Pulse or digital communications – Bandwidth reduction or expansion – Television or motion video signal
Reexamination Certificate
1999-09-07
2003-03-18
Kelley, Chris (Department: 2613)
Pulse or digital communications
Bandwidth reduction or expansion
Television or motion video signal
Reexamination Certificate
active
06535557
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
The present invention relates to a moving picture coding method for coding a moving picture image and a moving picture coding apparatus used for this method, and more particularly relates to a method and apparatus for coding a moving picture image used preferably for transmitting the moving picture image via a network by utilizing a variable length packet switching method or a variable bit rate TDM switching method.
2. Description of Related Art
Generally, in the case that a moving picture image is transmitted via a network, the moving picture image is coded with compression by means of for example MPEG (Moving Picture Experts Group) coding method, and the thus coded digital moving picture stream data are transmitted via the network by utilizing a variable length packet switching method (IP network) or a variable bit rate TDM (Time Division Multiplex) switching method.
Here, the MPEG coding method is an established standard as a method for coding a moving picture image. In the MPEG coding method, a highly effective compression is achieved by performing a motion vector prediction in forward and backward directions of the picture in addition to a picture compression of respective pictures by means of DCT and by performing a coding by utilizing a time interval of the moving picture image.
FIG. 6
is a schematic view showing the known hierarchical structure of MPEG picture. A piece of pictures comprises a plurality of slices. The slice is a minimum unit of data row having a synchronous code, and comprises a layer of macro blocks each having 16×16 pixels (arbitrary in MPEG1 and arbitrary length not crossing both side ends of picture in MPEG2). In the embodiment shown in
FIG. 6
, a synchronous code of a slice header includes a vertical information, and the coordinates in the picture can be determined from the synchronous code and that of the macro block continued to the slice header. Moreover, various predictions due to differential coding are all cleared as zero.
Moreover, the IP network is a variable length packet switching method which is a de facto standard by a development of internet. A switchboard (router) determines a route on the basis of an address (IP address) described in a header of the packet. A maximum packet length is determined corresponding to the network to be used.
Further, the variable bit rate TDM switching method performs a switching operation by storing data in a time slot having fixed length provided in TDM frame, as shown in FIG.
7
. As being different from the existing TDM switchboard, since the number of slots assigned to respective traffics is varied in response to the bit rate necessary at that time, it is possible to handle multi rate and variable bit rate traffics. Furthermore, it is possible to solve a problem about overhead or switching rate which is a problem in ATM switchboard by making a slot size larger corresponding to the network exclusive for the picture image, and also it is easy to achieve an assurance of synchronicity and a reduction of delay flicker by utilizing the TDM method. Moreover, it is possible to perform a control for optimizing a video quality in the switchboard, if the switchboard comprehends data contents stored in the time slot (correspondence with a video frame).
As mentioned above, in the case that the moving picture image is coded by means of the MPEG coding method utilizing the existing coding device and is transmitted by utilizing the IP network, there are following problems.
(1) Problems at receiving side due to data loss in the network:
When an MPEG stream is transmitted via the network, an existing encoder performs the coding irrelevant to a data transmission unit of the network, and the coded data are divided unconditionally irrelevant to the stream structure and are transmitted via the network. Therefore, in the case that a loss of packet or time slot occurs in the network, a stream unconformity occurs at a receiving side on and after a data drop portion. Such a loss of packet and time slot may also occur when the packet and so on can not be transmitted due to a network congestion or when the packet and so on can not be received due to a transmission error. In this case, since a variable length coding having no synchronous code is used in a macro block level, it is possible to perform a decode synchronizing at the receiving side, ant thus the decode synchronizing is maintained at a deteriorated state till the slice header including a synchronous point appears. Therefore, a decode algorithm is broken down, so that a larger distortion occurs in a decoded picture image.
(2) Problems on decoder at receiving side:
In the decoder at the receiving side, even if a loss of incoming data occurs, it is thought to be possible to perform an interpolation treatment of the data loss portion by utilizing nearby pixels or a picture in a previous frame. However, if the synchronicity is deteriorated due to the data loss, it is difficult to perform for example a detection of the data loss, a detection of the data loss region and so on, and thus the interpolation treatment can hardly be performed.
SUMMARY OF THE INVENTION
An object of the present invention is to eliminate the drawbacks mentioned above and to provide a method and apparatus for coding a moving picture image in which, even if a packet and a time slot are disposed on the network, it is possible to maintain a synchronicity at a receiving side or it is possible to rapidly recover the synchronicity at the receiving side.
According to the invention, a method for coding a moving picture image, in which the moving picture image is coded into a digital moving picture stream when the moving picture image is transmitted via a network, comprises a step of; coding the moving picture image in such a manner that a data unit including a synchronous signal corresponds or conforms to a data transmission unit on the network, or comprises a step of; coding the moving picture image in such a manner that a synchronous signal is included in a position near a head of respective data transmission units on the network.
Moreover, in a preferred embodiment of the method for coding a moving picture image according to the invention, a correspondence between the data unit and the data transmission unit is a conformity between the data unit and the data transmission unit, the network utilizes a variable length packet switching method such as an IP network or a variable bit rate TDM (Time Division Multiplex) method, and the coding of the moving picture image is based on MPEG (Moving Picture Experts Group) method.
Further, according to the invention, an apparatus for coding a moving picture image, comprises; an asynchronous layer encoder for performing a coding operation of the moving picture image till an uppermost layer having no synchronicity, in a hierarchical structure of a moving picture compression method; a buffer for storing coded data outputted from the asynchronous layer encoder in order; a synchronous layer encoder for picking up the coded data stored in the buffer and performing a coding operation of the moving picture image of a layer having synchronicity and till a layer positioned upper than that layer by inserting the synchronous signal, to obtain a digital moving picture stream; and a synchronous layer generation controller for observing a data amount of the coded data stored in the buffer and generating a coding instruction to the synchronous layer encoder when the data amount reaches to the data transmission unit on the network.
Furthermore, in a preferred embodiment of the apparatus for coding a moving picture image according to the invention, the generation of the coding instruction in the synchronous layer generation controller is performed by; (1) observing a bit amount generated from an output of the asynchronous layer encoder in each time; and predicting a timing at which the bit amount exceeds a target bit amount by the next output from the asynchronous layer encoder, (2) generating a synchronous layer by deleting a
Aida Hitoshi
Aoki Terumasa
Ohnishi Takayuki
Saito Tadao
Kelley Chris
Oliff & Berridg,e PLC
Senfi Behrooz
The University of Tokyo
LandOfFree
Method and apparatus for coding moving picture image does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Method and apparatus for coding moving picture image, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method and apparatus for coding moving picture image will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3018205