Television – Format conversion – Field rate type flicker compensating
Reexamination Certificate
1998-10-23
2001-11-13
Eisenzopf, Reinhard J. (Department: 2614)
Television
Format conversion
Field rate type flicker compensating
C348S607000, C382S299000
Reexamination Certificate
active
06317158
ABSTRACT:
BACKGROUND
An interlaced video signal is a common format for video information. In an interlaced video signal, each image or frame is defined by a set of lines of pixels. These lines are separated into two fields such that adjacent lines in the image or frame are in separate fields. Odd lines define an odd field; even lines define an even field. In an interlaced video display, only one field of each image is displayed each time the display is refreshed, alternating between odd and even fields within each image. Because the two fields usually closely resemble each other, and because the fields are displayed at twice the frame rate, a human eye generally does not perceive that each field actually is refreshed at the frame rate and thus does not perceive flickering.
Although an interlaced video signal generally provides good quality images and sufficient bandwidth, various artifacts may occur when a time varying effect is added to the interlaced video, particularly if the effect involves positioning an image, such as a progressive scan image, into the interlaced video. An input image may be positioned into the interlaced video signal by selecting, for each line in the output image, a corresponding line in the input image and a fraction indicating a subpixel position (or offset) into that line and the next (or previous) line in the input image. The intensity, color and brightness of each pixel in the output line is determined as a function of the subpixel position and the corresponding pixels in the selected lines in the input image. Thus, two lines in two different fields in the input image typically contribute to each line in an output field. This technique may be called “unfiltered subpixel positioning.” With unfiltered subpixel positioning, various artifacts may occur. Such artifacts include detail areas that appear to flicker (called twitter), such as with credits or other small text, or detail areas that appear and disappear over time.
The artifacts caused by unfiltered subpixel positioning may be reduced by positioning the input image into the interlaced video by determining the intensity, color and brightness of each pixel in the output line as a function of the subpixel position and the corresponding pixels in the corresponding line in one field of the input image and both of the adjacent lines in the other field of the input image. Thus, three lines in two different fields in the input image typically contribute to each line in an output field. This technique may be called “filtered subpixel positioning,” and has at least two disadvantages. First, the input image becomes slightly blurred. Second, each output line requires additional processing and storage, in comparison to unfiltered subpixel positioning, due to the use of three input lines.
SUMMARY
When processing a time varying effect that positions an input image into an interlaced output image, the input image is filtered first. The filtered input image then is positioned in the output image by mapping one line in the output image to one line in the filtered input image space. The input image may be filtered once and the filtered input image may be used to create all output fields for the effect. Thus, the overhead per output field used to create the output field from the input image is reduced in comparison to filtered subpixel positioning. In addition, equipment that is capable of performing only unfiltered subpixel positioning may use the filtered input image to reduce interlaced artifacts.
Accordingly, in one aspect, a process generates an interlaced video output frame comprising two fields from an input image according to an effect. In this process, the input image is filtered. A subpixel position in the filtered input field is determined for each output field according to the effect. The filtered input image is positioned in each output field according to the determined subpixel position. In one embodiment, the filtered input image is stored. The stored filtered input image is then positioned in the output field. In another embodiment, the subpixel position is a vertical position of each line of the filtered input image in the output image. Each pixel value in a line of the output field is computed according to a function of pixel values in two adjacent lines in the stored filtered input image and the vertical position.
In another aspect, a process generates a sequence of interlaced video output frames, wherein each output frame comprises two fields, from an input image according to an effect which varies over time. The input image is filtered and stored. For each field of each output frame, a subpixel position in the stored filtered input field is determined according to the effect, a position of the field in the output frame and a position of the output frame in the sequence. The stored filtered input image is positioned in the field of the output frame according to the determined subpixel position.
In yet another aspect, a process generates a sequence of interlaced video output frames, wherein each output frame comprises two fields, from an input image according to an effect which varies over time. For each field of each output frame, the input image is filtered. A subpixel position in the filtered input image is determined according to the effect, a position of the field in the output frame and a position of the output frame in the sequence. The filtered input image is positioned in the field according to the determined subpixel position.
In other embodiments of these aspects, the filtering may be achieved by averaging adjacent lines in the input image. Alternatively, filtering may be achieved by applying a low pass filter to the input image. Filtering also may involve applying a nonlinear filter to the input image. The image generally is filtered vertically but also may be filtered horizontally.
In other embodiments, the subpixel position may be determined by computing a vertical position of each line of the filtered input image in the output image according to the effect. In this embodiment, each pixel value in a line of the output field is computed according to a function of pixel values in two adjacent lines in the filtered input image and the computed vertical position.
In other embodiments, the filtering applied to the input image has an associated offset. The subpixel position is adjusted according to the offset. The filtered input image is positioned in each output field according to the adjusted subpixel position.
Another aspect is an editing system or other apparatus that performs such a process. The editing system also may include an effects processing system and a display. Another aspect is a computer program product in which computer program code is stored on a computer readable medium, wherein the computer program code. when executed, performs such a process. The process also may be performed by a combination of computer program code executed on a general purpose computer and one or more peripheral devices of the computer. Another aspect is a file format either for a video program including such an effect or for the effect, wherein the effect, the filtered image and any offset caused by the filtering are stored.
REFERENCES:
patent: 4603350 (1986-07-01), Arbeiter et al.
patent: 5097257 (1992-03-01), Clough et al.
patent: 5257102 (1993-10-01), Wilkinson
patent: 5280351 (1994-01-01), Wilkinson
patent: 5291280 (1994-03-01), Faroudja et al.
patent: 5892551 (1999-04-01), Uematsu
patent: 6069984 (2000-05-01), Sadler et al.
patent: 6094511 (2000-07-01), Metcalfe et al.
patent: 0 610 701A1 (1994-08-01), None
patent: 0 790 737A1 (1997-08-01), None
Wolberg, G., “Digital Image Warping”, IEEE Computer Society Press, Los Alamitos, CA, 1990, Chapter 5, Image Resampling, pp. 117-161.
Poynton, C.A., “A Technical Introduction to Digital Video”, John Wiley & Sons, Inc., 1996, pp. 11-13 and 43-79.
Abstract, “Advantages and Drawbacks of Interlaced and Progressive Scanning Formats”, Jun. 1995, p. 1 of 1.
Wilt, A., “Technical Difficulties—Fields and Frames”, DV.com. Magazine, Dec. 2000, pp. 1-6.
Guillotel P. et a
Avid Technology Inc.
Eisenzopf Reinhard J.
Gordon Peter J.
Tran Trang U.
LandOfFree
Method and apparatus for positioning an input image into... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Method and apparatus for positioning an input image into..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method and apparatus for positioning an input image into... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2570863