JP2002344947A - Image processing system, compression coding method for moving picture, decoding method for moving picture and their program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide an image processing system that enhances the effect of moving picture compression employing a time series difference in the case of using a frame switcher unit to switch images picked up by a plurality of cameras one after another. SOLUTION: By employing a plurality of memories M0-MF respectively corresponding to a plurality of cameras CAM0-CAMF, a time series difference is generated of each moving picture photographed individually by the cameras CAM0-CAMF separately by the memories M0-MF to apply compression coding to the moving picture while storing a key frame or the like. Similarly in the case of reproducing the moving picture, the moving pictures are separately decoded by the memories M0-MF. The time base correlation is taken between them to increase the compression rate.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

[0001] 1. Field of the Invention [0002] The present invention relates to an image processing system and related technology.

[0002]

2. Description of the Related Art In a video recording system for surveillance use, as shown in FIG. 10, a device called a frame switcher unit 1 controls a plurality of cameras CAM0-CAMF collectively. The frame switcher unit 1 has a function (multiplex function) of sequentially switching input images of a plurality of cameras CAM0 to CAMF for each frame and combining them into one image 2. By using the frame switcher unit 1, it is possible to efficiently record the input video from the plurality of cameras CAM0 to CAMF even with the time-lapse video 3 having only a single video input terminal.

At the time of reproducing the recorded video, the video 4 output from the time-lapse video 3 is passed through the frame switcher unit 1 as shown in FIG.
Output to a predetermined monitor device 5.

Here, the frame switcher unit 1
Displays the video from the time lapse video 3 on the monitor 5
When outputting to multiple cameras CAM0
“Camera number information” for identifying CAMFs from each other 6
Read. The camera number information is a signal added by the frame switcher unit 1 at the time of video recording, and is superimposed on, for example, a video signal during a vertical synchronization period in a non-video signal.

[0005] The frame switcher unit 1 reads the camera number information, and reads the camera number information.
It outputs only the images of AM0 to CAMF to the monitor device 5 and divides and displays the images of the cameras CAM0 to CAMF on one screen.

The frame switcher unit 1 is used for the purpose of recording and reproducing the images of a plurality of cameras CAM0-CAMF with one image processing system.

In FIG. 10, 16 cameras CAM0-CAM0
Although the CAMF is connected to the frame switcher unit 1, the number of cameras is not limited to 16, but can be two, three or more.
(The same applies to FIG. 12 described later).

[0008]

In recent years, image processing systems have been digitized, and in addition to the conventional analog recording system for a video deck, an image processing system of a digital recording system for a hard disk drive (HDD) has been developed. In these digital recording type image processing systems, a still image compression system (JPEG,
Weblet) is often used, but such a still image compression method has a lower compression efficiency than a moving image-based compression method, and when storing data in a hard disk drive,
The disk becomes full (Disk Full) in a short time.

By the way, with the improvement of LSI technology, it has become possible to develop a moving picture-based compression LSI at low cost. In recent years, there has been a movement to adopt a moving picture based compression method as a compression method. In such a moving image based compression method, compression is performed using difference information between adjacent videos.

However, when using the frame switcher unit 1 which is frequently used for monitoring, FIG.
As described above, the images individually captured by a plurality of different cameras CAM0 to CAMF are input to the image compressor 7 while being switched for each of the input fields V0 to VF. Therefore, the image input to the image compressor 7 has no correlation between the adjacent input fields V0 to VF. Therefore,
Even if the images V0 to VF are subjected to moving image compression by the image compressor 7, moving image compression is performed using difference information between images having no correlation on the time axis, and it is difficult to increase the compression ratio.

An object of the present invention is to provide an image processing method that can maximize the effect of moving image compression using a time-series difference when images sequentially captured by a plurality of cameras are sequentially switched using a frame switcher unit. An object of the present invention is to provide a processing system and related technology.

[0012]

In order to solve the above-mentioned problems,
The invention according to claim 1 includes a plurality of cameras, a frame switcher unit that switches and outputs a plurality of moving images individually captured by the plurality of cameras, and a moving image switched by the frame switcher unit. Control means for performing compression encoding using difference information between frames in at least a part thereof, wherein the control means sets a key frame to be referred to when generating the difference information between frames, by using a plurality of cameras. The difference information is stored separately for each of a plurality of memories respectively associated with each other, and the difference information for each key frame in each memory is separately generated in a state where the difference information is associated with each of the plurality of cameras.

