Pulse or digital communications – Bandwidth reduction or expansion – Television or motion video signal
Reexamination Certificate
1997-09-09
2002-07-09
Lee, Richard (Department: 2613)
Pulse or digital communications
Bandwidth reduction or expansion
Television or motion video signal
Reexamination Certificate
active
06418167
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
This invention relates to a moving picture compression apparatus and a moving picture expansion apparatus suitable for a system for transmitting or storing moving picture signals, such as a digital television broadcasting equipment or a digital video disc.
2. Description of the Related Art
In a system for transmitting or storing digital moving picture signals, picture signals are encoded (compressed) by exploiting intra- or inter-frame correlation of moving picture signals for efficient utilization of a transmission channel or a storage medium. As encoding (compression) techniques for moving picture signals, there is known a compression system standardized by a research organization termed Moving Picture Experts Group (MPEG) for encoding moving pictures for storage.
In the method for compressing picture signals by exploiting the intra-frame correlation, orthogonal transform, such as discrete cosine transform (DCT), capable of concentrating coefficients for encoding, is predominantly employed.
In the method for compressing picture signals by exploiting the inter-frame correlation, so-called motion-compensated inter-frame prediction is predominantly employed. The principle of the motion-compensated inter-frame prediction is now explained by referring to FIG.
1
. It is assumed that pictures P
1
and P
2
have been generated at time pints t
1
and t
2
, respectively, with the picture P
1
having been transmitted and with the picture P
2
being about to be transmitted, as shown in FIG.
1
. The picture P
2
is split into plural blocks for each of which the amount of motion (motion vector) between it and the picture P
1
is detected. The motion-compensated inter-frame prediction resides in finding a difference picture between a prediction picture and the block of the picture P
2
and encoding the difference picture and the motion vector. The prediction picture corresponds to the picture P
1
moved in translatory movement a distance equal to the motion vector.
FIGS. 2 and 3
shows a conventional moving picture compression device which takes advantage of the above-described intra- and inter-frame correlation, and the structure of a conventional moving picture expansion device, respectively.
The conventional moving picture compression device, shown in
FIG. 2
, compresses the input digital picture signal entering a picture input terminal
101
to output the compressed signal at a bitstream output terminal
109
.
In the conventional moving picture compression device, shown in
FIG. 2
, the picture signals entering the picture input terminal
101
is routed to a motion vector detector
112
where the motion vector is calculated. The motion vector information as found by the motion vector detector
112
is sent to a motion compensation unit
110
which then motion compensates a picture stored in a frame memory
111
, based on the motion vector, for formulating a prediction picture.
The digital picture signals entering the picture input terminal
101
are also routed to a difference calculation unit
102
which then calculates the difference between the picture signals entering the picture input terminal
101
and the prediction picture formulated by the motion compensation unit
110
. The difference signal, thus found by the difference calculation unit
102
, is routed to an orthogonal transform unit
103
for orthogonal transform. The signal orthogonal transformed by the orthogonal transform unit
103
is routed to a quantizer
104
where it is quantized for compression. The quantized data is routed to a multiplexer
108
where it is multiplexed with the motion vector information and outputted at a bitstream output terminal
109
.
The data quantized by the quantizer
104
is also routed to a dequantizer
105
where it is dequantized and then inverse orthogonal transformed by an inverse orthogonal transform unit
106
. This produces the same difference picture as that restored from the output bitstream. The signal of the difference picture and the signal of the prediction picture formulated by the motion compensation unit
110
are summed together by an adder
107
to produce picture signals which are entered to the frame memory
111
for the above-mentioned motion compensation.
On the other hand, the conventional picture expansion device shown in
FIG. 3
expands a bitstream entering an input terminal
121
to output the expanded bitstream at a picture output terminal
126
.
Referring to
FIG. 3
, the bitstream entering an bitstream input terminal
121
is sent to a motion vector separator
122
where the motion vector information is separated from the bitstream. This motion vector information is sent to a motion compensation unit
127
which then motion compensates a picture in the frame memory
128
for constructing a prediction picture.
The quantized data taken out of the bitstream by the motion vector separator
122
is routed to a dequantizer
123
for dequantization and thence supplied to an inverse orthogonal transform unit
124
for inverse orthogonal transform to generate a difference picture. The signals of the difference picture and those of the prediction picture produced by the motion compensation unit
127
are summed together by ah adder
125
to produce picture signals which are stored in a frame memory
128
while being outputted at the picture output terminal
126
.
The moving picture compression device and moving picture expansion device, as described above, are occasionally connected in series to each other, as shown in FIG.
4
. The compression and expansion devices, thus interconnected in tandem, as shown in
FIG. 4
, are equivalent to a device for repeatedly executing compression and expansion.
Specifically, the picture signals supplied to a picture input terminal
200
in
FIG. 4
are compressed by a moving picture compression device
201
and outputted at a bitstream output terminal
202
. This bitstream is supplied by for example broadcasting, communication or recording medium to a bitstream input terminal
220
and expanded by a moving picture expansion device
221
so as to be outputted at a picture output terminal
222
. The picture signals, outputted at the picture output terminal
222
, are entered via for example an edition unit, not shown, to a picture input terminal
240
. The moving picture signals supplied to the picture input terminal
240
are compressed by a moving picture compression device
241
so as to be outputted at a bitstream output terminal
242
. This bitstream is supplied by for example broadcasting, communication or recording medium to a bitstream input terminal
260
and expanded by a moving picture expansion device
261
so as to be outputted at a picture output terminal
262
. The picture signals outputted at the picture output terminal
222
are sent to for example the edition unit for edition.
In the arrangement shown in
FIG. 4
, the moving picture compression device
201
and the moving picture expansion device
221
execute first compression/expansion, while the moving picture compression device
241
and the moving picture expansion device
261
execute second compression/expansion. The same holds for the third and following compression/expansion operations.
If the picture is repeatedly compressed/expanded by the above-described compression/expansion system, the picture quality is deteriorated each time the operations are repeated.
Thus it is said to be advisable to match the picture coding type at the time of compression for suppressing picture quality deterioration brought about by repeated compression/expansion. That is, picture quality deterioration is thought to be suppressed by using the same encoded picture type, that is the intra-coded picture or I-picture devoid of motion compensation, a forward predictive encoded picture or P-picture obtained on motion compensation from a temporally previous frame or a bidirectional prediction encoded picture or B-picture obtained on motion compensation from a temporally previous frame and a temporally succeeding frame, as that us
Hosaka Kazuhisa
Kato Motoki
Miyahara Nobuyoshi
Yasuda Kazunori
Frommer William S.
Frommer & Lawrence & Haug LLP
Lee Richard
Shallenburger Joe H.
Sony Corporation
LandOfFree
Moving picture compression/expansion apparatus 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 compression/expansion apparatus, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Moving picture compression/expansion apparatus will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2862361