ColorKeyReplace registers 80, 180, and 280 for the R Y, G/U, B/V components, respectively to replace the original incoming values. As an example, a gray color may be preferred. Thus, the pixel values of the color components for a gray color are used as the replacements. By exchanging a key color by a replacement color (e.g., gray) before any filtering, the resulting image will have a change in the brightness, but not in the color tone. The outputs of multiplexers 86, 186 and 286 are provided to mixers (not shown) for combining with the background color.

On the other hand, if there is no match as determined by block 64, or if no color substitution is desired, a logic 0 is output by AND gate 72 and multiplexers 86, 186 and 286 directly output the incoming color components to the mixers without any modifications. The output of block 64 is also provided to the mixers.

After the key color is extracted, the data will go through conversion and filtering processes to limit visible artifacts in the final image. It should be noted that the present invention may also be used with other types of color components, such as Y, Cr, Cb.

While the invention has been described in conjunction with specific embodiments, it is evident that many alternatives, modifications and variations will be apparent to those skilled in the art in light of the foregoing description. Accordingly, it is intended to embrace all such alternatives, modifications and variations as fall within the spirit and scope of the appended claims.

Claims

CLAIMS:

1. A color key extraction apparatus for determining whether an incoming pixel having a plurality of color components matches a predefined key color, the apparatus comprising: a plurality of evaluation circuits (37, 137, 237) each configured to receive and evaluate one of the color components of the incoming pixel against a predefined color component intensity range and output a comparison signal; and a matching circuit (64, 68), operably coupled to each of the evaluation circuits, that is configured to determine whether a combination of the comparison signals meets a predefined condition and output a match signal if the condition is met, whereby the incoming pixel matches the predefined key color.

2. The apparatus of claim 1, wherein each of the evaluation circuits comprises: a first circuit (40) configured to receive a predetermined color component of the incoming pixel, mask the color component with a preselected mask value, and output a masked color component value; a second circuit (44) configured to mask a preselected upper boundary value of the predefined color component intensity range with the mask value and output a masked upper boundary value; a third circuit (50) configured to mask a preselected lower boundary value of the color component intensity range with the mask value and output a masked lower boundary value; a first comparator (46) that compares the masked color component value with the masked upper boundary value and outputs a first result; a second comparator (52) that compares the masked color component value with the masked lower boundary value and outputs a second result; and an output circuit (60) configured to receive the first and second results and output a first comparison signal.

3. The apparatus of claim 1, wherein the matching circuit includes a storage element (68) that stores a plurality of bits, the value of each bit indicating whether a particular combination of the comparison signals meets the predefined condition; and wherein the predefined condition for a given combination of the comparison signals is met when a bit in the storage element corresponding to the given combination is set.

4. The apparatus of claim 1, further comprising a color substitution circuit (80, 86, 180, 186, 280, 286), operably coupled to the matching circuit, that is configured to replace an incoming pixel with a replacement pixel if the matching circuit outputs a matching signal.

5. The apparatus of claim 4, wherein the color substitution circuit includes a plurality of multiplexers (86, 186, 286) each controlled via the matching signal and each configured to receive one of the color components of the incoming pixel and a corresponding replacement color component, and wherein upon receiving the matching signal each multiplexer outputs a corresponding replacement color component.

6. The apparatus of claim 4, further comprising an enabling circuit (72) configured to control enabling and disabling of the color substitution circuit.

7. A method for determining whether an incoming pixel having a plurality of color components matches a predefined key color, the method comprising the steps of:

(a) evaluating each of the color components of the incoming pixel against a predefined color component intensity range;

(b) outputting a comparison signal after evaluating each color component at step (a);

(c) determining whether a combination of the comparison signals meets a predefined condition; and (d) outputting a match signal ifthe condition is met, whereby the incoming pixel matches the predefined key color.

8. The method of claim 7, wherein step (a) comprises the steps of: masking a predetermined color component of the incoming pixel with a preselected mask value to produce a masked color component value; masking a preselected upper boundary value of the predefined color component intensity range with the mask value to produce a masked upper boundary value; masking a preselected lower boundary value of the color component intensity range with the mask value to produce a masked lower boundary value; comparing the masked color component value with the masked upper boundary value to produce a first result; comparing the masked color component value with the masked lower boundary value to produce a second result; and outputting a first comparison signal based on the first and second results.

9. The method of claim 7, wherein step (c) includes the step of providing a plurality of bits, the value of each bit indicating whether a particular combination of the comparison signals meets the predefined condition; and wherein the predefined condition for a given combination of the comparison signals is met when a bit in the plurality of bits corresponding to the given combination is set.

10. The method of claim 7, further comprising the step of replacing the incoming pixel with a replacement pixel if the matching signal is output.