JPH11341288A - Digital video system - Google Patents

Abstract

PROBLEM TO BE SOLVED: To attain reduction of a utility amount of a frame memory. SOLUTION: A subtraction/difference compression means 149 is provided, a difference between video data before being stored in a frame memory 16 and data before one picture element is obtained, subtraction/difference compression is performed by referring to a conversion table 148 and obtaining a table value corresponding to the differential value and a utility amount of a frame memory is reduced in comparison with a case in which uncompressed video data are stored in the frame as they are by storing this compression processing result in the frame memory.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a digital video technique, and more particularly to a compression and decompression technique therein, and more particularly to a technique effective when applied to a digital still camera.

[0002]

2. Description of the Related Art Generally, high-speed access is required between an image input / output unit of a digital still camera, which is an example of a digital video system, and a frame memory in synchronization with display timing and capture timing. Has not been compressed or expanded. For example, in a display system with a resolution of 800 × 600, when trying to perform non-interlaced display at 60 Hz vertical synchronization,
The dot clock frequency is 40 Hz or more. At this speed, it is extremely difficult to perform conventional operations related to image compression / decompression in real time. Most of the image compression / decompression algorithms themselves are based on the premise that the image data to be processed is stored in the frame memory. Data cannot be compressed in real time.

[0003] Examples of documents describing image compression / expansion techniques in digital still cameras include:
A color image and image quality evaluation (page 253) issued by the Technical Information Association on June 25, 1992. "

[0004]

When a system is constructed by a conventional frame memory system, assuming full color (24 bits / pixel), a system of 800 × 600 has a resolution of about 11 Mbits and a system of 1024 × 768 has a resolution of about 11 Mbits. A system of about 18 Mbits and 1600 × 1200 requires a memory capacity of about 44 Mbits. If the amount of data to be handled increases as the image quality further increases in the future, the storage capacity of the frame memory for storing the data must be increased.Therefore, there is a concern that the cost of the frame memory, the power consumption, and the mounting area will increase. You.

An object of the present invention is to provide a technique for reducing the used capacity of a frame memory included in a digital video system.

[0006]

The following is a brief description of an outline of a typical invention among the inventions disclosed in the present application.

That is, processing means (11, 13) for processing video data and a frame memory (1) for storing video data to be processed by the processing means.
6), a conversion table (148) storing table values referred to in differential compression and differential expansion processing, and the video data before being stored in the frame memory. Subtraction / difference compression means (149) for performing subtraction / difference compression by obtaining a difference from data one pixel before, and obtaining a table value corresponding to the difference value by referring to the conversion table.
And the output data of the subtraction / difference compression means is stored in the frame memory.

According to the above means, the subtraction / difference compression means obtains the difference between the video data before being stored in the frame memory and the data one pixel before, and refers to the conversion table. Then, subtraction / difference compression is performed by obtaining a table value corresponding to the difference value, and the result of the compression processing is stored in the frame memory. This achieves a reduction in the used capacity of the frame memory as compared with a case where uncompressed video data is stored in the frame memory as it is.

At this time, a differential expansion / addition means (147) for expanding the data compressed by the subtraction / difference compression means with reference to the conversion table can be provided.

Further, it is possible to provide a video input section (18) for capturing a video signal via an optical system and a video output section (17) for monitoring the captured video signal.

In order to perform the compression processing by the subtraction / difference compression means in real time, it is preferable to process the video data input to the subtraction / difference compression means on a raster basis.

Further, in order to make it possible to change the table value, it is preferable to form the conversion table in a rewritable memory.

In order to reduce the size of the conversion table, the step width of the conversion table may correspond to a quadratic curve.

[0014]

FIG. 1 shows a configuration example of a digital still camera which is an example of a digital video system according to the present invention.

Digital still camera 1 shown in FIG.
00 is not particularly limited, but a CPU (central processing unit)
Is realized by: It includes a video processing unit 11, a flash memory 12, a JPEG / LSI 13, a compression / decompression unit 14, a frame memory 16, a video output unit 17, and a video input unit 18.

