CN101026760A - Output Buffering Method for Pixel Data of Joint Photographic Experts Group Imagery - Google Patents

Abstract

An output buffer method for pixel data of a joint image expert group image. First, a minimum coding unit of the joint photographic experts group image is received, wherein the minimum coding unit comprises a plurality of pixel data. And carrying out proportion adjustment on the minimum coding unit to obtain a proportion-adjusted minimum coding unit. And determining a storage rule of the scaled minimum coding unit in a memory module according to the format of the joint photographic experts group image. Selecting a memory included in the memory module according to the determined storage rule, and writing the pixel data included in the minimum coding unit after the proportion adjustment into the selected memory. Determining the output mode of the minimum coding unit after the ratio adjustment according to the format of the joint image expert group image. Finally, according to the determined output mode, the pixel data contained in the minimum coding unit after the proportion adjustment is output to a frame buffer from the memory module.

Description

Output Buffering Method for Pixel Data of Joint Photographic Experts Group Imagery

technical field

The present invention relates to image processing, in particular to a processing module and method for Joint Photographic Experts Group (JPEG) images.

Background technique

The Joint Photographic Experts Group (JPEG) image standard is formulated by the International Standard Organization (ISO) and the Consultative Committee of International Telegraph & Telephone (CCITT), and is a commonly used image compression format. JPEG images are currently divided into several formats, including 420, 422V, 411, 422H, and 444. Since JPEG is a compressed image storage format, when a JPEG image is to be displayed on the screen, the JPEG image must first be decoded to restore the original image before being displayed on the screen.

Regardless of the JPEG format, JPEG images are decoded using the Minimum Coded Unit (MCU) as the unit. A minimum coding unit includes a plurality of pixel blocks. According to the fact that the pixel block contains 8 columns x 8 rows (that is, 64) pixel data, the pixel data can be further divided into brightness (intensity) component and color difference (color) component, wherein the brightness component represents the brightness of the pixel data, and the color difference component Represents the color of pixel data. Therefore, the pixel blocks can be divided into pixel blocks storing luminance components and pixel blocks storing color difference components.

FIG. 1A shows the minimum coding units of JPEG images in 420 and 422V formats. The pixel data of the JPEG image 100 whose format is 420 can be divided into luminance component 102, blue color difference component 104 and red color difference component 106, and a minimum coding unit of JPEG image 100 includes six pixel blocks 112-117, four of which are The pixel blocks Y0-Y3 store the pixel data of the luminance component, and the two pixel blocks Cb, Cr store the pixel data of the color difference component. Similarly, the pixel data of the JPEG image 120 whose format is 422V can be divided into a luminance component 122, a blue color difference component 124 and a red color difference component 126, and a minimum coding unit of the JPEG image 120 includes four pixel blocks 132-135, The two pixel blocks Y0 and Y2 store the pixel data of the luminance component, and the two pixel blocks Cb and Cr store the pixel data of the color difference component.

FIG. 1B shows minimum coding units of JPEG images with formats 411, 422H and 444. The pixel data of the JPEG image 140 whose format is 411 can be divided into luminance component 142, blue color difference component 144 and red color difference component 146, and a minimum coding unit of JPEG image 140 includes six pixel blocks 152-157, four of which The pixel blocks Y0 and Y1 store the pixel data of the luminance component, and the two pixel blocks Cb and Cr store the pixel data of the color difference component. Similarly, the pixel data of the JPEG image 160 whose format is 422H can be divided into a brightness component 162, a blue color difference component 164 and a red color difference component 166, and a minimum coding unit of the JPEG image 160 includes four pixel blocks 172-175, The two pixel blocks Y0 and Y1 store the pixel data of the luminance component, and the two pixel blocks Cb and Cr store the pixel data of the color difference component. In addition, the pixel data of the JPEG image 180 whose format is 444 can be divided into a brightness component 182, a blue color difference component 184 and a red color difference component 186, and a minimum coding unit of the JPEG image 180 includes three pixel blocks 192-194, wherein The pixel block Y0 stores the pixel data of the luminance component, and the two pixel blocks Cb and Cr store the pixel data of the color difference component.

When the decoding of the JPEG image is completed, since the size of the decoded JPEG image does not necessarily match the screen size, the JPEG image needs to be scaled in advance to reduce the resolution or the amount of pixel data of the JPEG image to fit Extent for displaying on screen. Next, the adjusted JPEG image data is first stored in a buffer module to wait for the screen to issue a request for obtaining pixel data, and then output the pixel data of the image to the frame buffer of the screen.

