Television – Format – Including additional information
Reexamination Certificate
1998-11-06
2001-04-10
Peng, John (Department: 2714)
Television
Format
Including additional information
C348S460000, C348S461000, C348S463000, C348S465000, C348S421100, C348S398100
Reexamination Certificate
active
06215526
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Technical Field
The invention relates to the encoding of data into analog television broadcast signals. More particularly, the invention relates to the in-band encoding of informational data into an analog television broadcast signal.
2. Description of the Prior Art
A large percentage of consumers now own Video Cassette Recorders (VCR). Almost every one of those consumers have encountered a common problem when recording broadcast programs. A consumer will set her VCR to record a television program that is either live or pre-recorded. The program is delayed or is broadcast early for some reason and the VCR records only a portion of the desired program.
The classic example concerns live broadcast programs such as boxing matches or football games. Boxing matches are notoriously known for unpredictable ending points. It is common for boxing matches to end unexpectedly because of an early knockout. Football, as well as baseball, games are delayed due to weather or extended because of overtime play. It is all too common for the last few moments of a game to be cut off when a consumer is recording it because the VCR is time based and ends the recording at a preset time.
The current approach of broadcast video recording uses time as the basis for the start/stop capture of programs. This is acceptable assuming: 1) there is an accurate time base available; 2) the data about start and stop times are accurate; and 3) the actual transmission of the program matches its stated transmission time. Even so, program guide times are accurate only to the minute, so precise capturing is difficult. In fact in such cases consumers may start capture early and end late to insure getting everything.
It would be far better if the program material included a “tag” directly within it indicating when a program begins or ends. Thus, it would be possible to capture programs with “frame accuracy,” i.e., with sufficient precision to capture the program beginning with the first frame of the program and ending with the last frame. Such a tag might include identifying information, or in fact any information of interest. In fact, the tag could mark any event of interest in the program, not just the beginning or ending of the program. For example, the beginnings and/or endings of commercial advertisements or special segments of the program might be so tagged.
The typical approach is to embed the tag into the Vertical Blanking Interval (VBI) of a succession of frames in the program. However, this approach does not guarantee that the VBI information will remain intact after transmission. The VBI is not considered part of the program material, so during the preparation and transmission of the program, the VBI may be modified in arbitrary ways. The only way to insure frame-accurate tagging using the VBI is to insert it directly into the program content at the last stage before transmission. This is difficult to achieve and expensive because it requires special equipment to be maintained at the broadcaster's site which can precisely identify the proper insertion points in the transmitted program and has available the full plurality of data that might be inserted. It is also difficult to manage because broadcasters must use a consistent format and notation for identical programs.
Similarly, the tag might be encoded into the audio portion of the signal. The audio signal is treated separately from the video portion through all stages of the preparation and transmission process, so it suffers from many of the weaknesses of VBI encoding. Further, audio encoding can only be accomplished by modifying the signal significantly so that the tag can be reliably retrieved, which results in unsatisfactory degradation of the final sound. Alternatively, a long time period can be used to send the tag, allowing minimal degradation of the audio signal, but this time period far exceeds the maximum latency required to achieve frame-accurate tagging.
A rich set of examples can be found in methods implementing automated broadcast verification systems, which use tagged video to trigger various mechanisms which record the tag for later reporting. Such systems are typically used for verifying that advertisements have been properly broadcast. A weakness of these systems is that the tag is usually out-of-band, placed either in the VBI, as described in U.S. Pat. No. 4,805,020 issued to Greenberg on Feb. 14, 1989, or in the audio portion of the signal, as described in U.S. Pat. No. 4,945,412 issued to Kramer on Jul. 31, 1990. These techniques suffer from the problems noted above.
Additionally, broadcast verification systems provide a rich source of examples for implementing in-band tagging. These typically break down into two approaches: signature recognition, in which the received signal is processed using various techniques to produce a “signature,” which is stored for later comparison with the original source signal; and embedding information into the signal by slightly modifying the signal. Signature recognition methods rely on comparison with a central database of signatures, which is infeasible within the scope of this invention. Embedding methods are computationally complex and expensive, and often fail in the face of normal transformations that occur during the transmission process.
It would be advantageous to provide an analog video tagging and encoding system that gives the programmer the ability to encode data in-band into the analog video stream at any point during the preparation and transmission process. It would further be advantageous to provide an analog video tagging and encoding system that utilizes a tagging approach that survives the changes that occur during the transmission process, including conversion to and from analog and digital television formats, and allows frame-accurate capture upon reception.
SUMMARY OF THE INVENTION
The invention provides an analog video tagging and encoding system. The invention provides a robust tagging system that encodes in-band data into an analog video broadcast signal. In addition, the invention allows frame-accurate capture of video frames upon reception.
A preferred embodiment of the invention uses the standard process of conversion of analog television signals to digital television signals in the Moving Picture Experts Group (MPEG) format to allow detection and capture of encoded tags. An analog tag frame is inserted into the analog video stream such that the luminance (Y) values are made to approach zero, i.e., black, allowing a potential tag frame to be quickly identified. Tag information is then encoded as a structured color pattern into the chrominance (Cb and Cr) portions of the frame, making the tag itself invisible to the casual viewer. The tag information can contain program start/stop information, program segment identification, or data downloads. The color pattern is chosen such that when it is converted into digital form according to the MPEG standard, the digital signal so produced may be easily and minimally parsed to recognize the presence or absence of a tag within a normal frame time period.
In a simple embodiment of the invention, the structured analog color pattern may appear as a pattern of colored blocks aligned with the macroblock segmentation performed on the analog signal by the MPEG encoding process. This causes the DC component of the encoded Cr and Cb macroblocks to precisely reflect the color pattern. The simplest color pattern uses two colors with extremely different Cr or Cb values, and derives a binary encoding sequence where a binary zero (0) might be represented by values near one color, and a binary one (1) represented by values near the other color. This is easily extended to create a number of different chrominance levels within each encoded macroblock, allowing multiple bits of tag information to be encoded by each macroblock. The number of chrominance levels available for data encoding are limited by the transmission accuracy and color distortion caused by the transmission process.
Another embodiment of the invention
Barton James M.
Look Howard
Desir Jean W.
Glenn Michael A.
Peng John
TiVo Inc.
Wong Kirk
LandOfFree
Analog video tagging and encoding system does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Analog video tagging and encoding system, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Analog video tagging and encoding system will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2464846