Television – Bandwidth reduction system – Data rate reduction
Reexamination Certificate
1998-02-23
2001-02-27
Peng, John K. (Department: 2714)
Television
Bandwidth reduction system
Data rate reduction
C348S398100, C348S396100, C348S390100, C348S607000, C382S275000, C382S237000
Reexamination Certificate
active
06195128
ABSTRACT:
This invention relates to a system for a method of processing digital video information in which inter alia video digital information is preferably read and captured in real time and then compressed. The invention also relates to a method of and system for decompressing and replaying such compressed video. The video can either be stored on a computer storage medium such as, but not be limited to, a CD-ROM, for future use or can be transmitted through a communications channel to a receiving station.
BACKGROUND
In WO95/16325 we described a method of processing digital video information and the present invention seeks to provide improvements of the method and techniques described in this publication.
WO94/14139 describes a method of reducing noise in digital video information which involves a frame-by-frame comparison of pixel values to find a key frame which is maintained until more than predetermined portion of the pixels have changed whereupon another frame becomes the key frame.
SUMMARY OF THE INVENTION
In one aspect of the invention there is provided a method of processing digital video information in an adapted format for transmission or storage, said method comprising:
reading digital data representing luminance and chrominance of individual picture elements (pixels) of a current video image frame as a series of binary coded words and idealising the data to reduce the affects of noise by evaluating at least preselected values of luminance of corresponding pixels over a number of frames which occur after said current frame to identify and choose the particular value which remains within a predetermined range of variations from the actual values of corresponding pixel luminance over the optimally greatest number of frames and which is not biased for several frames towards either one of the limits of the range for the optimally greatest number of frames to represent a fixed value of pixel luminance chosen for that optimal number of frames and where there is more than one particular luminance value thus chosen, further selecting the value which differs the least from the actual values of corresponding pixel luminance over the optimal number of frames following the current frame to represent the fixed value of luminance chosen for that optimal number of frames. These steps can be repeated for chrominance values and the evaluation and selection of both luminance and chrominance values can take place for further optional sets of frames. These fixed representation values can then be subjected to compression. Such a technique can involve deriving from the words representing pixel luminance and chrominance further words each describing first and second groups of pixels of which the first group is a block of 8×8 pixels and the second group is a block of 2×2 pixels derived from the first group, and each block of the second group is characterised as a single entity code word for further processing by averaging the chrominance values of all four pixels and by selecting just a pair of luminance values to represent all the four luminance values and by providing an indication of the particular luminance value to be adopted for each pixel. The invention also provides a method of processing digital video information in an adapted format for transmission or storage after compression; said method comprising:
reading digital data representing individual picture elements (pixels) of a video image frame as a series of binary coded words;
deriving from the words representing individual pixels further words each describing first and second blocks or groups of pixels of which the second group is a subset of the first group and includes a representation of the luminance of 2×2 pixel blocks in which a pair of luminance values is adopted together with an indication of which of these two values is to be chosen for each pixel;
comparing and evaluating the words of the second groups representing corresponding portions of one frame with another frame or other frames in a predetermined sequential order of the elements making up successive first groups to detect whether a significant change has occurred;
using the comparison to take the second group to represent all the individual pixels described by the group whenever the detected change is significant and
using variable length code words to represent the number of consecutive components of the second groups without significant change;
wherein the comparing and evaluating step comprises initially ascertaining whether the indications have changed to signify that a significant change has occurred and if not then information is derived by combining the pairs of luminance values and processing such combined values to produce fresh code words including the respective differences between the maximum and minimum luminance values and combining the fresh code words with said indications to produce further code words which are then processed to produce said information which represents the significance of the difference between the luminance values in the 2×2 pixel blocks being compared and utilizing this information to decide whether the significance of the difference should be taken as a significant change.
As appears hereinafter, the derived code words can be subjected to various processes, to achieve compression, some of which are defined in the dependent claims. Identical code words within each first block can be positionally correlated with a mask and a common code word stored just once for each block. During subsequent decompression the data representing the second blocks of pixels can be decompressed by converting the single entity code words into respective RGB pairs suitable for creating a display on a screen, each RGB pair being determined by the corresponding pair of luminance values in the associated single entity code words together with said indication which is taken to specify which of the RGB values of the RGB pair is to be adopted for each pixel in the subsequent re-constituted 2×2 pixel block.
EXAMPLE APPLICATIONS
In the digital domain, any application which can make use of video will be enhanced by the ability to process compressed video. Efficient compression and decompression is important in a number of applications, for example those outlined hereunder. Provided below are some descriptions, by way of example only, of how the invention in some of its various forms could be applied.
PSTN Videophone
In one version the invention can be used to provide the video processing required for a videophone (video telephone) through PSTN (analogue) telephone lines. In this case, the video is compressed in real time for transmission. Because of the constraints on bandwidth, the quality is not expected to be very high. For example. 160*120 pixel resolution, 10 frames per second and 32 intensity levels would be one example of image quality, although colour is possible. At this frame rate, there is considerable processor time both to improve compression and playback. Typical data rates are between 9.6 kb/s and 30 kb/s. In this case, a hardware based solution such as H.320 would be uneconomical, and current widely owned standard alternatives cannot deliver the required compression rates.
ISDN Videophone
In another implementation, the invention can be used to produce higher quality images and higher frame rates than in the PSTN videophone. These pictures may include colour. Typical data rates are between 38.4 kb/s and 128 kb/s. Other solutions exist to this problem (for example H.261), but they are typically much more costly to implement.
Radio Communications
In a further implementation of the invention, data rates of around 256 kb/s could be transmitted by radio to a receiving station. Because the codec can be implemented in software on a low cost computer system, it is relatively straightforward to add extra features, such as error correction, which may be needed over a noisy radio link. The radio communications could be used, for example, for live news or live sport transmissions to an aircraft. Hardware based solutions would not
Desir Jean W.
Eidos Technologies Limited
Peng John K.
Sughrue Mion Zinn Macpeak & Seas, PLLC
LandOfFree
Video processing for storage or transmission does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Video processing for storage or transmission, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Video processing for storage or transmission will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2603755