The output modes of the image data output from the buffer module to the frame buffer of the screen are 1 byte single output mode, 8 bytes burst write out (8 bytes burst write out) mode and 16 bytes burst write out (16 bytes burst write out) mode. write out) mode. In the 1-byte single output mode, the buffer module only outputs 1-byte pixel data to the frame buffer each time. In the 8-byte or 16-byte burst output mode, the buffer module can output 8-byte and 16-byte pixel data to the frame buffer respectively. When the amount of pixel data contained in a JPEG image is large, an 8-byte or 16-byte burst output mode must be used to meet the data requirements of the screen display image.

For example, a motion JPEG image (motion JPEG) with a resolution of 800×600 needs to replace 30 images per second, and the amount of data to be transmitted per second is 21.6M (800×600×30×1.5=21.6M)bytes, Among them, the parameter 1.5 is because each unit of luminance component needs to cooperate with 0.5 unit of color difference component. If the system operates with a clock signal with a frequency of 125MHz, the data of each byte needs to be 5.787 (125MHz/21.6M=5.787) The completion of the transfer within the clock cycle is a great load on the system. Conversely, in the 8-byte or 16-byte burst output mode, since the amount of data transmitted at one time is 8 times and 16 times respectively, each transmission of data can wait for 46.296 and 92.252 clock cycles respectively. Therefore, the buffer should adopt 8-byte or 16-byte burst output mode to transmit pixel data.

In addition, in order to fully store the pixel data displayed on the screen, the general buffer requires a storage capacity of tens of thousands to megabytes. The larger the storage capacity of the buffer, the higher the production cost required. If a buffer with a smaller storage capacity can be used, and the buffer can still perform the same function to cope with the required data volume of the transmitted data, the production cost can be reduced. Therefore, if a buffer with high-speed output and limited storage capacity can be provided, the production cost can be reduced and the performance of the buffer can be improved at the same time.

Contents of the invention

In view of this, the present invention provides an output buffering method for pixel data of a Joint Photographic Experts Group (JPEG) image, so as to solve the problems existing in the known technology. First, a minimum coding unit (MCU) of the JPEG image is received, wherein the minimum coding unit includes a plurality of pixel data. Then, scaling is performed on the minimum coding unit to obtain a scaled minimum coding unit. Then, according to the format of the JPEG image, a storage rule of the scale-adjusted minimum coding unit in a memory module is determined. Then, according to the determined storage rule, a memory included in the memory module is selected, so as to write the pixel data included in the scale-adjusted minimum coding unit into the selected memory. Next, according to the format of the JPEG image, an output mode for outputting the scale-adjusted minimum coding unit is determined. Finally, according to the determined output mode, the pixel data included in the scale-adjusted minimum coding unit is output from the memory module to a frame buffer.

The present invention further provides an output buffer module for pixel data of a Joint Photographic Experts Group (JPEG) image. The buffer module includes an adjustment module, a memory access control module, a memory module and an output buffer control module. The adjustment module receives a Minimum Coded Unit (MCU) of the JPEG image, and performs scaling on the Minimum Coded Unit to obtain a scaled Minimum Coded Unit, wherein the Minimum Coded Unit includes multiple pixel data. The memory module access control module determines the storage rule of the scale-adjusted minimum coding unit in a memory module according to the format of the JPEG image. The memory module selects a memory included in the memory module according to the determined storage rule, and writes the pixel data contained in the scale-adjusted minimum coding unit into the selected memory. The output buffer control module determines the output mode of the scale-adjusted minimum coding unit according to the format of the JPEG image, and according to the determined output mode, the pixel data contained in the scale-adjusted minimum coding unit is determined by The memory module outputs to a frame buffer.

The output buffering method of the pixel data of the Joint Photographic Experts Group image described in the present invention can simultaneously reduce the production cost of the buffer and improve the performance of the buffer.

Description of drawings

FIG. 1A shows the minimum coding unit of JPEG images with formats 420 and 422V;

FIG. 1B shows the minimum coding units of JPEG images with formats 411, 422H and 444;

2 is a block diagram of an output buffer module of pixel data of a JPEG image according to the present invention;

Fig. 3 is the flow chart of the output buffering method of the pixel data of JPEG image according to the present invention;

4 is a flow chart of a method for determining an output mode according to the present invention;

5 is a flow chart of a method for selecting memory storage rules according to the present invention;

FIG. 6A shows a first memory selection rule;

FIG. 6B shows a second memory selection rule;

FIG. 7A is a schematic diagram of a first memory selection rule operating in a memory module;

7B is a schematic diagram of the operation of the second memory selection rule in the memory module;

FIG. 7C is another schematic diagram of the operation of the second memory selection rule in the memory module.

