JP3119366B2 - Image processing apparatus and method - Google Patents

Description

Description: BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an image processing apparatus and method capable of transferring variable-length encoded moving image data from a memory.

2. Description of the Related Art Conventionally, an image processing apparatus processes a still image, and has a configuration in which image data having a large data amount, such as a moving image, cannot be directly handled. Although an image processing device capable of processing a moving image has been recently announced, this type of device directly connects a video camera and a video frame memory to directly display a moving image input from the video camera on a display. It is configured to display.

[Problems to be solved by the invention] However, in the case of a system that handles still images,
The size of the image memory for storing image data is A4 size (4752 x 3360 pixels x 3 colors for 400 dpi, approx. 46M Byt
e) is less than or equal to about
When 288 pixels x 3 colors are stored one-dimensionally as one frame, the number of frames becomes approximately 154 frames. When the image data is stored as a two-dimensional image data in a rectangular area, approximately 143 frames are stored.
It is about the frame.

That is, in the case of storing a moving image whose frame is updated every 1/30 seconds, the image memory can store only image data of about 5 seconds. There is a drawback that moving images of objects cannot be handled.

Also, every 1/30 second, one frame (360 × 288 pixels × 3
Color) image data from the image memory unit to the video frame memory, the transfer capacity is 72M / s.
It requires high-speed image transfer means of bits (= 360 × 288 pixels × 3 colors × 30 frames × 8 bits) or more.

For this reason, even if a high-speed image data transfer means is not provided or equipped, it is possible to change the transfer start address and transfer destination address to an arbitrary position or transfer an arbitrary image size. There was a drawback that you couldn't.

Furthermore, when moving image data is encoded, it has a disadvantage that it becomes a variable-length code and cannot be displayed from an arbitrary frame or displayed in a moving image such as reverse reproduction, fast forward, and reverse fast forward.

SUMMARY An advantage of some aspects of the invention is to provide an image processing apparatus and method capable of transferring variable-length encoded moving image data from a memory at a high speed.

[Means and Actions for Solving the Problems] In order to achieve the above object, the present invention is an image processing apparatus capable of transferring variable-length coded moving image data from a memory, comprising a variable-length coded moving image data. Means for managing a storage address of each frame stored in the image memory of the moving image data, and first data for transferring variable-length coded moving image data managed by the managing means to the decompression means. The image processing apparatus further includes a transfer unit and a second data transfer unit that transfers the moving image data expanded by the expansion unit to the image monitor display device so that the moving image is displayed on the image monitor display device.

Hereinafter, a preferred embodiment according to the present invention will be described in detail with reference to the accompanying drawings.

<Description of Configuration (FIG. 1)> FIG. 1 is a schematic block diagram showing the configuration of the image processing apparatus in the present embodiment. In FIG. 1, reference numeral 1 denotes a system bus, which is used when 11 CPUs and 12 CPUs transfer system data, a system address, a system control signal, and the like, and control the entire system. An image data high-speed bus 2 transfers image data exclusively between the image memory unit and the video frame memory unit. An image memory unit 5 stores image data and controls image transfer.
In the image memory section 5, reference numeral 3 denotes an image memory;
Image data of a maximum of 4096 × 4096 pixels is stored. Reference numeral 4 denotes an address generator, which performs unique address control for high-speed transfer of image data.

A display control unit 9 receives the transfer of the image data and controls the display of the image data. In the display control unit 9, reference numeral 6 denotes an address generator, which performs unique address control for receiving transfer of image data.
A video frame memory 7 stores image (video) data of at least one display screen (for example, 1280 × 1024 pixels). Reference numeral 8 denotes a display converter, which converts the video data stored in the video frame memory into an analog image signal by D / D.
A conversion etc. Reference numeral 10 denotes, for example, a CRT display device, which displays an image according to an analog image signal. An image data storage unit 15 stores a large amount of image data. In the image data storage unit 15, a disk controller 13 converts a disk access control signal from the CPU 11, for example, to a control signal for an actual disk. Reference numeral 14 denotes a disk, which is a disk device such as a hard disk or a magneto-optical disk.

<Address Generator (FIG. 2)> Next, the configuration of the address generator in this embodiment is shown in FIG.
This will be described with reference to the drawings.

It should be noted that the address generators 4 and 6 in the present embodiment have functions of a transmitting side and a receiving side of image data.

FIG. 2 is a detailed block diagram showing the configuration of the address generators 4 and 6. In the figure, reference numeral 200 denotes a bidirectional data multi-multiplexer (MPX) for switching the flow of system data and image data. A multiplexer 201 switches between a system address and an internally generated address. The image memory 5 is composed of, for example, a DRAM, and reads and writes image data at high speed in accordance with a strobe signal.

