JP2002152775A - Method and device for encoding - Google Patents

Abstract

PROBLEM TO BE SOLVED: To smoothly perform converting processing the high speed to pixel data having various numbers of lateral pixels without preparing a table or the like for supporting converting processing. SOLUTION: The number of lateral pixels in a picture of pixel data of 4:2:2 is divided by '48' to obtain the first quotient, pixel data fetched with the lateral width of 48 pixels into a memory inside an IC is converted from 4:2:2 to 4:2:0 with the lateral width of 48 pixels the number of the first quotient times. Next, the number of pixels being the residual resulting from the above division by '48' is divided by '16' to obtain the second quotient, pixel data fetched with the lateral width of 16 pixels into the memory inside the IC is converted from 4:2:2 to 4:2:0 with the lateral width of 16 pixels the number of the second quotient times.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an encoding method and an encoding device for performing image compression.

[0002]

2. Description of the Related Art MPEG (Moving Picture Experts Gro
up) When the main profile main level image compression is performed in 2, 4: 2: 2 pixel data which is a digital video signal in which each sampling frequency of the two types of color difference signals is １ ／ of that of the luminance signal From the above, the sampling frequency of each of the two types of color difference signals is の of that of the luminance signal, but it is necessary to perform pre-processing for pixel conversion into 4: 2: 0 pixel data that transmits the color difference signals line-sequentially. There is. Generally, if a filter is directly applied on the SDRAM to perform this pre-processing, it takes a very long time due to page-crossing.

Therefore, 4: 2: 2 pixel data is once fetched from the SDRAM, and a 4-tap vertical digital filter as shown in FIG. 4 is applied to the internal memory of the encoder IC (see FIG. 2). A general method is to write back 4: 2: 0 pixel data (the number of color components is halved) after processing and write it back to the SDRAM. In this case, if the pixel data to be converted is at the upper or lower part of the screen as shown in FIG. 5, it is necessary to copy one line at the upper or lower part due to filtering. In the example of FIG. 5, the line (1) is copied to form the line (2). In the case of the center of the screen, the overlap area is fetched one by one in the upper and lower lines.

[0004]

SUMMARY OF THE INVENTION MPEG as described above
In the pre-processing of image compression in 2, in order to reduce the number of accesses to the SDRAM, the larger the range to be processed at a time, the better. However, there is actually a limit to the capacity of the memory inside the IC, and in reality, 34 × 48.
Pixels are the limit.

In MPEG, the number of pixels to be handled is a multiple of 16 in LUMA (brightness), so that Cr (color difference signal of red system),
Even when processed together with Cb (color difference signal of blue system), 48
In a pixel unit, there is a problem that the remaining portion cannot be processed well.

Therefore, in the current encoder, a table is prepared for what kind of processing is performed for the required number of horizontal pixels, and the processing of 48 pixels is performed several times, and then the processing of 16 pixels is performed several times. It is sufficient to prepare a table for the number of types of horizontal pixels that support whether or not to perform the conversion. In this case, however, if there is pixel data of a number of horizontal pixels that is not included in the table, the conversion process will fail. To avoid this, processing may be performed in units of 16 pixels, but this has the problem that the number of accesses to the SDRAM increases and the processing speed decreases.

SUMMARY OF THE INVENTION The present invention has been made to solve the above-mentioned conventional problems, and an object of the present invention is to prepare pixel data having various numbers of horizontal pixels without preparing a table for supporting pixel conversion processing. An object of the present invention is to provide an encoding method and an encoding device capable of performing smooth and high-speed pixel conversion.

[0008]

In order to achieve the above object, a feature of the present invention is that a 4: 2: 2 pixel data is fetched from an external memory to an internal memory to form a vertical digital filter. In the MPEG2 encoding method of converting the pixel data into 4: 2: 0 using the 4: 2: 2 pixel data, the quotient and the remainder are obtained by dividing the number of horizontal pixels of the 4: 2: 2 pixel data by 48. Performing the conversion with a width of 48 pixels by the number of obtained quotients, and dividing the obtained remainder by 16 and performing the conversion with a width of 16 pixels by the number of obtained quotients. It is in.