According to a second aspect of the present invention, in the image processing system according to the first aspect, camera number information uniquely associated with the camera is stored in a non-video signal of the moving image. The control means separately generates the difference information for each key frame in each memory based on the camera number information in a state where the difference information is associated with a plurality of cameras.

According to a third aspect of the present invention, there is provided the image processing system according to the first or second aspect, wherein the control means is configured to execute the processing based on the difference information for each of the key frames in each of the memories. , Each of which is separately decoded in a state of being associated with a plurality of the cameras.

According to a fourth aspect of the present invention, a plurality of moving images individually picked up by a plurality of cameras are switched by a frame switcher unit and output, and at least a part of the moving images switched by the frame switcher unit is output. In the compression encoding method of a moving image to be compression-encoded using the difference information between frames, a key frame to be referred when generating the difference information between frames,
A first step of storing each of a plurality of memories respectively associated with each of the plurality of cameras, and separately storing the difference information for each key frame in each of the memories in a state where each of the difference information is associated with each of the plurality of cameras. And a second step of generating.

According to a fifth aspect of the present invention, there is provided the moving picture compression encoding method according to the fourth aspect, wherein a camera number uniquely associated with the camera is included in a non-video signal of the moving picture. Information is stored, the first step stores each key frame in each memory based on the camera number information, and the second step calculates the difference information based on the camera number information. , Each of which is separately generated in a state of being associated with a plurality of the cameras.

According to a sixth aspect of the present invention, there is provided a moving picture decoding method for decoding a moving picture compressed and encoded by the moving picture compression encoding method according to the fourth or fifth aspect. And decoding separately in a state associated with each of the plurality of cameras based on the difference information for each of the key frames in each of the memories.

According to a seventh aspect of the present invention, in order to realize the method according to any one of the fourth to sixth aspects on a computer, the computer executes each of the steps.

[0019]

DESCRIPTION OF THE PREFERRED EMBODIMENTS <Outline> FIG. 1 is a block diagram showing an image processing system according to an embodiment of the present invention. As shown in FIG. 1, this image processing system compresses and records a moving image by using difference information along a time axis. In particular, the image processing system individually corresponds to each of a plurality of cameras CAM0 to CAMF. The memories M0 to MF are provided, information necessary for moving image compression is stored in each of the memories M0 to MF, and compression is performed in parallel while obtaining correlation between frames in each of the memories M0 to MF. is there.

In this image processing system, images captured by the plurality of cameras CAM0 to CAMF are input to the compression / expansion device 8 (control means) while the frame switcher unit 1 replaces the images for each of the input fields V0 to VF.

The compression / expansion device 8 reads out the “camera number information” 6 superimposed on the video signal during the vertical synchronization period in the non-video signal, out of the signals input via the frame switcher unit 1, Based on the camera number information, compression / expansion of the moving image is performed for each of the input fields V0 to VF corresponding to the individual cameras CAM0 to CAMF.

At this time, the correlation data (key frame and difference information) for each of the adjacent images having no correlation with each other, which is input from the frame switcher unit 1, is converted into individual images for the number of cameras CAM0 to CAMF. The data is separately stored in the memories M0 to MF. The compression / expansion device 8 performs compression / expansion on the basis of the time series (between frames) and the correlation within the frame using the data stored in the individual memories M0 to MF, and incorporates a hard disk or the like. The image is recorded in the moving image compression digital image recording device 9.

As a result, even when the frame switcher unit 1 is used, the effect of moving image compression can be obtained.

The processing procedure (method) described below is as follows.
The program is governed by a software program stored in advance in a storage device (not shown) in the compression / decompression device 8, and is executed by the operation of a CPU (computer) according to the software program.

<Compression encoding method of moving image> Each camera CA
An example of a compression encoding method for each data that is imaged for each of M0 to CAMF and stored in the associated individual memory M0 to MF will be described. FIG. 2 is a schematic diagram for explaining the compression encoding method, and FIG. 3 is a flowchart for realizing the compression encoding method.

In this compression encoding method, as shown in FIG. 2, a plurality of still images (frames) f1, f1, which are sequentially captured along the time axis in each of the cameras CAM0 to CAMF, respectively.
are input to the compression / decompression device 8 for each of the input fields V0 to VF (step ST1 in FIG. 3).