Reference numeral 203 denotes a Y register, which holds a transfer start Y address of image data. A row counter 204 is loaded with the contents of the Y register 203 by a vertical synchronization (V-Sync) signal 211 for transfer, and thereafter a horizontal synchronization (H-Sync) signal for transfer. Incremented by 210. An X register 206 holds an X address at which image data is transferred. Reference numeral 205 denotes a column counter which is loaded with the contents of the X register by the H-Sync signal 210 and thereafter incremented by the clock (CLK) signal 212 for transfer.

Reference numerals 208 and 209 denote shifters that shift the outputs (internally generated addresses) of the counters 204 and 205, respectively, in accordance with a command from a control register 207 described later. Reference numeral 207 denotes a control register which holds information indicating in which direction and how many bits the outputs of the counters 204 and 205 are to be shifted. That is,
If the value of the register 207 is “0”, there is no shift, and “−”
A shift direction and the number of bits, such as 1 bit in the lower direction if "1" and 1 bit in the upper direction if "1", are registers 603 and 604, which have passed through the shifters 208 and 209. Address data to be added to the address information are respectively held.601 and 602 are adders, and the address information register 60 passed through the shifters 208 and 209.
Add the contents of 3,604.

On the other hand, reference numeral 303 denotes a Y length register, which holds the transfer Y length of the image data. 307 is a counter,
The signal is cleared by the V-Sync signal 211 and then the H-Sync signal 210
Is incremented by one. Reference numeral 305 denotes a comparator, which outputs a signal of logic "1" level while the content of the counter 307 is smaller than the content of the Y length register 303. Reference numeral 304 denotes an X length register that holds the transfer X length of image data. 308 is a counter, H-Syn
Cleared by c signal 210 and then incremented by one by CLK signal 212. Reference numeral 306 denotes a comparator, and a counter 308
Is smaller than the content of the X length register 304, a signal of logic "1" level is output. Reference numeral 309 denotes an AND circuit, and while the outputs from the comparators 305 and 306 are both at the logical “1” level, the video frame memory 7 and the image memory 5 (RA
M), and outputs a chip enable signal CE. Therefore, during this time, image data is written by the strobe signal.

The video frame memory 7 is composed of, for example, a two-port RAM, and connects the output of the serial access port to the display converter 8 and connects the random access port to the image data writing side (MPX200). This allows RA
Writing and reading of image data to M7 can be performed simultaneously,
Monitor image data in real time.

<Moving Image Management Table (FIG. 3)> FIG. 3 is a diagram showing a general format of the moving image management table in the present embodiment. In the figure, a frame number field 31 stores a frame number of moving image data for each frame. When moving image data is stored in the image memory unit 5, the storage position information field 32 stores the start address information in the image memory 3 in the image data storage unit 15.
If the image data is stored in the
14 stores information such as the number of tracks and the number of sectors.

<Description of Operation> A data transfer operation in the present apparatus having the above configuration will be described below.

To simplify the description, the resolution of the display is 1280 × 1024 pixels, the size of the moving image to be handled is 360 × 288 pixels × 3 colors, and the size of the image memory is 4096 × 4096 pixels × 3 colors.
Then, it is assumed that the size of the video frame memory is 1280 × 1024 pixels × 3 colors for one screen. Further, an example will be described in which moving image data is stored in the image memory unit 5 in an uncoded data format.

FIG. 5 is a view showing an image displayed on the display 10. As shown in FIG. In the figure, a portion 52 is, for example, a state in which a still image such as a map of Japan is displayed, and the operator specifies a rectangular area 51 with a pointing device such as a mouse (not shown), and the portion 52 is displayed. To display a certain moving image. CPU11 that input this instruction
Reads out the corresponding moving image data from the image data storage unit 15 or the like, and loads the moving image data into the image memory 3 in the order from frame 0 to 1, 2,... As shown in FIG. 4a. Also,
The CPU 11 creates a moving image management table as shown in FIG. 4b in parallel with this operation.

Here, for example, the moving image data of the frame 0 is loaded to the position 40 in the image memory 3, and the moving image data of the frame 1 is moved to the position 41.
The data of 1,... Are loaded to the corresponding positions. The rectangular area 42 indicates an area of one frame of moving image data. At this time, when the X and Y addresses corresponding to the position 40 in the image memory 3 are set to “0”, for example, the X-address is added to the storage position information corresponding to the frame 0 of the table. = 0,
Y address = 0, X address = 360, Y address = 0 for frame 1, X address = 720, Y address =
0,... X address = 0, Y address = 288,.
Is set as shown in FIG. 4b.

Here, when the above-described loading process is completed, the process proceeds to a process of displaying a moving image at the designated position 53 on the display 10 next.