Detailed ways

In order to make the above and other objects, features and advantages of the present invention more comprehensible, a few preferred embodiments are specifically cited below, together with the accompanying drawings, and are described in detail as follows.

FIG. 2 is a block diagram of a buffer module 200 for outputting pixel data (pixel data) of a JPEG image according to the present invention. The buffer module 200 includes an adjustment module 202 , a memory module 204 , a memory access control module 206 , and an output buffer control module 208 . The buffer module 200 can still buffer pixel data decoded from an inverse discrete cosine transform (Inverse Discrete Cosine Transformation, IDCT) module under the condition of limited memory capacity, and output the pixel data to the frame buffer of the screen at a high data rate device (frame buffer). The memory module 204 of the buffer module 200 may include six memories (not shown), namely the first memory to the sixth memory. Each memory can just store a pixel block of 8 columns×8 rows, that is, the capacity of each memory is 64 (8×8=64) bytes, and one pixel occupies a storage capacity of one byte. Therefore, under the specific design of the present invention, the six memories of the memory module 204 only need 384 (64×6=384) bytes in total, which is enough to buffer the required data volume of the JPEG image, without the need of Generally requires a large amount of storage capacity.

FIG. 3 is a flowchart of an output buffering method 300 for pixel data of a JPEG image according to the present invention. The output buffering method 300 can be used to illustrate the internal operation of the buffering module 200 of FIG. 2 . As mentioned above, before the JPEG image is output to the screen for display, the resolution and size of the JPEG image must be adjusted according to the size of the screen. Therefore, the adjustment module 202 first receives a minimum coding unit (MCU) of the JPEG image in step 302, wherein the minimum coding unit includes a plurality of pixel data. At this time, it is necessary to determine whether to perform scaling on the minimum coding unit according to whether the size of the image matches the size of the frame buffer to adjust the resolution of the image. If the image needs to be adjusted, the adjustment module 202 performs scale adjustment on the minimum coding unit in step 304 to obtain a scaled minimum coding unit.

The pixel data of the scale-adjusted LCU is then sent to the memory module 204 for storage. In the memory module 204, the first memory to the third memory are used to store the pixel data of the luminance component (intensity) of the scale-adjusted minimum coding unit, and the fourth memory to the sixth memory are used to store the scale-adjusted minimum coding unit. Pixel data for the color difference component (color) of the unit. Certainly, if the minimum coding unit is not scaled by the adjustment module 202 , it is also stored in the memory module 204 . Since the capacity of the memory module 204 is only 384 bytes, a memory access control module 206 needs to determine the storage location of the input pixel data in the memory module 204 to properly complete the work of buffering the pixel data. Firstly, the memory access control module 206 determines the specific storage rule of the pixel data in the memory module 204 according to the format of the JPEG image in step 306 . Then, the memory module 204 can select a memory from the memory module 204 according to the determined storage rule in step 308, so as to write the pixel data into the selected memory. The storage rules of the memory will be described in detail with reference to FIGS. 5 , 6 , and 7A-7C in subsequent paragraphs.

Next, the output buffer control module 208 is responsible for retrieving the pixel data from the memory module 204 and outputting the pixel data to the frame buffer of the screen for image display. The output buffer control module 208 determines the output mode of the output pixel data according to the format of the JPEG image in step 310 , and the output mode can be an 8-byte or 16-byte burst output mode. The method of determining the output mode will be described with reference to FIG. 4 . Next, the output buffer control module 208 outputs the pixel data from the memory module 204 to the frame buffer according to the output mode in step 312 . Steps 302 to 312 can be executed repeatedly until all the pixel data of the JPEG image are output.

