JPH1141607A - Moving image processor - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a moving image processor in which a memory capacity for MPEG decoding is saved. SOLUTION: A moving image processor has an MPEG decoding part 2 that decodes data coded in compliance with the MPEG standards, a memory 10 that stores image data, and a filter part 7 that magnifies/reduces image data. In the case of reducing an output image size with respect to a decoded image size, an inverse discrete cosine transform part (IDCT part) 5 inside the MPEG decoding section 2 conducts processing and the filter section 7 reduces the image size and the result is stored in the memory 10.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

The present invention relates to a moving picture processing apparatus, and more particularly to a moving picture processing apparatus having an MPEG decoding unit having a filter function.

[0002]

2. Description of the Related Art A conventional example will be described with reference to the drawings.

FIG. 2 is a block diagram of a conventional moving image processing apparatus having an MPEG decoding section having a filter function. In FIG. 2, a code input 21 is input to a moving image processing apparatus, and the moving image processing apparatus includes an MPEG decoding unit 22 and a data bus 29.
, A memory 30, a filter unit 27, and an output unit 28, and a video output 34 from the moving image processing apparatus.
Here, the MPEG decoding unit 22 includes the VLD unit 23 and the IQ
The memory 30 is composed of a reference memory 0.31, a memory 24, an IDCT unit 25, and a motion compensation unit 26.
, Reference memories 1 and 32, and a B picture display memory 33.

[0004] The video code input from the code input 21 is subjected to variable length code decoding and zigzag scanning in the VLD section 23 and passed to the IQ section 24. The IQ unit 24 performs inverse quantization and inputs the result to the IDCT unit 25. The IDCT unit 25 performs an inverse discrete cosine transform (hereinafter abbreviated as an inverse DCT transform) on the input data. When the image subjected to the inverse DCT conversion is an I picture, the reference memory 0.31 or the reference memory 1.3 is connected via the data bus 29.
Write to 2. When the image subjected to the inverse DCT transformation is a P picture or a B picture, the picture is sent to the motion compensation unit 26 and motion compensated. If the motion-compensated image is a P-picture, it is written to the reference memory 0.31 or reference memory 1-32. If the motion-compensated image is a B-picture, it is written to the B-picture display memory 33.

[0005] The image decoded by the MPEG decoding unit 22 is:
Reference memory 0 ・ 31, reference memory 1 ・
32, written into the B picture display memory 33. The image data once written to the memory is transmitted to the filter unit 27.
And is output from the video output 34 through the output unit 28.

The amount of memory written in the reference memories 0 and 31, the reference memories 1 and 32, and the B-picture display memory 33 is 720 pixels × 480 pixels × 8 bits × 720 pixels in the case of 720 horizontal pixels and 480 vertical pixels, respectively.
1.5 = 4,147,200 bits, total 12,44
1600 bits.

[0007]

In the conventional MPEG decoding unit, even when the image is reduced and output in the image output filter unit, the image is written to the memory at the same magnification, so that the size of the image actually output is large. The problem of occupying more memory than that has arisen.

An object of the present invention is to provide a moving picture processing apparatus capable of reducing the amount of memory in MPEG decoding.

[0009]

According to the present invention, there is provided a moving picture processing apparatus comprising: an MPEG decoding unit for decoding data encoded according to the MPEG standard; a memory for storing image data; If the output image size is reduced with respect to the decoded image size
After processing by the inverse discrete cosine transform unit inside the G decoding unit,
It has means for reducing by a filter and storing it in a memory.

When the output image size is made equal to or enlarged with respect to the decoded image size, the image decoded by the MPEG decoding unit is converted into an I-picture or a P-picture.
Once stored in a reference memory inside the memory, and in the case of a B picture, once stored in a display memory inside the memory, and then read out from the memory and output after applying an equal-size or enlargement filter. Is reduced by a filter after processing by an inverse discrete cosine transform unit inside the MPEG decoding unit, and is stored in the reference memory as it is for an I picture, and the motion compensation inside the MPEG decoding unit is used for a P picture. After compensating the motion in the section, store it in the reference memory,
In the case of a B picture, there may be provided a means for storing in a display memory after motion compensation by a motion compensator, and thereafter reading and outputting from the memory.

Therefore, when the output image is reduced to less than one time, two reference memories and a B picture display memory can be reduced.

[0012]

Embodiments of the present invention will be described with reference to the drawings. FIG. 1 is a configuration diagram of a moving image processing apparatus having an MPEG decoding unit having a filter function according to an embodiment of the present invention. In FIG. 1, a code input 1 is input to a moving image processing apparatus, and the moving image processing apparatus
, A data bus 9, a memory 10, a filter unit 7,
The video output 14 is provided from the moving image processing apparatus. Here, the MPEG decoding unit 2 includes a VLD unit 3
, An IQ unit 4, an IDCT unit 5, and a motion compensation unit 6.
, Reference memories 1 and 12 and a display memory 13 for B pictures.

