Television – Bandwidth reduction system – Data rate reduction
Patent
1995-10-04
1999-08-17
Tung, Bryan
Television
Bandwidth reduction system
Data rate reduction
H04N 732
Patent
active
059401303
DESCRIPTION:
BRIEF SUMMARY
BACKGROUND OF THE INVENTION
1. Field of the Invention
This invention relates to transcoders for the conversion of signals between a first and a second coding scheme. The invention is particularly suitable for converting video signals.
2. Related Art
There are many occasions when it is necessary to transmit moving picture television over long distances via a transmission link. Broadcast quality television requires in excess of 100 Mbit/s when transmitted in digital form which is expensive to transmit and requires links of high bandwidth. An acceptable degree of degradation in the picture quality can be introduced to reduce the information content being transmitted. Additionally or alternatively, compression coding techniques may be used which take advantage of the high degree of spatial and temporal redundancies in the video signals being encoded. So, for example, for video conferencing applications compression down to a bit rate of a few hundred kbit/s is possible whereas videophone-quality pictures including sound can be compressed down to only 64 bit/s, equivalent to a single telephone circuit.
Redundancy reduction techniques assume there is spatial and/or temporal correlation between neighbouring pixels or blocks of pixels. The details of correlation are encoded as well as the differences between the assumptions and the actual pixels or blocks. Typically each frame of an image to be coded comprises an array of picture elements (pixels) which are divided into blocks of N.times.M pixels.
Predictive coding exploits the assumption that a value within a frame is related to some neighbouring values, in the same or a different frame, and the value may therefore be calculated at the receiver instead of being transmitted. It is only necessary to transmit the prediction error arising from such an assumption. For instance the first pixel of a frame may be transmitted exactly whilst each subsequent pixel is transmitted as a difference from its predecessor. In more complex schemes the prediction may be a combination of a number of pixels.
Transform coding exploits the correlation of pixel magnitudes within a frame by transforming the magnitudes into another set of values, many of which are relatively small and can therefore be coded using fewer bits. The most common form of transform coding uses the Discrete Cosine Transforn (DCT). A block of N.times.M pixels is transformed into an array of N.times.M transform coefficients. The resulting array of coefficients is then quantised by dividing each coefficient by a variable quantisation factor. The quantised coefficients may be coded in variable length code, for instance a Huffman code.
Another coding technique is motion compensation in which a picture is divided into blocks of pixels and each block of the current frame is compared with the corresponding block of a reference frame, which may be a previous or a subsequent frame, and with regions shifted in position from that block, and that region of the reference frame which the block most closely resembles is identified.
The vector difference in position between the identified region and the block in question is termed a motion vector and is used to shift the identified region of the reference frame into the position of the relevant block in the current frame. Motion vectors are generated for all the blocks of the current frame and these are used to derive a predicted frame from the reference frame. The differences between the current and predicted frame are, on average, smaller than those between the current and reference frame and can be encoded using a lower bit rate. A decoder which already has the reference frame stored can thus reproduce the current frame using the motion vectors and the difference values. A signal may be coded using any of the aforementioned coding techniques, either separately or in combination.
There are circumstances when it is desirable to employ a transcoder which can accept a received data stream encoded according to a first scheme and output an encoded data stream encoded according to a s
REFERENCES:
patent: 5072295 (1991-12-01), Murakami et al.
patent: 5228028 (1993-07-01), Cucchi et al.
patent: 5278646 (1994-01-01), Civanlar
patent: 5291284 (1994-03-01), Carr et al.
patent: 5293229 (1994-03-01), Iu
patent: 5367336 (1994-11-01), Kustka
patent: 5446495 (1995-08-01), Tourtier et al.
patent: 5537440 (1996-07-01), Eyuboglu et al.
patent: 5544266 (1996-08-01), Koppelmans et al.
Communications of the ACM, vol. 34, No. 4, Apr. 1, 1991, New York, USA, pp. 103-112; M. Leibhold et al: "Toward an Open Environment for Digital Video".
Patent Abstracts of Japan, vol. 16, No. 520 (P-1444), Oct. 26, 1992 & JP-A-04 192 080 (Mitsubishi Electric Corp.) Jul. 10, 1992 (Abstract).
Patent Abstracts of Japan, vol. 7, No. 10 (P-18) Jan. 14, 1983 & JP-A-57 166 640 (Hitachi Seisakusho KK) Oct. 14, 1982 (Abstract).
17th International Television Symposium, Broadcast Sessions, Jun. 13, 1991, Montreux, CH, pp. 552-558; B. Pank: "Picture conversion of HD Graphics".
18th International Television Symposium, Broadcast Sessions, Jun. 10, 1993, pp. 432-447; David et al: "A Third Generation Algorithm For Use in High Definition Frame Rate Convertors".
Granbari Mohammed
Morrison David Geoffrey
Nilsson Michael Erling
British Telecommunications public limited company
Tung Bryan
LandOfFree
Video transcoder with by-pass transfer of extracted motion compe 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 transcoder with by-pass transfer of extracted motion compe, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Video transcoder with by-pass transfer of extracted motion compe will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-319823