FIG. 4 is a flowchart of a method 400 for determining an output mode according to the present invention, and the output buffer control module 208 can select an output mode of pixel data according to this method. Firstly, in step 401 , it is determined whether the pixel data stored in the memory module 204 has been processed by the adjustment module 202 for scaling. If there is no scaling process, it is directly determined in step 408 to output the pixel data in the 16-byte burst output mode. If the pixel data is scaled, the output mode of the pixel data can be selected according to the image format. If the format of the JPEG image in step 402 is 411 , 422H or 444 in FIG. 1B , then the pixel data storing the luminance component will be output in step 408 in the 16-byte burst output mode. If the format of the JPEG image in step 402 is 420 or 422V in FIG. 1A , then the pixel data storing the luminance component will be output in 8-byte burst output mode in step 406 . In addition, regardless of the format of the JPEG image, if the pixel data storing the color difference component has been scaled, it will also be output in the 8-byte burst output mode in step 406 (not shown).

FIG. 5 is a flow chart of a method 500 for selecting a memory storage rule according to the present invention. The memory access control module 206 determines the storage rule of pixel data in the memory module 204 according to this method. The storage rules are divided into the first memory selection rule and the second memory selection rule. The details will be described with FIGS. The memory access control module 206 first determines in step 501 whether the pixel data of the JPEG image has been scaled by the adjustment module 202 . If the minimum coding unit of the input image has not been scaled, no matter whether it is the pixel block of the brightness component or the pixel block of the color difference component of the minimum coding unit, it is determined in step 510 to select the minimum coding unit according to the second storage unit selection rule. The storage unit to which the remaining pixels of the coding unit are to be written.

If the minimum coding unit of the input image has been scaled, it is determined in step 502 whether the input pixel block is a luma component or a color difference component. If it is a color difference component, still select the second storage unit selection rule as the storage rule in step 510 . If it is a brightness component, then in step 503, different storage rules are selected according to the format of the JPEG image. If it is judged in step 504 that the JPEG image format is 420 or 422V, and it is judged in step 506 that the three memory stores storing the luminance component are not filled with two (for example, only one memory is filled), then in step 506 In 508, the first storage unit selection rule is selected as a storage rule. Otherwise, if it is judged that the JPEG image format is 411, 422H or 444 in step 504, or it is judged that the JPEG image format is 420 or 422V in step 504 but it is judged in step 506 that the three memories storing the brightness component have filled two , the second storage unit selection rule is selected as the storage rule in step 510 .

As mentioned above, the memory module 204 includes the first memory to the sixth memory, and each memory can accommodate a pixel block of (8×8) pixels. The first to third memories store pixel data of brightness components, and the fourth to sixth memories store pixel data of color difference components. Since the pixel block of the minimum coding unit of the image has been scaled to reduce the resolution before storage, the pixel block will be reduced from the pixel block originally containing (8×8) pixels to include (n×n) pixels A pixel block of pixels, where n is a natural number less than or equal to 8. Therefore, the storage capacity required for the scale-adjusted pixel block is less than the storage capacity of a memory, and the remaining storage capacity of the memory can be used to store other pixel blocks. Therefore, when the pixel blocks are sequentially stored in the memory, part of the pixel blocks will be cut into two parts, which are respectively stored in two different memories. When the previous part of the pixel block has been written to the edge of the currently written memory, it is necessary to select the next memory according to the determined storage rule to store the next remaining part of the pixel block.

6A and FIG. 6B show the first memory selection rule 600 and the second memory selection rule 650 respectively, wherein the rules 600 and 650 are only illustrated by the first memory, the second memory, and the third memory for storing luminance components. Under the first memory selection rule 600 in FIG. 6A , when the currently written memories are the first memory, the second memory, and the third memory, the memory access control module 206 will sequentially select the third memory, the first The memory and the second memory are the next memory to write the remaining pixels. However, under the second memory selection rule 650 in FIG. 6B , when the currently written memories are the first memory, the second memory, and the third memory, the memory access control module 206 will sequentially select the second memory respectively. , the third memory and the first memory are next memory for writing remaining pixels. Similarly, if the first memory and the second memory selection rules are implemented as the fourth memory, the fifth memory, and the sixth memory for storing color difference components, then the first memory, the second memory currently written in Fig. 6A and Fig. 6B The memory and the third memory correspond to the fourth memory, the fifth memory, and the sixth memory, respectively.

Next, an example will be used to illustrate the actual operation of the first and second memory selection rules. FIG. 7A is a schematic diagram of the operation of the first memory selection rule 600 in the memory module 204 . Assuming that the input pixel block is the luminance component of the minimum coding unit of JPEG image format 420, such as Y0-Y3 of the minimum coding unit 100 in FIG. 1A, the memory access control module 206 will select the first memory selection according to the selection method 500 Rules 600 are storage rules. Firstly, the pixel block Y0 is stored in the first memory. Since the first memory still has remaining space, the previous part of the next pixel block Y1 is also stored in the first memory. At this time, the next memory must be selected according to the first memory selection rule 600 in FIG. 6A to store the second part of pixels of the pixel block Y1. At this time, the third memory is selected as the next memory according to the rule 600, so the second part of the pixel block Y1 is stored in the third memory.

