Pulse or digital communications – Bandwidth reduction or expansion – Television or motion video signal
Reexamination Certificate
1997-11-25
2001-07-10
Kelley, Chris (Department: 2613)
Pulse or digital communications
Bandwidth reduction or expansion
Television or motion video signal
C382S248000
Reexamination Certificate
active
06259739
ABSTRACT:
FIELD OF THE INVENTION
The present invention relates to a moving picture variable bit rate coding apparatus, a moving picture variable bit rate coding method, and a recording medium for variable bit rate coding program, especially to a moving picture variable bit rate coding apparatus, a moving picture variable bit rate coding method, and a recording medium for moving picture variable bit rate coding program, all capable of coding at a variable bit rate in a real time processing in response to capturing of picture data.
BACKGROUND OF THE INVENTION
Recently, two MPEG methods (Moving Picture Experts Group) are generally used as high-efficient coding methods for moving pictures. In coding moving pictures using either of the two MPEG methods, initially, a blocking processing is performed in which digital picture data corresponding to one picture (one frame) is divided into 16×16 macroblocks, and each macroblock is subdivided into 8×8 blocks. Each block is subjected to discrete cosine transform to obtain a discrete cosine transform coefficient (DCT). Next, a quantization processing is performed in which the obtained transform coefficient is divided by a quantization scale and a 8 by 8 quantization matrix corresponding to each frequency component, to generate a quantized transform coefficient. Then, the quantized transform coefficient is subjected to variable length coding to obtain a bit stream which is coded data as the result of the coding.
In such a series of processings, a value of a quantization scale used for the quantization processing highly affects a compression ratio in the coding processing: the larger the value of the quantization scale, the higher the compression ratio, resulting in reduced quantity of coded data; the smaller the value, the lower the compression ratio, resulting in an increased quantity of coded data. The value of the quantization scale can be set and changed in unit of macroblock of the data as a processing object, thereby controlling the generated code quantity.
FIG. 11
is a diagram for explaining relations among a bit rate as a data transmission rate, a quantization scale, a generated code quantity, an allocated code quantity per frame, and a play-back picture quality. Since the quantization processing is fundamentally a division processing as described above, a generated code quantity decreases as a quantization scale increases, whereby a bit rate as a transmission rate is reduced. In this case, since an allocated code quantity per frame increases, a superior play-back picture quality of coded data is available. In contrast with the case, when the quantization scale is smaller, the quantity of allocated code is reduced because a larger quantity of code and a higher bit rate area available, resulting in degraded picture quality of coded data.
Here provided as a control method for a generated code quantity which is used in MPEG2 TEST MODEL3 is a fixed bit rate method in which the generated code quantity is controlled in GOP (Group of Picture) units so that the quantity is held constant.
As described above, when the bit rate is high, a better picture quality is available despite of low compression rate, which is desirable to dynamic moving pictures, complicated moving pictures, i.e., moving pictures having a large amount of information. On the other hand, in static moving pictures, simple moving pictures, i.e., moving pictures having a small amount of information, the degradation of picture quality caused by high compression rate is not recognized explicitly, and thus it is possible to obtain coded data with a high compression rate by decreasing a bit rate and to reduce a processing burden for a device for use.
In fixed bit rate method, since the bit rate is controlled to a prescribed value (a set rate), it is possible to easily perform a real time processing in response to capturing of moving pictures by setting a bit rate corresponding to a property of a moving picture and a processing performance of a device.
However, in the fixed bit rate coding, an uniform quantity of code is allocated not only to coding objects having a small amount of information but to those having a large amount of information, in spite of a fact that motion scale and complexity of moving pictures are variable.
Therefore, if a set rate is smaller than expected, visual degradation of picture quality occurs in pictures having a large amount of information. With a high set rate, although the degradation of picture quality is prevented, extra quantity of code is allocated to a picture having a small amount of information, resulting in reduced coding efficiency. This leads to a problem that in the case of recording bit streams into a recording medium, a reproduction time for the recorded moving picture data, that is, a recording time becomes shorter. Further, if this method is applied to a picture whose picture quality is not affected even if the picture is further compressed, unnecessary data is processed and recorded due to an excessively low compression rate for this picture, whereby the device resource of a coding apparatus and recording medium cannot be used efficiently in this coding.
To solve the above-described problems, Japanese Patent Application Number Hei. 6-141298 provides a coding apparatus using a variable bit rate method which intends to improve picture quality without reducing coding efficiency by controlling a generated code quantity by controlling a set rate in unit times such that a total quantity of code in a moving picture has a prescribed value. In a variable bit rate coding apparatus according to such prior art technique, an input picture is subjected to two coding processing in total. Initially, as a first coding processing to an input picture, provisional coding is performed using a fixed quantization scale. For a bit stream generated by the provisional coding, a generated code quantity is computed in unit times and the result is stored as a provisional transmission rate. Then, based on the provisional transmission rate, desired transmission rate is set such that a total code quantity in an input picture has a given value. Next, as a second coding processing to the input picture, real coding is performed controlling a generated code quantity corresponding to a desired transmission rate per unit time.
This processing enables to use a transmission rate corresponding to a property of a moving picture as a coding object, thereby optimizing both picture quality and coding efficiency which the above-described fixed bit rate coding cannot.
As described above, in the coding processing or moving picture according to the prior art technique, using a fixed bit rate method enables to perform coding in a real time processing in response to the input of a moving picture and record the result of the coding. However, the coding processing with a fixed bit rate cannot correspond to the property of the moving picture, and might degrade picture quality and coding efficiency by changing a property of moving picture.
This variable bit rate coding is provided to solve these problems, and can improve both picture quality and coding efficiency by performing a provisional coding for setting an appropriate bit rate and a real coding performed on the basis of the appropriate bit rate in combination.
Nevertheless, the conventional variable bit rate coding apparatus needs two coding operations to obtain bit streams, and thus at least a time twice as long as a total time length of a moving picture is required. Further, since every moving picture is subjected to provisional coding before real coding, a storage unit, such as a memory, has to have a large capacity, and a high-performance control unit is required. In addition, since two coding processings are performed, it is impossible to perform real time processing in response to an input of a moving picture.
Performing real time processing and recording in response to capturing of moving pictures or the like with ease with general device performance is necessary for general users at home to record multimedia data includi
Kelley Chris
Matsushita Electric - Industrial Co., Ltd.
Parkhurst & Wendel L.L.P.
Vo Tung
LandOfFree
Moving picture variable bit rate coding apparatus, moving... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Moving picture variable bit rate coding apparatus, moving..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Moving picture variable bit rate coding apparatus, moving... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2449465