In this case, the frame switcher unit 1
Is used, as shown in FIG. 1, a plurality of different cameras CA
Each of the images individually captured by the M0-CAMF is replaced with each of the input fields V0-VF while the compression / decompression device 8
Is input to Then, based on the camera number information 6 (see FIG. 10) superimposed on the video signal during the vertical synchronization period in the non-video signal, the compression / decompression device 8 outputs a different camera CA.
Input fields V0 respectively corresponding to M0 to CAMF
, VF, and VF are separately subjected to the compression process according to the processing procedure described below. The key frame f shown in FIG.
1, 1, f2... (Described later) correspond to input fields (reference numeral V0 in the example of FIG. 1) given from the same camera (reference numeral CAM0 in the example of FIG. 1).

Here, in FIG. 3, reference numeral 10A denotes
One of the key frame memories 11A, which is set in each of the memories M0 to MF, is the same as the memory M0.
MF shows one of the key block memories respectively set in MF.

Before the frame f1 is input to the key block memory 11A, a predetermined specific area (key GOB =
A reference frame f0 (not shown) composed of a Group of Block (to be described in detail later) is stored. This reference frame f0 is stored in a key block memory 1 for decryption described later.
After being compressed and transmitted to 1B, it is decoded and stored.

The compression / decompression device 8 used in this compression / encoding method includes input frames f1, f2, f3, f
Key frames are designated periodically from 4,. Here, the key frame refers to a frame serving as a reference image when generating difference information of each input frame of a moving image along a time axis, and is generated such that a key frame is generated periodically. . In FIG.
The input frame f1 is a key frame.

In step ST2 in FIG. 3, it is determined whether the input frame f1 is a key frame.
In FIG. 2, since the frame f1 is a key frame as described above, the process proceeds to step ST3 as it is, and the intra-frame (intra) / inter-frame encoding process is executed.

FIG. 4 shows the processing procedure of step ST3. As shown in FIG. 4, the difference between the pixel value between the reference frame f0 (not shown) and the input frame f1 (key frame) stored in the key block memory 11A in step ST4, and the sum of the absolute values of the difference values (Sum of absolute difference values) S is calculated. In this case, the compression / decompression device 8
Based on the camera number information 6 (see FIG. 10) superimposed on the video signal during the vertical synchronization period in the non-video signal, the key block memories 11 in the corresponding memories M0 to MF, respectively.
A includes a reference frame f0 (not shown) and an input frame f.
1 (key frame) and a sum of absolute values of the difference values (sum of absolute differences) S are stored.

Next, in step ST5 in FIG. 4, it is determined whether the sum of absolute differences stored in the key block memory 11A for each of the memories M0 to MF is equal to or smaller than a threshold. For example, when the difference value is represented by ΔP _i (i: a number corresponding to each pixel), the sum of absolute differences S is given by S = | ΔP
₁ | + | ΔP ₂ | +... + | ΔP _n | (n: the number of pixels). If the sum of absolute differences S is equal to or smaller than the threshold value, it is determined that the temporal change between the two frames f0 and f1 is small, and the input frame f1 is subjected to inter-frame encoding using the frame f0 (step ST6). Specifically, DCT is applied to the difference signal between the input frame f1 and the frame f0.
An orthogonal transform such as (discrete cosine transform) is performed, and a quantized coefficient obtained by quantizing the transform coefficient is calculated. Further, such an inter-frame encoding process is performed for 8 × 8 pixels or 16 × 16 pixels.
It is executed in small block units having a size such as a pixel.
The same applies to the subsequent processing. In this embodiment, orthogonal transform such as DCT is adopted as a transform method.
DWT (discrete wavelet transform) may be employed instead of CT. In this case, instead of performing the inter-frame encoding process in units of the small blocks, the inter-frame encoding process may be performed in units of frames in consideration of the execution memory capacity or the like, or may be performed by dividing the frame into a plurality of regions called tiles and in units of tiles. .

On the other hand, in step ST5 in FIG.
If the difference absolute value sum S calculated in step ST4 exceeds the threshold, the process proceeds to step ST7. Then, the input frame f1 (key frame) is subjected to intra-frame encoding in which only the information in the frame is encoded. Specifically, DCT is applied to the pixel value of frame f1.
Orthogonal transform such as is performed, and the quantized coefficient obtained by quantizing the transform coefficient is calculated.

