Television – Image signal processing circuitry specific to television – Chrominance-luminance signal separation
Reexamination Certificate
1998-12-17
2001-06-05
Eisenzopf, Reinhard J. (Department: 2614)
Television
Image signal processing circuitry specific to television
Chrominance-luminance signal separation
C348S667000, C348S712000
Reexamination Certificate
active
06243144
ABSTRACT:
FIELD OF THE INVENTION
The invention relates generally to systems and methods for compensating against false color from composite video information, and more particularly to systems and methods for compensating against false color occurrences using adaptive filters.
BACKGROUND OF THE INVENTION
It is well known that video encoders typically combine luminance and chrominance information by adding them together. The result is that the chrominance and high frequency luminance signals occupy the same portion of the frequency spectrum. Consequently, video decoders typically use some form of frequency separation filtering to separate the luminance information from the chrominance information in the composite video source information. When some luminance information is decoded as color information, cross color or false coloring conditions can occur.
Many luminance and chrominance separators are known. One type is a two dimensional (2-D) adaptive comb filter. Such separators are typically used because conventional comb filters have problems with diagonal lines and vertical color changes. Typically, with diagonal lines, after luminance and chrominance separation, the chrominance information may also include the difference between adjacent luminance values which may be interpreted by a decoder as chrominance information. The result may be false color artifacts along the edge of a line. A general discussion of an example of two-dimensional adaptive luminance and chrominance separators may be found for example in a book entitled “Video De-Mystified” authored by Keith Jack (1997), pages 294-298. Other examples of various filters may be found on pages 179 and pages 288301.
Conventional 2-D adaptive luminance and chrominance separators typically look at vertical chrominance data over multiple lines and also evaluate for horizontal chrominance information. Such filters are adaptive in that they evaluate if there is a difference between vertical and horizontal chrominance information. However, a problem arises with these separators because these separators typically choose horizontal chrominance information when there is a difference. An additional problem arises if the video image is black and white diagonal luminance information, since the luminance information can still bleed into the chrominance data.
The use of notch filtering is also known to notch out luminance information near the color burst frequency, such as about 3.58 MHz for NTSC video. However, notching out the requisite color burst frequency (hence chrominance information) also notches out luminance information at overlapping frequencies. This can result in distorted images such as a black and white striped shirt appearing gray to an observer.
Video graphic controllers and other video processing devices are known that provide false color filtering in an effort to reduce video display degradation. Video filters may encode signals that are used to send out to television monitors or other monitors such as signals in the format of NTSC or PAL composite video signals. In the digital color space, as known in the art, luminance information may be represented as Y, and color (chrominance) information may be represented as Cb, U, Pb or Y-B and Cr, V, Pr or Y-R, as appropriate. The chrominance information has a frequency spread at about 3.58 MHz. The NTSC and PAL standards use a phase change of the color information from line to line to reduce false coloring. A problem arises when graphics is displayed on a television tuner that normally receives and decodes television signals. Hard edge graphics produce a lot of false coloring to the naked eye. Therefore it is desirable to reduce false color without reducing text legibility excessively. Many television filters mistake the fine detail of text for color images.
Comb filters attempt to see if line to line averaging yields false coloring and if so, lines are blended to reduce false coloring. However, this can result in the loss of resolution and the human eye will perceive blurs along the edges of sharp text. Comb filters typically can remove false color with video signals, however, with text such as the letter “W” the averaging of line to line may produce false color because if the top and bottom of a text character is bounded by a blank line, there may be nothing to average. Other techniques such as “clean encoding” attempt to remove the false color before image information is sent. However, such systems are typically expensive to implement.
Consequently, a need exists for an improved video filter that attempts to adapt to changes in the video information to compensate for false color.
REFERENCES:
patent: 4240105 (1980-12-01), Faroudja
patent: 4617589 (1986-10-01), Weckenbrock
patent: 4731660 (1988-03-01), Faroudje et al.
patent: 5097322 (1992-03-01), Fairhurst
patent: 5663771 (1997-09-01), Raby
patent: 5841934 (1998-11-01), Chmielewski et al.
patent: 5870153 (1999-02-01), Murray et al.
Video Demystified—A Handbook for the Digital Engineer, Keith Jack, pp. 179, 298-301, 1996.
ATI International SRL
Eisenzopf Reinhard J.
Tran Trang U.
Vedder Price Kaufman & Kammholz
LandOfFree
Adaptive false color video filter and method does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Adaptive false color video filter and method, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Adaptive false color video filter and method will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2486791