Television – Image signal processing circuitry specific to television – Motion dependent key signal generation or scene change...
Reexamination Certificate
1999-12-30
2003-11-25
Philippe, Gims S. (Department: 2613)
Television
Image signal processing circuitry specific to television
Motion dependent key signal generation or scene change...
C375S240160, C382S236000
Reexamination Certificate
active
06654067
ABSTRACT:
FIELD OF THE INVENTION
The field of the present invention relates to video image processing and, more particularly, to the detection of short-duration scenes (or objects within scenes) that appear within a longer video sequence.
BACKGROUND OF THE INVENTION
Virtually all conventional moving video images are generated by displaying a series of still images one after another in rapid succession. Each of these still video images is commonly referred to as a frame. When the frame rate is low (e.g. five frames per second), the resulting video image will flicker objectionably. But when the frame rate is increased sufficiently, the resulting sequence of still images will be perceived by human observers as a smooth, flicker-free moving image, due to the normal operation of the human eye and brain. In the United States, a commonly used frame rate that provides flicker-free moving video images is thirty frames per second (fps).
At the standard 30 fps rate, however, video information can be inserted into a single frame of a video clip, and still remain undetectable to ordinary observers. This effect occurs because the video information in a single frame is not displayed for a sufficiently long time to be noticed by viewers. When the video clip is displayed in slow motion or stop motion, however, the otherwise undetectable video information is revealed. This situation provides an opportunity for disgruntled workers and practical jokers to insert inappropriate and potentially embarrassing material (e.g. pornography) into moving video images.
The standard 30 fps rate also provides an opportunity to insert subliminal messages into moving video images that can not be detected by ordinary observers. In fact, patents have even been issued (e.g., U.S. Pat. No. 5,017,143) describing ways to produce this type of subliminal message.
For quality control purposes, producers of moving video images videos may wish to check that no inappropriate, embarrassing, or subliminal images have been included in their products, even for a single frame. Until now, however, the only way to check for these undesired images has been to watch the entire video one frame at a time. But because frame-by-frame checking is a painstakingly slow process, it is commonly skipped, and products are released without checking for undesired frames of video information.
Numerous techniques for finding a point in a moving video image where one scene ends and another scene begins (hereinafter referred to as a “scene cut”) can be found in the prior art. Examples include International Publication WO 98/23085; “A hierarchical multi-resolution video transition detection scheme” by H. Yu and W. Wolf, to appear in the Journal of Computer Vision and Image Understanding, Academic Press; “A Feature-Based Algorithm for Detecting and Classifying Scene Breaks” by R. Zabih, J. Miller, and K. Mai, Proc. ACM Multimedia, San Francisco, 1995, pp. 189-200; and “Automatic Partitioning of Full-Motion Video” by H. J. Zhang, A. Kankanhalli, and S. W. Smoliar, Multimedia Systems, ACM-Springer, Vol. 1, No. 1, 1993, pp. 10-28, each of which is incorporated herein by reference.
The prior art also discusses detecting when a region within a scene changes, even when the remainder of the scene remains constant (hereinafter referred to as a “region cut”), e.g., in “Video Keyframe Extraction and Filtering: A Keyframe is not a Keyframe to Everyone” by N. Dimitrova, T. McGee, and H. Elenbaas, Proc. ACM Conf. on Knowledge and Information Management, 1997, pp.113-120, which is incorporated herein by reference.
Techniques for eliminating undesirable video information from a moving video image are also described in the prior art. For example, “Parsing TV Programs for Identification and Removal of Non-story Segments” by T. McGee and N. Dimitrova, presented at Storage and Retrieval for Image and Video Data Bases VII in San Jose, Calif., in January 1999 describes removing non-story segments (e.g., commercials) from within the broadcasted material. These prior art techniques, however, are not directed at finding and eliminating short-duration segments of video information that may be hidden within a longer video work.
SUMMARY OF THE INVENTION
The present invention relates to finding and reporting short-duration portions of video information that are contained within longer video works.
The present invention operates by locating cuts that occur in rapid succession. When two cuts in rapid succession are located, the video information contained between those two cuts is reported by, for example, displaying it on a monitor.
One aspect of the present invention relates to a method of finding short-duration images within a series of frames of video data. The method includes the steps of locating a first cut in the video data, locating a second cut in the video data, determining a number of frames of the video data that are located between the first cut and the second cut, and outputting an indication when the number of frames determined in said determining step is less than a threshold.
Another aspect of the present invention relates to a method of finding short-duration images within a series of frames of video data. The method includes the steps of locating a first cut in the video data corresponding to a first time, locating a second cut in the video data corresponding to a second time, and outputting an indication when a difference between the first time and the second time is less than a threshold.
REFERENCES:
patent: 5017143 (1991-05-01), Backus et al.
patent: 5083860 (1992-01-01), Miyatake et al.
patent: 5493345 (1996-02-01), Ishikawa et al.
patent: 6341168 (2002-01-01), Nagasaka et al.
patent: 0690413 (1996-01-01), None
patent: WO9823085 (1998-05-01), None
patent: WO9823087 (1998-05-01), None
Abstract for JP09074575A dated Mar. 18, 1997.
Abstract for RD 408046 dated Apr. 10, 1998.
Dimitrova Nevenka
Mc Gee Thomas
Koninklijke Philips Electronics , N.V.
Philippe Gims S.
LandOfFree
Finding hidden images in video based on detected proximity... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Finding hidden images in video based on detected proximity..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Finding hidden images in video based on detected proximity... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3160653