In practice, from the viewpoint of increasing the compression ratio of the frame, the input frame is subjected to color space conversion before the above-mentioned intra-frame coding (step ST7) or inter-frame coding (step ST6). Is applied. For example, if the original signal is composed of RGB spaces of “R (red component)”, “G (green component)”, and “B (blue component)”, this is referred to as NTSC (National Television System Com.
mittee) YUV coordinate system, YI adopted in the method
Q coordinate system, it may be used and YC _b C _r coordinate system. For example, YC _b C _r When using the coordinate system, YC made from the RGB component luminance signal Y and two color difference signals C _b, and C _{r b}
It is transformed into the coordinate system of the _Cr component. YC _b C _r component is RG
Since the correlation between the components is smaller than that of the B component, the image size can be compressed.

After execution of the intra-frame / inter-frame encoding process (step ST3, ie, the process of FIG. 4), the process proceeds to step ST19 of FIG. At this time, the quantized coefficients calculated in steps ST6 and ST7 (FIG. 4) are subjected to entropy coding including Huffman coding and the like, and thereafter, image information such as the image size of the frame and the number of quantization bits, and the like. Multiplexed together with compression information such as an encoding table and an encoding method (intra-frame encoding, inter-frame encoding) of each small block area, and output as a bit stream. The quantized coefficients calculated in steps ST6 and ST7 (FIG. 4) are locally decoded (inverse orthogonal transform such as inverse quantization and inverse DCT) in step ST8 in FIG. Is stored in the key frame memory 10A. Here, the key frame changed by including the quantization error through the encoding (ST6, ST7) and the local decoding (ST8) is stored in the key frame memory 10A. In this case, the compression / decompression device 8 outputs the camera number information 6 superimposed on the video signal during the vertical synchronization period in the non-video signal.
(See FIG. 10), the corresponding memory M
The key frames changed including the quantization error are stored in the key frame memory 10A in the 0 to MF. As a result, the image of the key frame becomes the same as the image of the key frame referred to when decoding (inter-frame decoding) as described later, and the image quality of the moving image to be decoded does not deteriorate. . Thus, the compression encoding process for the input frame f1 (key frame) is completed.

Next, as shown in FIG.
Subsequent to 1, the frame f2 is input to the compression / decompression device 8. Then, in step ST2 in FIG. 3, it is conditionally determined whether or not the frame f2 is a key frame. In FIG. 2, frame f2 is not a key frame. Therefore, the process proceeds to step ST9 in FIG.

In step ST9, the frame f2 is divided into a plurality of block areas (hereinafter referred to as "GOB").

Here, the GOB is, for example, divided into 16 or 32 in one frame image.
This means a specific area as a division unit for dividing in the vertical direction by the number of powers of “2”.

Next, in step ST10, one or more specific areas ("key G
OB ”).

FIG. 6A schematically shows a frame f2 divided into four GOBs. The frame f2 is divided in the vertical direction into four GOBs in units of several tens of pixels to several tens of pixels, and the first-stage GOB is designated as a key GOB (“key GOB1”). 2B to 2D.
As shown in the figure, the frames f3 to f5 sequentially input to the compression / decompression device 8 following the frame f2 are also divided into a plurality of GOBs. However, as each key GOB, the second-stage GOB of frame f3 (“key GOB
2 ”), the third-stage GOB of frame f4 (“ key GOB
3 "), the fourth GOB of frame f5 (" key GOB
4 ”) is specified. Each of these frames f2 to f5 including the key frame f1 and the keys GOB1 to GOB4
Are arranged along the time axis as shown in FIG.

The description returns to the frame f2. At this point, the process proceeds to step ST11 in FIG.
In frame f2, GOB is further divided into 8 × 8 pixels or 16 × 1 pixels.
Processing is performed sequentially for each small block divided into basic processing units of about six pixels. However, this step ST1
In 1, the condition is determined as to whether the block to be processed belongs to the key GOB. If the block belongs to the key GOB, the process proceeds to step ST12. Then, after the block is subjected to the intra-frame encoding, the block is entropy-encoded in step ST19, multiplexed with the image information and the compression information, and output as a bit stream. The quantized coefficients output by intra-frame encoding of the block in step ST12 are subjected to local decoding (inverse orthogonal transformation such as inverse quantization and inverse DCT) in step ST13, and thereafter are performed in the key block memory 11A.
Is accumulated in In this case, based on the camera number information 6 (see FIG. 10) superimposed on the video signal during the vertical synchronization period in the non-video signal, the compression / decompression device 8 stores the key block memories in the corresponding memories M0 to MF. The quantization coefficient is stored in 11A.