Next, since the current memory is the third memory, according to the rule 600, the second memory is selected as the next memory to store the pixel block Y2. Since the second memory still has remaining space, the previous part of the next pixel block Y3 is also stored in the second memory. At this time, since the first memory is full, the previous part of the pixel blocks Y0 and Y1 stored in the first memory has been output to the frame buffer, so the first memory can store data again. Therefore, according to the rule 600, the first memory is selected as the next memory to store the second part of the pixel block Y3.

FIG. 7B is a schematic diagram of the operation of the second memory selection rule 650 in the memory module 204 . Assume that the memory module 204 continues to receive the luminance component pixel block of the next smallest coding unit of the JPEG image whose format is 420 . At this time, the currently written memory is the third memory, and the second memory is still selected as the next memory according to the rule 600 to store the previous part of the pixel block Y1. At this time, because the edges of the third memory and the second memory are full, two of the three memory have filled the pixel block to the edge, so the storage rule must be changed to the second memory selection rule according to the selection method 500 650, as shown in Figure 6B. Since the third memory is full before, the previous part of the pixel block Y0 and other pixel blocks stored in the third memory have been output to the frame buffer, so the third memory can store data again. At this time, the current memory is the second memory, so according to the rule 650, the third memory is selected as the next memory to store the remaining part of the pixel block Y1.

Since the currently written memory is the third memory, the first memory is selected as the next memory according to the rule 650 to store the previous part of the next pixel block Y2. Then, according to the rule 650, the second memory is selected as the next memory to store the second part of the pixel block Y2 and continue to fill in the former part of the next pixel block Y3. At this time, since the first memory has been cleared, only the second memory has been filled to the edge of the pixel block among the three memories. Therefore, according to the selection method 500 , the storage rule must be changed to the first memory selection rule 600 . Therefore, according to the rule 600, the first memory is selected as the next memory to store the second part of the pixel block Y3.

FIG. 7C is another schematic diagram of the operation of the second memory selection rule 650 in the memory module 204 . Assuming that the input pixel block is the luminance component of the smallest coding unit with the JPEG image format of 422H, such as Y0-Y1 of the smallest coding unit 160 in FIG. Rules 650 are storage rules. Assuming that the currently written memory is the first memory, firstly, the first memory is used to store the previous part of the pixel block Y0. Then according to the rule 650, the second memory is selected as the next memory to store the second part of the pixel block Y0 and continue to fill the former part of the next pixel block Y1. Then, according to the rule 650, the third memory is selected as the next memory to store the second part of the pixel block Y1.

The present invention provides an output buffering method of pixel data of JPEG image, which can still exert high data output rate under limited storage capacity, so as to satisfy the JPEG image format with increasing data demand speed. According to different JPEG image formats, the present invention provides different data output modes and memory storage rules, so as to perform different buffering processes on pixel data according to image formats under the limitation of storage capacity. Therefore, the present invention can simultaneously reduce the production cost of the buffer and improve the efficiency of the buffer.

The above description is only a preferred embodiment of the present invention, but it is not intended to limit the scope of the present invention. Any person familiar with this technology can make further improvements on this basis without departing from the spirit and scope of the present invention. Improvements and changes, so the protection scope of the present invention should be defined by the claims of the present application.

A brief description of the symbols in the drawings is as follows:

200: output buffer module

202: Adjustment module

204: memory module

206: memory access control module

208: Output buffer control module

Claims (10)

