Pulse or digital communications – Bandwidth reduction or expansion – Television or motion video signal
Reexamination Certificate
1998-12-14
2002-01-15
Le, Vu (Department: 2613)
Pulse or digital communications
Bandwidth reduction or expansion
Television or motion video signal
C348S699000, C348S700000
Reexamination Certificate
active
06339617
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
The present invention relates to a moving picture compressing apparatus and a moving picture compressing method and in particular, to a moving picture compressing apparatus and a moving picture compressing method for compression-encoding a picture signal on real time basis corresponding to a compression-encoding system mainly using a discrete cosine transform (DCT) process and a quantizing process.
2. Description of the Prior Art
When picture information is digitized and transmitted to a communication medium such as a telephone line, it is transmitted after it is compressed and encoded since the data amount of the picture information is vast. When a picture is compressed, a DCT encoding system has been widely used. In the DCT encoding system, picture data is compressed using the characteristic that magnitudes of spatial frequencies thereof tend to concentrate on lower spatial frequencies. This system has been adopted in international standards such as MPEG (Moving Picture Experts Group) video and ITU-T Recommendation H.263.
FIGS. 9A
,
9
B,
9
C,
9
D,
9
E, and
9
F are schematic diagrams showing a hierarchy of code formats of MPEG video.
FIG. 1A
shows a video sequence as the top hierarchical level. The video sequence is composed of a plurality of GOPs (Group of Pictures). Each GOP is composed of a plurality of pictures as shown in FIG.
1
B. Each picture represents one frame. Pictures are categorized into three types of pictures that are I picture, P picture, and B picture. The I picture is composed of only intra-frame codes. The P picture is composed of inter-frame codes in forward direction as well as intra-frame codes. The B picture is composed of inter-frame codes in both of forward and backward directions as well as intra-frame code.
Each picture is composed of a plurality of slices arbitrarily divided into areas as shown in FIG.
1
C. Each slice is composed of a plurality of macro-blocks arranged rightwardly or downwardly. Macro-blocks are categorized as intra-blocks and inter-blocks. The intra-blocks are composed of intra-frame codes. The inter-blocks are composed of inter-frame codes in forward direction or two directions. The I picture is composed of only intra-blocks. On the other hand, the P picture and B picture are composed of inter-blocks as well as intra-blocks.
Each macro-block is composed of a total of six blocks that are luminance components (Y
1
, Y
2
, Y
3
, and Y
4
) and two color difference components (Cb and Cr) as shown in FIG.
1
E. Each of the six blocks is composed of 8×8 pixels. A block of 8×8 pixels shown in
FIG. 1F
is the minimum encoding unit.
Next, with reference to
FIG. 2
, a conventional picture compressing process will be explained.
FIG. 10
is a block diagram showing an example of the structure of a conventional picture compressing apparatus. In the apparatus shown in
FIG. 10
, central processing unit (CPU)
2
executes a program in apparatus controlling means
1
so as to control the whole of the apparatus. A user inputs a desired command to CPU
2
through keyboard
3
. Picture data is compression-encoded by picture compressing means
4
. The resultant compressed code is transmitted to a communication line.
Picture data is supplied to color converting means
5
in picture compressing means
4
. Color converting means
5
converts the picture data into three types of data that are Y, Cr, and Cb (hereinafter referred to as YCrCb data as a whole). A motion estimating means
6
searches a macro-block in the preceding/following frame so that the difference between the macro-block and a macro-block in the current frame becomes the minimum and calculates a motion vector corresponding to the motion between he two macro-blocks. When the difference is small, predicting means
7
calculates the difference between the frames so as to perform the inter-frame compressing process.
Output data of motion estimating means
6
is supplied to DCT means
8
along with the output of predicting means
7
. When the intra-frame compressing process is performed, DCT means
8
performs a DCT process for the YCrCb data. When the inter-frame compressing process is performed, DCT means
8
performs the DCT process for the difference data. Thereafter, quantizing means
9
quantizes the resultant data. Next, variable length code encoding means
10
encodes the resultant data into variable length code.
The quantized data that is outputted from quantizing means
9
is supplied to inversely quantizing means
14
. Inversely quantizing means
14
inversely quantizes the quantized data. Inverse DCT means
13
performs an inverse DCT process for the resultant data. When the intra-frame compressing process is performed, the resultant data is stored in reference frame portion
11
. When the inter-frame compressing process is performed, the difference in the inverse DCT process is added to the macro-block compensated in motion with the motion vector stored in reference frame portion
11
. The resultant data is stored in reference frame portion
11
.
The picture code that has been compression-encoded on the transmitting side is transmitted to the communication line. The picture to be transmitted need to be compressed in a compression ratio corresponding to the bit rate of the communication line. In particular, when picture data and audio data are transmitted at a low bit rate of 64 kbps used in a TV telephone system, they should be compressed at a high compression rate. Thus, the ratio of inter-frame compressed blocks to intra-frame compressed blocks is increased because code amount of an inter-frame compressed blocks is smaller than that of an intra-frame compressed block.
For example, in the first frame, all macro-blocks are compressed in the intra-frame compressing process. In the second or later frames, all the macro-blocks are compressed in the inter-frame compressing process unless the difference between different frames is large. Thus, the ratio of macro-blocks that are compressed in the intra-frame compressing process is decreased in second or later frames. However, when a scene change takes place, a code amount abruptly increases since the difference between different frames becomes large. Thus, it takes a long time to reproduce frames following a scene change on the reproducing side. Consequently, the picture looks like a still picture.
In order to solve such a problem, various picture compressing apparatuses that effectively compress pictures preceded by a scene change have been proposed in, for example, JPA-8-56361, JPA-2-174387, JPA-7-38895, and JPA-3-13792. In the picture compressing apparatus disclosed in JPA-8-56361, several consecutive frames preceded by a scene change including frames which are originally compressed in the intra-frame compressing process are forcedly compressed in the inter-frame compressing process, so that a code amount is decreased. In the picture compressing apparatus disclosed in JPA-2-174387, the intra-frame encoding process or the inter-frame prediction encoding process is used so that the prediction error becomes minimum and blocks are encoded using codebooks of which each is dedicated for intra-frame encoding process or the inter-frame prediction encoding. In the picture compressing apparatus disclosed in JPA-7-38895, after a scene change, higher-spatial-frequency components of blocks that are compressed in the intra-frame compressing process are deleted. In the picture compressing apparatus disclosed in JPA-3-13792, after a scene change due to a switching operation of a camera, picture data of several frames are sub-sample encoded so as to decrease a code amount generated.
Among the foregoing prior art references, in the picture compressing apparatus disclosed in JPA-8-56361, the ratio of macro-blocks that are compressed in the inter-frame encoding process is increased so as to transmit code at a low bit rate. But, the effect of decreasing a code amount is small. In addition, when data is compressed at a high compression rate in order to reduce a code amount, the
Le Vu
Ostrolenk Faber Gerb & Soffen
LandOfFree
Moving picture compressing apparatus and moving picture... 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 compressing apparatus and moving picture..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Moving picture compressing apparatus and moving picture... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2835157