Although the image input unit 18 is not particularly limited,
Optical system and CCD (Charge Coupled Device)
ce), an A / D (analog / digital) converter for converting a video signal from the solid-state imaging device into a digital signal, and a digital signal output from the A / D converter being γ ( DSP (digital / gamma) for performing various image processing such as gamma correction
Signal processor). This video input unit 18
The video signal output from is composed of 8 bits each for RGB or YUV, is indicated by “8 bits × 3”, and is transmitted to the compression / expansion unit 14 at the subsequent stage.

The compression / decompression unit 14 performs a compression process on the pre-compression data to reduce the data amount, and a decompression for decompressing the data (pre-decompression data) compressed by the compression function. Including functions. The data compressed by the compression function is stored in the frame memory 16, read out as needed, decompressed by the decompression function, and then stored in the JPEG / LSI 13,
It is transmitted to the video processing unit 11. After the compressed data is expanded by the expansion function, the compressed data is displayed on an appropriate display unit such as a liquid crystal display via the video output unit 17.

Here, a detailed configuration example of the compression / expansion unit 14 will be described.

A subtraction / difference compression unit 149 for performing subtraction / difference compression processing on the pre-compression data is provided, and a difference decompression / addition processing unit 147 for performing the differential expansion / addition processing on the pre-decompression data. Is provided. The processing in the subtraction / difference compression unit 149 and the difference expansion / addition processing unit 147
The differential compression / expansion conversion table 148 is referred to for the processing in. The subtraction / difference compression section 149 includes an input level latch 151 for holding previous pixel data used for difference processing, and a write data latch 150 provided exclusively for processing on the video processing section 11 side. Are combined. The difference expansion / addition processing unit 147 includes an output level latch 146 for holding previous pixel data,
And a read data latch 145 provided exclusively for processing on the video processing section 11 side.

In the capture operation, first, the video input unit 1
8 is input to the subtraction / difference compression unit 149. Then, the subtraction / difference compression unit 14
9 is the input data before compression and the input level latch 1
The difference between the previous pixel data held in 51 and the obtained difference is obtained, and the obtained difference is subjected to compression processing using the difference compression / expansion conversion table 148. Although the differential compression / expansion conversion table 148 is not particularly limited, it is a rewritable dynamic random access memory or static random access memory, and the table values are rewritten according to the video data handled by the present system. It has become. The rewriting of this table value is performed by the C
This is performed by a PU (central processing unit).

A buffer memory 142 is provided, and read / write access to video data from the video processing unit 11 is performed via the buffer memory 142. A memory transfer controller 143 is provided, and the address control when data stored in the frame memory 16 is accessed from the video processing unit 11 is performed by the memory transfer controller 143. A display / capture controller 141 is provided, and a subtraction / difference compression unit 149 is provided.
The address control when transferring the compressed data from the frame memory 16 to the frame memory 16 and the reading of the data before decompression from the frame memory 16 are performed by the address control of the display / capture controller 141. A video address signal output from the display / capture controller 141 and a transfer address signal output from the memory transfer controller 143 are selectively selected from the frame memory 16.
An address bus switching unit 144 for transmitting the data to the address terminal of the first embodiment is provided. Address bus switching unit 1
The operation of 44 is performed by the CPU in which the video processing unit 11 is realized.

In the display operation, the data before expansion read from the frame memory 16 is input to the difference expansion / addition unit 147. The differential expansion / addition unit 147 performs differential expansion processing with reference to the differential compression / expansion conversion table 148, and the difference obtained by the differential expansion processing is added to the previous pixel data held by the output level latch 146. The added data is transmitted to the video output unit 17 as decompressed data.

Here, the display / capture controller 141, the memory transfer controller 143, and the address bus switching unit 144 are provided as described above, and the address control of the frame memory 16 is performed by the address switching of the address bus switching unit 144. For the following reasons.

Read / write access of image data from the video processing unit 11 is performed via the buffer memory 142. Since only the differentially compressed data is stored in the frame memory 16, decompression processing is always required.
Moreover, during the display operation and the capture operation, the video processing unit 11
Performing the above-described decompression processing by the video processing unit 11 is very inefficient because the side must wait.