First, if the upper left position 53 of the moving image display area 51 specified by the operator is, for example, (X, Y) = (200,100),
In order to display the moving image data loaded in the image memory 3 in the rectangular area 51 on the display 10, the position 40 in the image memory 3 is shifted from the position 40 in the video frame memory 7.
The image data may be sequentially transferred to the 53, from the position 41 to the position 53,... Every 1/30 second. That is, the CPU 11 makes the following initial settings while sequentially referring to the top of the moving image management table, that is, the storage position information of the frame 0.

<Image memory unit 5> X register 206 = 0 Y register 203 = 0 MPX200 = High-speed bus connection MPX201 = Use internal address Control register 207 = 0 Register 603 = 0,604 = 0 X length register 304 = 360 Y length register 303 = 288 < Display control unit 9> X register 206 = 200 Y register 203 = 100 MPX200 = use high speed bus MPX201 = use internal address Register 207 = 0,603 = 0,604 = 0 X length register 304 = 360 Y length register 303 = 288 After completion, when the CPU 11 starts, the address (X, Y) of the image memory 3 = (0, 0)
(360 × 288) pixels of image data (frame 0
Of the video frame memory 7 (20 image data).
(0,100) to the rectangular area 53 starting at high speed. By performing this operation every 1/30 second according to the storage position information in the moving image management table, the moving image stored in the image memory 3 is sequentially transferred to the frame 1, the frame 2, and so on. It can be displayed in a specified rectangular area 53 on the display 10.

In the above-described embodiment, normal moving image reproduction has been described. However, reproduction in the reverse direction can be performed by referring to the moving image management table from the end and performing the above-described transfer operation. In this case, when the moving image is fast-forwarded, reverse fast-forwarded, or the like, the moving image management table is referred to every few entries instead of being incremented or decremented by one. For example, if you refer to every third,
It can be played back at twice the speed, and every third
You can play at twice the speed.

According to the present embodiment, by providing a high-speed image data transfer / transfer unit for transferring image data between the image memory unit and the video frame memory at high speed separately from the system bus, one frame can be transferred every 1/30 second. Image data can be transferred.

The format of the image data stored in the image memory is
Having a management information table for moving image data in frame units stored in the image memory, regardless of whether it is coded or uncoded, and in the case of coding, regardless of fixed-length or variable-length coding Accordingly, normal video playback can be performed by sequentially transferring the video frames from the beginning to the video frame memory according to this table. Furthermore, by skipping this table several times, or by referring back,
Moving images such as reverse fast forward and reverse playback can be displayed.

Also, by changing the address in the video frame memory of the transfer destination, the position at which the moving image is displayed on the display can be changed to an arbitrary position. Furthermore, if still image data is transferred to the video frame memory before moving image transfer, and then moving image data is transferred, combined display of a still image and a moving image is also possible. By providing enlargement / reduction means in the image data transfer means, processing such as enlargement / reduction of a moving image can be performed in real time.

Another Embodiment Next, another embodiment according to the present invention will be described below with reference to the accompanying drawings.

In this embodiment, an example will be described in which encoded moving image data is directly loaded from the image data storage unit 15 into the video frame memory 7 and the moving image data is displayed on the display 10.

Here, the moving picture coding method is, for example, a method proposed by a standardization committee such as motion assurance + DCT coding, and is not particularly limited. It is also assumed that the image data storage unit 15 stores image data that has been encoded in advance by a device (not shown) or the CPU 11 or the like.

Further, it is assumed that the moving image management table for the moving image is created by the CPU 11 or the like when the moving image data is encoded. Since the moving image management table in this case is a table for the image data storage unit 15, the storage position information field is start information on the disk 14. For example, it is assumed that the number of sectors, the number of tracks, and the like of the disk 14 are stored and stored in an area different from the image data area in the image data storage unit 15 in advance.

Hereinafter, the operation principle of another embodiment will be described in detail with reference to FIG.

FIG. 6 is a schematic block diagram showing the configuration of an image processing apparatus according to another embodiment. Note that the same operations as those shown in FIG. 1 are denoted by the same reference numerals, and description thereof will be omitted.

In the figure, reference numeral 64 denotes a block for expanding encoded image data. Reference numeral 61 denotes an expander for expanding the encoded data in accordance with the encoding method, 63 denotes a buffer memory for several lines which temporarily stores the expanded image data output from the expander 61, and 62 denotes an address generator shown in FIG. This is an address generator that performs unique address control so as to enable high-speed image data transfer similarly to the devices 4 and 6.

FIG. 7 is a diagram showing a configuration of a moving image management table in this embodiment. Here, similarly to the above-described embodiment, when an instruction to display a moving image is specified by a device (not shown) from the operator, the moving image management table of the corresponding moving image is stored in the work memory of the CPU 11 from the image data storage unit 15. The image data of the corresponding frame is read out and transferred to the decompressor 61 via the system bus 2 as needed in accordance with this table. Then, the decompressor 61 decompresses the transferred image data as needed, and develops it in the buffer memory 63.