A feature of the second aspect of the present invention is that an external memory for storing 4: 2: 2 pixel data, and fetching the 4: 2: 2 pixel data and using a digital filter in the vertical direction to store 4: 2: 2 pixel data. In an MPEG2 encoding apparatus having an internal memory for converting into 2: 0 pixel data, the number of horizontal pixels of the 4: 2: 2 pixel data in the external memory is obtained and divided by 48. First dividing means for obtaining a quotient and a remainder, variable acquiring means for fetching the 4: 2: 2 pixel data from the external memory into the internal memory with a width of 48 pixels or 16 pixels, and a width of 48 pixels 4: 2: 2 fetched into the internal memory at
The first conversion means for converting the pixel data of 4: into pixel data of 4: 2: 0 using the digital filter in the vertical direction with a width of 48 pixels, and the second for obtaining the quotient by dividing the remainder by 16 The dividing means and the 4: 2: 2 pixel data fetched into the internal memory with a width of 16 pixels by using the vertical digital filter with a width of 16 pixels.
A second conversion means for converting the pixel data into 2: 0 pixel data and a data return means for returning the converted 4: 2: 0 pixel data from the internal memory to the external memory are provided.

[0010]

Embodiments of the present invention will be described below with reference to the drawings. FIG. 1 is a block diagram showing a configuration according to an embodiment of the encoding device of the present invention. The MPEG2 encoding apparatus according to this embodiment includes an SDRAM 1 for storing pixel data, and an encoder IC 2 for performing pixel conversion processing of the pixel data.

The encoder IC2 has a function of dividing the number of pixels of pixel data, a function of exchanging pixel data with the SDRAM 1, a function of converting pixel data, and the like.
It has various functions necessary to convert 2: 2 pixel data to 4: 2: 0 pixel data.

FIG. 2 shows an example of the configuration of the internal memory 3 of the encoder IC 2 described above, and a data storage area 31 of 4: 2: 2.
And a converted data storage area 32 of 4: 2: 0.

Next, the operation of this embodiment will be described.
In this example, when the horizontal pixels to be processed at one time are 48 pixels (16 × 3) including Cr and Cb as in the current technology, what kind of processing is performed is described with reference to the flowchart of FIG. Will be explained.

First, the number of horizontal pixels (variable) on the screen of 4: 2: 2 pixel data from the SDRAM 1 is determined by an encoder IC.
2 (step 301), the number of horizontal pixels is divided by 48, and the quotient is set as loop (step 302). Next, the number of conversions (count)
Is set to 0 (step 303).

The 4: 2: 2 pixel data is fetched from the SDRAM 1 with a width of 48 pixels, and the encoder I
The data is stored in the data storage area 31 of the internal memory 3 of C2, and these data are stored in the horizontal direction by using a digital filter in the vertical direction.
The pixel data is converted from 4: 2: 2 to 4: 2: 0 with eight pixels, and the converted pixel data is stored in the data storage area 32 (step 304). Next, the number of conversions count is lo.
is determined (step 305), and if smaller, count is incremented by +1 (step 3).
06), the process returns to the step 304.

The above conversion process is repeated, and the number of conversions cou
When nt becomes loop (step 305), the process proceeds to step 307, where the remainder obtained by dividing the number of horizontal pixels of the 4: 2: 2 pixel data by 48 is divided by 16, and the obtained quotient is defined as loop. (Step 307). Next, the initial value of the number of conversions (count) is set to 0 (step 30).
8).