Therefore, using the difference expansion / addition unit 147, the memory transfer controller 143 and the buffer memory 1 are used.
According to 42, access is made collectively in block units. Since the compression / expansion processing is not performed during the flyback period even during the display operation or the capture operation, the memory transfer controller 143 accesses intensively during that period. That is, during the flyback period, the address bus switching unit 144
, The memory transfer controller 143 is selected, and data is read based on the transfer address. Since the access may not be completed in one retrace period depending on the transfer size and the pixel position, the output level latch 146 and the input level latch 151 are not used for holding the previous pixel data, and are provided separately therefrom. The read data latch 145 and the write data latch 150 are used.

The JPEG / LSI 13 performs a compression process and a decompression process on a video image according to a JPEG predetermined encoding system which is an international standard, and is formed as one semiconductor integrated circuit. There are three types of element technologies in this compression. That is, "DCT (discrete cosine transform)" for converting the value of the spatial coordinates into frequency, "quantization" for reducing the data amount by truncating the word length of the converted coefficient, and There are three types of "variable length coding" for assigning a code having a length suitable for the frequency of occurrence of data. In DCT, the original image is divided into small blocks,
The pixel values of each block are multiplied by a coefficient unique to DCT and added. As a result, the spatial coordinate data is converted into frequency data. Although DCT alone cannot compress,
Because it enables efficient compression tailored to human vision,
A high compression rate can be realized by the subsequent quantization. JPEG
The data to be processed in the LSI 13 is the frame memory 1
6 and expanded by the differential expansion / addition unit 147, that is, data after expansion. JPEG L
The data compressed by the SI 13 is stored in the flash memory 12, which is a nonvolatile memory. The data is transferred to a personal computer or the like via a serial interface or the like.

FIG. 2 shows a configuration example of the differential compression / expansion conversion table 148.

When 8-bit data is compressed into 4-bit data, the table value has 16 levels. For example, + 255-
If the data is tabulated in the maximum range of 255, it is necessary to cover even an impossible difference value, and the table value is wasted. Therefore, as shown in FIG. 2, a maximum difference value in the positive direction and a minimum difference value in the negative direction are obtained based on the previous pixel position, and 16 types of tables TB1 to TB16 are prepared. If there is room in the circuit, it is best to use 256 types. As shown in FIG. 3, the table values are divided into 16 sections from the maximum difference value in the positive direction to the minimum value in the negative direction, and correspond to a quadratic curve, so that subtle changes are possible. The distribution is made so that the width is fine and the step width is increased for steep changes. This is JP
As in the case of EG and the like, the emphasis is placed on low-frequency components that are easy for the human eye to perceive, and by increasing the step size for abrupt changes, the differential compression / expansion table 1
48 is reduced in size.

FIG. 4 shows that the table value shown in FIG.
In this case, if this table value is used, there may be cases where it is not possible to respond to sudden changes,
Conversely, the reproducibility of the low frequency component is improved.

FIG. 5 shows the reproducibility for high-frequency input when the table values shown in FIG. 2 are used.
FIG. 7 shows the reproducibility for a low frequency input when the table values shown in FIG. 2 are used, and FIG. 7 shows the reproducibility corresponding to the high frequency input when the table values shown in FIG. 4 are used. FIG. 8 shows the reproducibility corresponding to the low-frequency input when the table values shown in FIG. 4 are used. 5 to 8, the horizontal axis is a horizontal pixel position,
The vertical axis is the signal level.

For high-frequency input, the reproducibility is higher when the table values shown in FIG. 2 are used (see FIG. 5) than when the table values shown in FIG. 4 are used (see FIG. 7). Are better. However, for low frequency inputs,
When using the table values shown in FIG. 2 (see FIG. 5)
Rather than using the table values shown in FIG. 4 (see FIG. 7), the reproducibility is superior. Therefore, it is very efficient to determine the table value according to the image data to be handled.

According to the above example, the following functions and effects can be obtained.

(1) The subtraction / difference compression unit 149 and the difference expansion / addition unit 147 provided between the video input / output unit 18 and the frame memory 16 hold the signal level of the pixel output at that time. By doing so, it is not necessary to store the absolute level of the signal in the frame memory 16 and only the difference from the previous pixel needs to be stored. Although the data amount of the difference value is a maximum of (original data + 1) bits, there is a merit that the previous signal level can be expressed even if it is compressed. In addition, by making the step size finer in response to minute changes and increasing the step size for steep changes, non-linear compression is performed. Can be reproduced.

