Pulse or digital communications – Bandwidth reduction or expansion – Television or motion video signal
Reexamination Certificate
2000-05-08
2004-04-20
Rao, Andy (Department: 2713)
Pulse or digital communications
Bandwidth reduction or expansion
Television or motion video signal
C375S240230
Reexamination Certificate
active
06724821
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
The present invention relates to a picture coding apparatus, and more particularly to a picture coding apparatus encoding a digital picture data.
2. Description of the Related Art
As the technique for encoding picture data in the high efficiency, a motion compensation interframe prediction coding system is conventionally standardized and widely used (for example, MPEG and so on). The system uses the property that the correlation between video signals of continuing frames is generally large. Only the differential signal between a current frame (hereinafter, to be referred to as a predicted frame) and the past or future frame (hereinafter, to be referred to as a reference frame) is coded to reduce temporal redundancy.
In such an motion compensation interframe prediction coding system, the previously coded reference frame is moved based on a motion vector and the moved reference frame is used for an prediction signal. Subsequently, the difference value (hereinafter, to be referred to as a motion vector) between the prediction signal and a signal for the current frame is encoded. Thus, the increase of the code amount caused by the change of picture data is restrained.
Here, a general image processing will be described using MPEG as an example.
FIGS. 7A
to
7
F are diagrams showing the image processing in the MPEG. As shown in
FIG. 7A
, the whole video program is referred as an image sequence and is composed of a plurality of GOPs (Group Of Pictures). The image sequence starts with a sequence header (not shown) and ends with a sequence end (not shown). Data of the whole image sequence such as a data indicative of the size of the picture, the number of frames to be coded for one second, and a communication rate are stored in the sequence header.
The GOP is composed of a GOP header (not shown) and a plurality of pictures following the GOP header. Each picture is either of an I picture, a P picture, and a B picture. The I picture is the picture which is coded using only the picture signal in the frame. The P picture is the picture which is predicted from the coded picture (I or P picture) immediately before the P picture and is used for the motion prediction only in the front direction in the image sequence. The B picture is the picture which is predicted from two neighbor coded pictures (I and/or P pictures) and is used for the motion prediction in the front and back directions. It should be noted that the B picture is not used for the prediction of another picture.
Also, each picture corresponds to each screen of a picture signal and is composed of a picture header (not shown) and a plurality of slices following the picture header. Each slice is composed of a plurality of macro blocks. The macro block is composed of pixel blocks of 16 pixels×16 pixels.
Next, a conventional motion compensation interframe prediction coding apparatus will be described.
FIG. 1
is a block diagram showing a conventional motion compensation interframe prediction coding apparatus. As shown in
FIG. 1
, the conventional apparatus is composed of an input buffer
102
with an input terminal
101
, a subtracting circuit
103
, a motion vector detecting unit
104
, a motion compensating unit
105
, a DCT unit
106
, a quantization unit
107
, a variable length coding unit
108
, an inverse quantization unit
109
, an inverse DCT unit
110
, an adding circuit
111
, a frame memory
112
, a vector coding unit
113
, and a multiplexing unit
114
.
An input picture data is supplied to and stored in the input buffer
102
via the input terminal
101
. The input buffer
102
supplies the stored picture data to the subtracting circuit
103
and the motion vector detecting unit
104
in units of predetermined pixel blocks (i.e., in units of the above-mentioned macro blocks). The subtracting circuit
103
is supplied with the picture data outputted from the input buffer
102
and a block data (a reference macro block data) of a reference frame which is subjected to the motion compensation by the motion compensating unit
105
. The subtracting circuit
103
determines an interframe differential signal using them and supplies to the DCT unit
106
.
After the DCT unit
106
carries out a DCT (discrete cosine transformation) process to the supplied interframe differential signal to separate the frequency components in the horizontal and vertical directions, and to output the separated components to the quantization unit
107
. The quantization unit
107
quantizes the output of the DCT unit
106
to reduce a bit rate and supplies the quantized data to the variable length coding unit
108
and the inverse quantization unit
109
. The inverse quantization unit
109
inversely quantizes the output of the quantization unit
107
and then supplies to the inverse DCT unit
110
.
The inverse DCT circuit
110
carries out an inverse DCT process to the output of the inverse quantization unit
109
and then supplies to the adding circuit
111
. That is, a decoding process is carried out by the inverse quantization unit
109
and the inverse DCT unit
110
, and a data similar to the interframe differential signal before the coding is obtained. The adding circuit
111
is supplied with the block data of the reference frame which is subjected to the motion compensation by the motion compensating unit
105
and the output of the inverse DCT unit
110
. The adding circuit
111
adds them to reproduce the predicted block data of a current frame. The predicted block data is supplied to the frame memory
112
.
The frame memory
112
stores the supplied block data and then supplies to the motion compensating unit
105
and the motion vector detecting unit
104
as the data of a new reference frame. The motion vector detecting unit
104
is supplied with the block data of the current frame from the input buffer
102
, and the block data of the reference frame from the frame memory
112
. The motion vector detecting unit
104
detects the motion vector between the reference frame and the current frame using the block data supplied thereto and supplies the detected motion vector to the motion compensating unit
105
.
The block data of the reference frame is supplied from the frame memory
112
to the motion compensating circuit
105
. Therefore, the motion compensating circuit
105
carries out motion compensation to the block data of the reference frame based on the motion vector from the motion vector detecting circuit
104
. The motion compensating circuit
105
produces and supplies the block data of the reference frame subjected to the motion compensation (the reference macro block data) to the subtracting circuit
103
.
On the other hand, the variable length coding unit
108
encodes the output of the quantization unit
107
into a variable length code and supplies the variable length code to the multiplexing unit
114
. The vector coding unit
113
encodes the motion vector detected by the motion vector detecting unit
104
into a variable length code to supply to the multiplexing unit
114
. The multiplexing unit
114
multiplexes the motion vector value encoded in the variable length code by the vector coding unit
113
as a header data and the picture data encoded in the variable length code by the variable length coding unit
108
and output the multiplexed code.
Next, the operation principle of such an motion compensation interframe prediction coding apparatus will be described.
FIGS. 2A
to
2
C are diagrams showing a reference frame and a predicted frames when an object moves to a direction at a constant speed. Generally, the movement of the object in the picture is felt as if it is very complicated as movement of a person in a TV picture. However, when the change of the picture is considered in units of pixels, it is not always correct. Because the motion compensation is carried out in units of predetermined pixel blocks, most of the pixels changes almost at the constant speed in compression of a picture, as shown in
FIGS. 2B and 2C
. Therefore, it could be considere
Dickstein , Shapiro, Morin & Oshinsky, LLP
NEC Corporation
Rao Andy
LandOfFree
Picture coding apparatus with coding efficiency improved does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Picture coding apparatus with coding efficiency improved, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Picture coding apparatus with coding efficiency improved will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3211476