The video code input from code input 1 is
The variable length code is decoded and zigzag scanned by the VLD unit 3 and passed to the IQ unit 4. The IQ unit 4 performs inverse quantization and inputs the result to the IDCT unit 5. The IDCT unit 5 performs inverse DCT on the input data.

If the size of the image output from the video output 14 is at least one time and the image subjected to the inverse DCT conversion is an I picture, the reference memory 0. Write to. When the image size output from the video output 14 is 1 or more and the image subjected to the inverse DCT transformation is a P picture or a B picture, the picture is sent to the motion compensation unit 6 and motion compensated. If the motion-compensated image is a P-picture,
The data is written to the reference memories 0 and 11 or the reference memories 1 and 12. If the motion-compensated image is a B picture, the data is written to the B picture display memory 13.

The images decoded by the MPEG decoding unit 2 are stored in reference memories 0 and 11 and reference memories 1 and 1.
2. The data is written to the B picture display memory 13. The image data once written to the memory is enlarged or magnified by the filter unit 7 and output from the video output 14 through the output unit 8.

If the size of the image output from the video output 14 is less than one time, the inverse DCT transformed data is input to the filter unit 7 and reduced. When the image reduced by the filter unit 7 is an I picture, the reference memory 0. 11 or the reference memory 1.
12 is written. When the image reduced by the filter unit 7 is a P-picture or a B-picture, the image is passed to the motion compensation unit 6 and subjected to motion compensation. If the motion-compensated image is a P picture, it is written to the reference memory 0. 11 or reference memory 1, 12. If it is a B picture, it is written to the B picture display memory 13.

The image data written in the reference memories 0 and 11, the reference memories 1 and 12 and the display memory for the B picture are output from the video output 14 through the output unit 8.

If the video output size is 1 or more times, the reference memories 0 and 11 and the reference memories 1 and 1
2. In the case of 720 horizontal pixels and 480 vertical pixels, the amount of memory written in the B picture display memory is 720 pixels × 480 pixels × 8 bits × 1. 5 = 4,1
47,200 bits, for a total of 12,441,600 bits. If the video output size is less than one time, the reference memories 0 and 11, the reference memories 1 and 12, B
The amount of memory written to the picture display memory decreases according to the reduction ratio.

When reducing 720 horizontal pixels and 480 vertical pixels by 縦 each in the vertical and horizontal directions, 720 pixels ×
480 pixels x 8 bits x 1.5 x 1/2 x 1/2 = 1,
036,800 bits, for a total of 3,110,400 bits, 1/4 of the conventional total of 12,441,600 bits
It is.

By reducing the amount of memory in MPEG decoding, OSD (On Screen Display)
For example, it is possible to allocate more memory capacity to other functions.

[0021]

As described above, according to the present invention, when the output image is reduced to less than one time, the image is processed by the inverse discrete cosine transform unit inside the MPEG decoding unit, then reduced by the filter, and stored in the memory. By having the means, there is an effect that two reference memories and a B-picture display memory can be reduced.

[0022]

[Brief description of the drawings]

[0023]

FIG. 1 is a diagram showing an M having a filter function according to an embodiment of the present invention.
FIG. 2 is a configuration diagram of a moving image processing device having a PEG decoding unit.

[0024]

FIG. 2 is a conventional configuration diagram of a moving image processing apparatus having an MPEG decoding unit having a filter function.

[0025]

[Explanation of symbols]

 1, 21 Code input 2, 22 MPEG decoding unit 3, 23 VLD unit 4, 24 IQ unit 5, 25 IDCT unit 6, 26 Motion compensation unit 7, 27 Filter unit 8, 28 Output unit 9, 29 Data bus 10, 30 Memory 11, 31 Reference memory 0 12, 32 Reference memory 1 13, 33 Display memory for B picture 14, 34 Video output

Claims (2)

[Claims] 1. An image processing apparatus comprising: an MPEG decoding unit for decoding data encoded according to the MPEG standard; a memory for storing image data; and a filter for enlarging / reducing the image data. When reducing the size, the moving image processing apparatus further comprises means for performing processing in an inverse discrete cosine transform unit inside the MPEG decoding unit, reducing the size by the filter, and storing the data in the memory. 2. An image decoded by an MPEG decoding unit when the output image size is equal to or enlarged with respect to the decoded image size, and is temporarily stored in a reference memory inside the memory in the case of an I picture or a P picture. , In the case of a B picture, means for temporarily storing in a display memory inside the memory, and thereafter reading out from the memory and applying an equal-size or enlargement filter to output, and reducing the output image size with respect to the decoded image size. Is processed by an inverse discrete cosine transform unit inside the MPEG decoding unit, then reduced by the filter, and stored in the reference memory as it is for an I picture, and a motion compensation inside the MPEG decoding unit for a P picture. After performing motion compensation in the section, the data is stored in the reference memory. In stored in the display memory after motion compensation, the moving image processing device according to then the memory in claim 1, characterized in that it comprises means for reading output.