Pulse or digital communications – Bandwidth reduction or expansion – Television or motion video signal
Reexamination Certificate
1997-05-07
2001-10-16
Lee, Young (Department: 2613)
Pulse or digital communications
Bandwidth reduction or expansion
Television or motion video signal
Reexamination Certificate
active
06304605
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
The invention relates to a video compression method, and especially to a video compression method that is intended to be applied in the transmission of a video picture as a digital video signal at very low transmission rates.
2. Description of the Prior Art
A video picture is digitized by dividing it into pixels and by giving the pixels digitized values. In a black and white picture, the value of a pixel can simply be the value of brightness of the corresponding pixel in digital form, for example, given in 8 bits. Several signals are required in order to show a color picture, and thus, the digital presentation requires, for example, Y, U and V pixels and their digitized values, which contain the information about the brightness and colors of the picture. The invention is applied in the same way in the compression of all pixel information, and that is why this application primarily refers to pixels, pixel values and value information in general.
Because of the limited capacity of the channels used specifically for transferring digital picture information, the information must be compressed before transfer. For example, in practice, when transferring a normal television picture on a UHF channel, which has a transfer capacity of 32 Mbit/s, a compression of picture information is required in a ratio of 1:10-1:20. When a video picture is transferred on a channel, which has a very low transfer rate, for example, 8 kbit/s, a very efficient compression is required even if some of the picture quality is sacrificed.
Several different techniques have been developed for compressing a video picture. These techniques are used in coding the picture information in the transmission encoder and, correspondingly, in decoding the picture information and reconstructing the picture at the receiving end. These methods include, for example, variable length coding (VLC), predictive coding, movement compensation, run-length coding, and transform coding, such as discrete cosine transform coding (DCT). To make calculations easier, a picture is usually coded in blocks. A generally used block size in 8×8 pixels. The above-mentioned methods are familiar to the persons skilled in the art, and even though some of the methods, for example, variable length coding, can also be used in the video compression method of the invention to enhance the compression, they are not described in detail here, because it is not necessary in order to understand the invention.
When transferring a video picture as a digital video signal at low transfer rates, a generally used compression method is one, in which the picture is divided in blocks of n×m pixels. The blocks are compared to the corresponding blocks of the previously processed picture. The changed blocks are identified, and their coded information and the address data, which indicate their location, are transferred. At the receiving end, this information and the information of the said previously processed picture are used to reconstruct the picture. This compression method is very economical when transferring a video picture at low transfer rates, and that is why the method of the invention is considered to be preferably applied, but not in any way limited, to this application.
The next explains the conditions of transferring a video picture in a case, in which the available capacity of the transfer channel is very low, and in which it is economical to apply the method of the invention. The QCIF resolution of the picture is 176 pixels/line, and the picture has 144 lines. The picture is divided into macro blocks of 16×16 pixels, the total amount of which is 9×11, that is 99. Each macro block contains four Y blocks (8×8 pixels) and one U block and one V block (8×8 pixels). So, the picture has a total of 4×99+2×99=594 blocks. If we assume that the capacity of the channel is 8 kbit/s and the frequency of the picture is 8.3 pictures per second, there are 963 bits available per picture. If we assume further, that the share of the changed macro blocks is 10% of the picture, there are 0.1×594, that is, about 60 blocks to be coded. Of the available 963 bits, about 50 are used to address the changed macro blocks, for example, by use of a binary run-length coded bit map. The other 910 bits are left for the picture information, so there are 910/60, that is, 15 bits available per block.
At very low transfer rates, or if the amount of changes in the picture is greater than the 10% assumed previously, one generally used method is to reduce the picture rate so that the amount of transferred picture information and, in that way, the resolution can be retained. The reduction in the picture rate is known to reduce the quality of the motion the eye can see. An alternative to reducing the picture rate is a more efficient compression, in which the aim is to present the information contained in a picture block with a smaller amount of information than previously.
OBJECT OF THE INVENTION
The object of the invention is to develop a video compression method, with which it would be possible to transfer picture information coded with a very small number of bits, and still retain the quality of the picture at a reasonable level. One object of the invention is to code with 14 bits the picture information of a block with 8×8 pixels.
Additionally, the compression method of the invention and the corresponding reconstruction method aim at retaining the continuity of the picture's contours and the contrast. The method also makes it possible to enhance the resolution of the picture by interpolating during the reconstruction, and to reconstruct the picture for a greater number of pixels without the contours becoming staggered.
In order to achieve these objects, the method of the invention is characterized in what is presented in claim 1.
SUMMARY OF THE INVENTION
In a preferred embodiment of the method of the invention, the picture is coded in blocks of 8×8 pixels or, if the coding is not accurate enough, in sub-blocks of 4×4 pixels. The information contained in the block in divided into pattern information and the value information connected with it. Usually, a block contains a pattern formed by one or two contours, in which case the pattern is defined by the contours and the borders of the block. In the method of the invention, a relatively small group of pattern models are defined beforehand, and each model is given in corresponding, relatively short code of, for example, 8 bits. The pattern models are created, so that, they approximate accurately enough as many different patterns as possible. In a preferred embodiment, a pattern model that resembles most the pattern of the block is searched for, and the code defined for this patterns model is set as the pattern information code. The pattern information code is combined with a value information code corresponding with the value information connected with the pattern information.
Also, one possibility for more accurate coding is to divide the block into four smaller sub-blocks and to apply to these sub-blocks the invention's method for coding the picture information of a block by searching for the most resembling pattern models for the sub-blocks from a predefined relatively small group of sub-block pattern models, which have been defined in addition to the block pattern models.
Previously coded blocks located next to the block being coded can also be used to help the coding, and to ensure that the contours are continuous.
The method of reconstructing a block that has been compressed by using the method of the invention is characterized in what is presented in claim 12. In an advantageous embodiment of the invention the block compressed by a method according to the invention may be reconstructed to have a more pixels than the original block, without causing staggering of the contours in the block.
In the following, the invention is described in more detail, and references are made to the included figures of
Haikonen Pentti
Juhola Janne M.
Latva-Rasku Petri
Lee Young
Nokia Mobile Phones Ltd.
Perman & Green LLP
LandOfFree
Video compressing method wherein the direction and location... 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 compressing method wherein the direction and location..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Video compressing method wherein the direction and location... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2608784