The address generators 62 and 6 are provided with X,
Initialization of registers such as the start address of each Y, the X and Y lengths, the high-speed bus connection, and the use of internal addresses are performed, and image data transfer is started.

The description of the portion for performing the same initial setting as in the above-described embodiment will be omitted. Here, the different part is that, when the coding method of the moving image is block coding such as DCT, the buffer memory 63 needs the number of lines for the sub-scanning for this block, and the CPU side initializes the source side. Is rotated by this number of lines.

By performing the above processing, a moving image can be directly displayed on the display 10 from the image data storage unit 15, as in the above-described embodiment.

Further, as in the above-described embodiment, by changing the method of referring to the moving image management table, a reproduction method such as reverse reproduction, fast-forward, and reverse-forward can be performed.

Further, in the above two embodiments, the image data storage unit
Although the moving image is displayed on the display 10 from 15, both the image memory unit 5 and the image data storage unit 15
It is also possible to transfer image data and display a plurality of moving images on the display 10 at once. Further, it can be easily inferred that a plurality of moving images can be displayed on the display 10 only by the image memory unit 5 or the image data storage unit 15 alone.

Further, the information in the storage position information field in the moving image management table includes the start address,
It can be easily inferred that the case of the image data storage section 15 is not limited to the number of disks and the number of sectors, but may be various information.

The moving image management table has been described for the case where moving images are stored in the image memory unit 5 or the image data storage unit 15 such as a disk. However, the present invention is not limited to this. It is easy to guess that we can respond to

As described above, the moving image encoding method is not particularly limited, and can be dealt with by changing a decompressor or the like according to an encoding algorithm.

Furthermore, when encoding image data, if the encoded image data amount is smaller than the transfer capability of the image storage unit or the transfer capability of the system bus, an encoding decompressor is provided at the entrance of the video frame memory, The image can be directly transferred from the image storage unit to the video frame memory, and the moving image can be displayed on the display without being transferred to the image memory unit once.

Further, large-scale hardware is required to handle moving images. However, according to the present invention, special hardware for displaying moving images is not particularly required. It is possible to process the image. Also,
By handling moving images on a frame basis, various playback methods are possible.

[Effects of the Invention] As described above, according to the present invention, the storage address of each frame in the image memory of the moving image data that has been subjected to the variable length coding is managed, and the managed moving image data is transmitted to the decompression means. Since the data is transferred, decompressed and output to the monitor, a moving image corresponding to the moving image data decompressed at high speed can be reproduced as a visible image on the monitor with a simple configuration.

[Brief description of the drawings]

FIG. 1 is a schematic block diagram showing a configuration of an image processing apparatus according to this embodiment, FIG. 2 is a detailed block diagram showing a configuration of an address generator shown in FIG. 1, and FIG. 4a and 4b show a state where moving images are stored in an image memory unit, and FIGS. 4a and 4b show contents of a moving image management table. FIG. 5 shows moving images displayed on a display. FIG. 6 is a schematic block diagram showing the configuration of an image processing apparatus in another embodiment. FIG. 7 is a diagram showing the contents of a moving image management table in another embodiment. In the figure, 1 ... system bus, 2 ... high-speed image data bus, 3 ... image memory, 4, 6 ... address generator, 7 ...
... frame memory, 8 ... display converter, 10 ... display, 11 ... CPU, 13 ... disk controller, 14 ...
... a disk.

──────────────────────────────────────────────────続 き Continued on the front page (58) Fields surveyed (Int.Cl. ⁷ , DB name) H04N 7/24 H04N 1/21 H04N 5/76-5/907

Claims (3)

(57) [Claims] An image processing apparatus capable of transferring variable-length coded moving image data from a memory, comprising: a storage address of each frame stored in an image memory of the variable-length coded moving image data. A first data transfer unit that transfers variable-length encoded moving image data managed by the management unit to a decompression unit; and a decompression unit that displays a moving image on an image monitor display device. An image processing apparatus comprising: a second data transfer unit configured to transfer the expanded moving image data to an image monitor display device. 2. The image processing apparatus according to claim 1, wherein said first data transfer means and said second data transfer means are means for transferring data via different buses. 3. An image processing method capable of transferring variable-length coded moving image data from a memory, comprising: a storage address of each frame stored in an image memory of the variable-length coded moving image data; A first data transfer step of transferring the variable-length coded moving image data managed in the management step to a decompression means; and a decompression means for displaying a moving image on an image monitor display device. Transferring the decompressed moving image data to an image monitor display device.