The 4: 2: 2 pixel data is fetched from the SDRAM 1 with a width of 16 pixels, and the encoder I
The data is stored in the data storage area 31 of the internal memory 3 of C2, and these data are stored in the horizontal direction by using a digital filter in the vertical direction.
The pixel data is converted from 4: 2: 2 to 4: 2: 0 with six pixels, and the converted pixel data is stored in the data storage area 32 (step 309). Next, the number of conversions count is lo
It is determined whether it is smaller than op (step 310), and if smaller, count is incremented by +1 (step 3).
11), the process returns to step 309. Number of conversions cou
When nt becomes loop (step 310), the process ends. The converted data stored in the data storage area 32 is sequentially returned to the SDRAM 1, and when the conversion process is completed, all the converted pixel data is stored in the SDRAM 1.

According to the present embodiment, the number of horizontal pixels on the screen is divided by 48, and the pixel conversion from 4: 2: 2 to 4: 2: 0 in the horizontal width of 48 pixels is performed by the number of times of the quotient by 48 times. By performing in a memory, further dividing the remaining pixel by 48 by 16 and performing the pixel conversion process with the width of 16 pixels by the number of quotients, all pixel data is processed every 16 pixels, Alternatively, it is possible to avoid having a table corresponding to the number of horizontal pixels of the pixel data, and it is possible to perform processing according to the number of horizontal pixels at high speed.

In the above embodiment, the present invention is applied to the processing in the horizontal direction. However, the same effect can be obtained by similarly applying the present invention to the processing in the vertical direction.

The present invention is not limited to the above-described embodiment, and may be embodied in various other forms in terms of specific configuration, function, operation, and effect without departing from the gist of the invention. .

[0021]

As described above in detail, according to the encoding method and the encoding apparatus of the present invention, pixel data having various numbers of horizontal pixels can be smoothly processed without preparing a table for supporting the conversion processing. In addition, pixel conversion processing can be performed at high speed.

[Brief description of the drawings]

FIG. 1 is a block diagram illustrating a configuration according to an embodiment of an encoding device of the present invention.

FIG. 2 is a block diagram showing a configuration example of an internal memory of the encoder IC shown in FIG.

FIG. 3 is a flowchart showing a processing procedure according to an embodiment of the encoding method of the present invention.

FIG. 4 is a diagram illustrating conversion from 4: 2: 2 to 4: 2: 0 by a conventional digital filter.

FIG. 5 is a diagram showing an example of one-line copying at the top of the screen when converting from 4: 2: 2 to 4: 2: 0 using a conventional digital filter.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 1 SDRAM 2 Encoder IC 3 Internal memory 31 4: 2: 2 data storage area 32 4: 2: 0 data storage area

Claims (2)

[Claims] 1. An encoding method for fetching 4: 2: 2 pixel data from an external memory to an internal memory and converting the fetched pixel data into 4: 2: 0 pixel data using a vertical digital filter. Dividing the number of pixels of the 2: 2 pixel data in the horizontal direction of the screen by 48 to obtain a quotient and a remainder, and performing the conversion with a width of 48 pixels by the number of obtained quotients; Dividing the remainder by 16 and performing the conversion with a width of 16 pixels by the number of obtained quotients. 2. An external memory for storing 4: 2: 2 pixel data, and fetching the 4: 2: 2 pixel data into a 4: 2: 0 pixel data using a vertical digital filter. A coding apparatus having an internal memory for conversion, wherein the number of pixels of the 4: 2: 2 pixel data in the external memory in the horizontal direction of the screen is obtained and divided by 48 to obtain a quotient and a remainder. And the external memory stores the 4: 2: 2 pixel data in 48
A variable acquisition unit for fetching the internal memory with a width of pixels or 16 pixels to the internal memory; First converting means for converting the pixel data into 4: 2: 0 pixel data, second dividing means for dividing the remainder by 16 to obtain a quotient,
Second conversion means for converting the 2: 2 pixel data into 4: 2: 0 pixel data with a width of 16 pixels using the vertical digital filter; and the converted 4: 2: 0 pixel data. Encoding means for returning data from the internal memory to the external memory.