(2) Since the amount of data stored in the frame memory 16 can be reduced by the operation and effect of the above (1), high image quality can be achieved without increasing the storage capacity of the frame memory 16. can do.

(3) Differential compression /
By forming the expansion conversion table 148, the table values can be changed, and the differential compression / expansion conversion table 1 can be changed according to the frequency component of the video data to be processed.
Forty-eight table values can be changed, and compression and decompression processing suitable for video data to be processed can be performed.

Although the invention made by the inventor has been specifically described based on the embodiments, the present invention is not limited to the above embodiments, and various modifications can be made without departing from the gist of the invention. Needless to say.

For example, the JPEG processing may be performed by the video processing unit 11.

In the above description, the case where the invention made by the present inventor is applied to a digital still camera, which is the field of application, has been described.
The present invention is not limited to this, and can be widely applied to various digital video systems.

The present invention can be applied on condition that it includes at least processing means for processing video data and a frame memory for storing video data to be processed by the processing means. .

[0040]

The effects obtained by typical ones of the inventions disclosed in the present application will be briefly described as follows.

That is, the difference between the video data before being stored in the frame memory and the data one pixel before is obtained by the subtraction / difference compression means, and a table corresponding to the above-mentioned difference value is referred to by referring to the conversion table. The subtraction / difference compression is performed by obtaining the value, and the result of the compression processing is stored in the frame memory. Therefore, the used capacity of the frame memory is reduced as compared with the case where the uncompressed video data is stored in the frame as it is. Can be achieved.

[Brief description of the drawings]

FIG. 1 is a block diagram illustrating a configuration example of a digital still camera as an example of a digital video system according to the present invention.

FIG. 2 is an explanatory diagram of a configuration example of a differential compression / expansion conversion table shown in FIG. 1;

FIG. 3 is a characteristic diagram illustrating a method of calculating a value of the differential compression / expansion conversion table.

FIG. 4 is an explanatory diagram of another configuration example of the differential compression / expansion conversion table shown in FIG. 1;

5 is an explanatory diagram of reproducibility with respect to a high-frequency input when the table values shown in FIG. 2 are used.

FIG. 6 is an explanatory diagram of reproducibility for a low frequency input when the table values shown in FIG. 2 are used.

7 is an explanatory diagram of reproducibility corresponding to a high-frequency input when the table values shown in FIG. 4 are used.

8 is an explanatory diagram of reproducibility corresponding to a low-frequency input when the table values shown in FIG. 4 are used.

[Explanation of symbols]

 Reference Signs List 11 video processing unit 12 flash memory 13 JPEG / LSI 14 compression / expansion unit 16 frame memory 17 video output unit 18 video input unit 100 digital still camera 141 display / capture controller 142 buffer memory 143 memory transfer controller 144 address bus switching unit 145 read data Latch 146 Output level latch 147 Difference expansion / addition unit 148 Difference compression / expansion conversion table 149 Subtraction / difference compression unit 150 Write data latch 151 Input level latch

Claims (6)

[Claims] In a digital video system including a processing unit for processing video data and a frame memory for storing video data to be processed by the processing unit, the digital video system performs differential compression and differential expansion processing. A conversion table storing a table value to be referred to, and a difference between the video data before being stored in the frame memory and data of one pixel before, and calculating the difference value by referring to the conversion table And subtraction / difference compression means for performing subtraction / differential compression by obtaining a table value corresponding to the above. The output data of the subtraction / differential compression means is stored in the frame memory. A digital video system characterized by the following. 2. The digital video system according to claim 1, further comprising a difference decompression / addition means for decompressing the data compressed by said subtraction / difference compression means with reference to said conversion table. 3. An image input unit for capturing a video signal via an optical system, and a video output unit for monitoring the captured video signal, wherein the data expanded by the differential expansion / addition means is provided. 3. The digital video system according to claim 2, wherein said digital video system is processed by said processing means. 4. The digital video system according to claim 1, wherein the compression processing by said subtraction / difference compression means is performed in raster units of input video data. 5. The digital video system according to claim 1, wherein said conversion table is stored in a rewritable memory. 6. The digital video system according to claim 1, wherein a step width of said conversion table corresponds to a quadratic curve.