1. an output buffering method of the pixel data of the Joint Photographic Experts Group image, it is characterized in that, comprising: receiving a minimum coding unit of the Joint Photographic Experts Group image, wherein the minimum coding unit includes a plurality of pixel data; performing scale adjustment on the minimum coding unit to obtain a scale-adjusted minimum coding unit; determining a storage rule for the scale-adjusted minimum coding unit in a memory module according to the format of the Joint Photographic Experts Group image; selecting a memory included in the memory module according to the determined storage rule, so as to write the pixel data included in the scale-adjusted minimum coding unit into the selected memory; According to the format of the Joint Photographic Experts Group image, determine an output mode for outputting the scale-adjusted minimum coding unit; and According to the determined output mode, the pixel data included in the scale-adjusted minimum coding unit is output from the memory module to a frame buffer. 2. The output buffering method of the pixel data of the Joint Photographic Experts Group image according to claim 1, wherein the scale adjustment of the minimum coding unit is to make the output size of the Joint Photographic Experts Group image conform to the frame buffer size of. 3. The output buffering method of the pixel data of the Joint Photographic Experts Group image according to claim 1, wherein the memory module includes six memories, and the three memories included in the first group of memories are used to store the The pixel data of the luminance component of the scale-adjusted minimum coding unit are respectively a first memory, a second memory and a third memory, and the three memories included in the second group of memories are used to store the scale-adjusted minimum coding unit The pixel data of the color difference components are respectively the fourth memory, the fifth memory and the sixth memory. 4. The output buffering method of the pixel data of the Joint Photographic Experts Group image according to claim 3, characterized in that, the six memories can respectively store a pixel block data volume of m columns×m rows, not scaled The pixel data of the brightness component or color difference component of the adjusted minimum coding unit is a plurality of pixel blocks of m columns×m rows, and the pixel data of the brightness component or color difference component of the scale-adjusted minimum coding unit is a plurality of n columns×m rows of pixel blocks. A pixel block of n rows, where n is a natural number less than or equal to m. 5. The output buffering method of the pixel data of the Joint Photographic Experts Group image according to claim 3, wherein the storage rules include a first memory selection rule and a second memory selection rule, wherein when the currently written memory is respectively When it is the first memory, the second memory and the third memory, the first memory selection rule respectively selects the third memory, the first memory and the second memory as the next memory to write the remaining pixels, and when the current write When the memories are respectively the first memory, the second memory and the third memory, the second memory selection rule respectively selects the second memory, the third memory and the first memory as the next memory for writing the remaining pixels. 6. The output buffering method of the pixel data of the Joint Photographic Experts Group image according to claim 5, characterized in that, when the currently written memories are respectively the fourth memory, the fifth memory and the sixth memory, the first A memory selection rule respectively selects the sixth memory, the fourth memory and the fifth memory as the next memory to write the remaining pixels, and when the currently written memory is the fourth memory, the fifth memory and the sixth memory respectively , the second memory selection rule respectively selects the fifth memory, the sixth memory and the fourth memory as the next memory for writing the remaining pixels. 7. The output buffering method of pixel data of a Joint Photographic Experts Group image according to claim 5, wherein if a target pixel data input to the memory module is the luminance component of the scale-adjusted minimum coding unit, The storage rule of the proportion-adjusted minimum coding unit is determined according to the following steps, so that when the current memory for writing the target pixel data is full, it is used to select the next memory for writing the remaining pixels of the target pixel data : If the format of the Joint Photographic Experts Group image is 420 or 422V, and two of the three memories of the first set of memories are not filled, then using the first memory selection rule to determine the next memory; if the format of the Joint Photographic Experts Group image is 420, and two of the three memories of the first set of memories are filled, then using the second memory selection rule to determine the next memory; and If the format of the Joint Photographic Experts Group image is 422H, 444 or 411, then use the second memory selection rule to determine the next memory. 8. The output buffering method of the pixel data of the Joint Photographic Experts Group image according to claim 6, wherein if a target pixel data input to the memory module is the color difference component of the scale-adjusted minimum coding unit, Then the second memory selection rule is determined as the storage rule, so that when the current memory for writing the target pixel data is full, it is used to select the next memory for writing the remaining pixels of the target pixel data. 9. The method for output buffering pixel data of a Joint Photographic Experts Group image according to claim 6, wherein if the smallest coding unit has not been scaled, the second storage selection rule is determined to be the storage rule, In order to select the next memory for writing the remaining pixels of the target pixel data when the current memory for writing a target pixel data of the minimum coding unit is full. 10. The output buffering method of pixel data of the Joint Photographic Experts Group image according to claim 1, wherein the step of outputting the pixel data from the memory module to the frame buffer further comprises: The output mode in the case of the scale-adjusted LCU is according to: When the format of the Joint Photographic Experts Group image is 420 or 422V, adopt the 8-byte burst output mode as the output mode; and When the format of the Joint Photographic Experts Group image is 444, 411, 422H, adopt the 16-byte burst output mode as the output mode; and If it is the smallest coding unit without scale adjustment processing, the 16-byte burst output mode is determined to be the output mode.

Images