Image decompression system capable of reducing memory capacity

Details Image decompression system capable of reducing memory capacity Image decompression system capable of reducing memory capacity

1997-12-23

2001-05-29

Tran, Phuoc (Department: 2621)

Image analysis

Image compression or coding

Including details of decompression

C382S236000

Reexamination Certificate

active

06240210

ABSTRACT:

BACKGROUND OF THE INVENTION
The present invention relates generally to an image decompression system for decoding and decompressing compression-coded image data. More specifically, the invention relates to an image decompression system for decoding and decompressing image data using the same system as a compression system.
The storage and coding for color moving pictures have been standardized by MPEGs (Moving Picture Expert Groups). The moving picture compression technique requires code data for buffering, reference images for the motion prediction, and images for display, as data to be stored in a memory. Among these data, the reference images and the display images have large data capacities. Therefore, it is effective to reduce these data capacities in order to reduce the costs of an image decompression system.
In general, ISSCC95 (International Solid State Circuits Conference 95)/SESSION 17/VIDEO SIGNAL PROCESSING/PAPER FA17.1: An MPEG-1 Audio/Video Decoder with Run-Length Compressed Antialiased Video Overlays, discloses that “1. Decoded B frames are compressed before being written to DRAM (Dynamic Random Access Memory) to save about 200 kb of DRAM space” and that “2. Decoded B frames are compressed with a lossy DPCM compression technique to save DRAM space”.
That is, in MPEG-1, only B frames (Bidirectional Predicted Frames) are compressed using the lossy DPCM (Differential Pulse Code Modulation) technique to save about 200 kb of memory capacity.
The term “MPEG” means an organization for advancing the standardization of a coding system for storing color moving pictures, and also means the name of a coding system, for which the standardization work has been done by the MPEG. MPEG coding systems are classified into three types of MPEG-1, MPEG-2 and MPEG-4 to advance the standardization thereof (MPEG-3 for tens Mbit/second of high definition images was absorbed into the MPEG-2). The MPEG-2 is a high efficiency coding technique for internationally standardized multimedia data, which is intended to be applied to current broadcasts, high definition televisions (HDTVs) and AV equipments, and has an image transfer rate of 4 Mbit/second to 60 Mbit/second (4 Mbit/second to 10 Mbit/second before absorbing the MPEG-3). Therefore, the MPEG-2 is a motive power for the spread of digital multimedia for home and for software making. Similar to the MPEG-1, the images managed by the MPEG-2 include three types of pictures, i.e., B pictures (Bidirectional predictive frame pictures), P pictures (Predictive frame pictures) and I pictures (Intra-frame pictures).
FIG. 1
shows the frame arrangements when the respective frames of the B, P and I pictures are compressed to be reproduced. In the upper arrangement of frames for a current image, the upper alphabet denotes the kind of the image, and the lower numeral denotes the order of the arrangement of frames. In the middle arrangement of frames for a compressed image, an I picture capable of being produced only by its frame is preferentially arranged, and the arrangement of frames is recombined in the order of B, B and P pictures after the I picture. In the lower arrangement of frames for a reproduced image, an image having the same arrangement of frames as that of the current image is reproduced using a prediction according to the kind of the image of the frame.
For example, when an I2 frame is reproduced, it can be reproduced by itself, and when a P5 frame is reproduced, its picture is reproduced using the preceding I2 frame picture. When a B6 frame is reproduced, it is reproduced using the P5 frame picture and a P8 frame picture in a compressed image since it uses the last and next P picture frames in the current image.
Referring to the block diagram of
FIG. 2
, the conventional image decompression system, which operates by such a principle, will be described. In
FIG. 2
, an image decompression system
100
comprises: a decoding part
101
for decompressing an inputted compressed image to output a decoded image; a reference image storing part
102
for storing, as a reference image, the decompressed decoded image outputted from the decoding part
101
; an adder
103
for adding the decoded image outputted from the decoding part
101
, to the reference image stored in the reference image storing part
102
to output a synthesized decoded image; and a switch
104
for selecting one of the outputs of the decoding part
101
and the adder
103
.
The conventional image decompression system
100
shown in
FIG. 2
decompresses a compressed image using the respective frame pictures of the above described I, P and B pictures to output a decompressed image. Specifically, when the I frame of the compressed image is decompressed, the decompressed I picture is directly displayed and outputted, and the reference image storing part
102
stores the I picture frame using a region for a single picture. When the P picture frame is inputted as a compressed image, the reference image of the I picture frame is outputted, and the P picture is decompressed using the I picture by means of the adder
103
.
When the inputted compressed image to be decoded is a B picture, the decompression is carried out using two pictures, i.e., the last and next pictures of I pictures or P pictures in the current image. Therefore, for example, in the case of the B6, the pictures of two frames, i.e., P5 and P8 frames, are stored in the reference image storing part
102
. Thus, in the conventional image decompression system, the reference image storing part
102
must have a memory capacity capable of storing at least two frames of the decompressed and expanded decoded image.
However, in the conventional image decompression system described above, first, since the image has two kinds of pictures respectively having high and low compression rates, there is a problem in that a certain memory capacity can not be assured.
Second, since a method called “the lossy DPCM” is used, if the same of a certain capacity is attempted, the dispersion in picture quality occurs by the degree of deterioration of the picture quality due to some properties of the image, so that there is a problem in that the whole picture quality of the reproduced image is deteriorated.
Third, the saved memory capacity is 200 kb at the most, so that there is a problem in that the reduction of the memory capacity is insufficient. Since commercially available DRAMs have discrete values, e.g., 1 Mbit, 4 Mbit and 16 Mbit, only a part of the memory region of a single DRAM is saved, so that it does not contribute to the reduction of the number of DRAMs at all.
SUMMARY OF THE INVENTION
It is therefore an object of the present invention to eliminate the aforementioned problems and to provide an image decompression system, which can increase a memory capacity capable of being saved by the compression of data, and which can prevent the deterioration of a picture quality, to save the memory capacity and improve the picture quality of a reproduced image.
In order to accomplish the aforementioned and other objects, according to the present invention, there is provided an image decompression system, which can improve the memory capacity to be saved and prevent the deterioration of the picture quality by compressing and decompressing decoded image data using the same system as a compressing system used for the compression of image data.
Specifically, according to an aspect of the present invention, there is provided an image decompression system comprising: decoding means including at least one decoding part for decoding an inputted compressed image by the same system as a coding system of the compressed image; coding means for coding a decoded image, which has been decoded by the decoding means, to reproduce a compressed image; and output means for selectively outputting one of the decoded image outputted from the decoding means and a decoded image derived by decoding a compressed reference image compressed by the coding means.
The image decompression system may further comprise compressed reference image storing means for storing the

Affiliated with

Also associated with

Rating

Image decompression system capable of reducing memory capacity does not yet have a rating. At this time, there are no reviews or comments for this patent.

If you have personal experience with Image decompression system capable of reducing memory capacity, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Image decompression system capable of reducing memory capacity will most certainly appreciate the feedback.

Rate now

Comments { 0 }

Profile ID: LFUS-PAI-O-2556210