On the other hand, if the block does not belong to the key GOB in step ST11, the process proceeds to step ST14.
The processing shifts to the intra-frame / inter-frame encoding processing. Here, FIG. 5 is a flowchart showing a processing procedure of a subroutine of intra-frame / inter-frame encoding processing. As shown in FIG. 5, first, in step ST15, the corresponding block of the input frame and the key frame memory 10A
Is calculated, and a difference absolute value sum S is calculated. In this case, based on the camera number information 6 (see FIG. 10) superimposed on the video signal during the vertical synchronization period in the non-video signal, the compression / decompression device 8 stores the key frame memories in the corresponding memories M0 to MF. For each 10A, a difference value and a sum S of difference absolute values are calculated.

Next, in step ST16, a condition judgment is made as to whether or not the sum of absolute differences S is equal to or smaller than a threshold. If the sum of absolute differences S is equal to or smaller than the threshold, the process proceeds to step ST17. The camera number information 6 (FIG. 10) superimposed on the video signal during the vertical synchronization period in the non-video signal
Memory M0 to MF, respectively, based on
The inter-frame coding is performed with reference to the individual key frames stored in the key frame memory 10A. On the other hand, if the sum of absolute differences S exceeds the threshold, the process proceeds to step ST18, and the block is subjected to the intra-frame encoding. Thus, step ST
The quantized coefficients encoded in steps 17 and 18 are subjected to variable-length encoding (entropy encoding) and the multiplexing process in step ST19 shown in FIG. 3, and output as a bit stream. Thus, the compression encoding process for the input frame f2 is completed.

Next, in FIG. 2, frames f3, f4 and f4 input to the compression / decompression device 8 subsequent to the frame f2.
Are processed in the same manner as in the case of the frame f2 until a key frame is input. However, as described above, the key GOB (key G) in each frame f2, f3, f4,.
The positions of OB1 to key GOB4) are different from each other.
In this manner, the keys GOB1 to GOB4 locally decrypted in step ST13 are accumulated for one frame in the key block memory 11A, and as schematically shown in FIG. 2, the keys GOB1 to GOB4 are: The reference frame A is synthesized in the key block memory 11A. The reference frame A is used when a key frame to be input later is inter-frame encoded in the subroutine of step ST3. In this case, the compression / decompression device 8
Is based on the camera number information 6 (see FIG. 10) superimposed on the video signal during the vertical synchronization period in the non-video signal, and refers to the reference frame A for each key block memory 11A in the corresponding memories M0 to MF. Are synthesized.

As described above, each of the cameras CAM0-CAMF
Are processed separately using a plurality of memories M0 to MF respectively associated with the moving images, so that the input fields V0 to VF are compressed using the frame switcher unit 1. Even in the case where the video data is converted, the moving image can be compressed using the difference information between frames in the moving image captured for each camera.

The intra-frame coding and the inter-frame coding are selectively executed according to the difference between the reference frame stored in the key block memory 11A in step ST3 and the reference frame stored in the key block memory 11A. An input frame is divided into a plurality of GOBs, a key GOB is designated, and a key GOB for one frame is dispersed in a plurality of frames along a time axis, and each of these keys GOB is intra-coded. As a result, the intra-frame encoding processing amount is temporally dispersed, and a sudden increase in the compression encoding processing amount is suppressed, and the encoding processing amount is temporally flattened. Can be compressed and transmitted without changing the quality. In particular, the present invention is effective for transmission paths with limited bandwidth such as the Internet.

In the key block memory 11A, the keys GOB dispersed in a plurality of frames are stored, and a reference frame A including the keys GOB is formed. This reference frame A is an aggregate of key GOBs at different times. In this embodiment, intra-frame coding and inter-frame coding are selectively executed according to the difference between the reference frame A and the key frame. For this reason, the difference in image quality between GOBs caused by using the reference frame A including the key GOBs at different times is reduced, and high-quality moving images can be compressed and transmitted.

<Decoding Method for Reproducing Moving Image> Next, a decoding method for reproducing a moving image will be described in detail. FIG. 7 is a schematic diagram for explaining the decoding method, and FIG. 8 is a flowchart for realizing the decoding method. In FIG. 8 and the like, reference numeral 10B denotes one of key frame memories set in each of the memories M0 to MF, and reference numeral 11B denotes
Similarly, one of the key block memories set in each of the memories M0 to MF is shown.

This decoding method also uses different cameras CAM0-CAMF in the compression / decompression device 8 based on the camera number information 6 (see FIG. 10) superimposed on the video signal during the vertical synchronization period in the non-video signal. The decompression process is performed separately according to the processing procedure described below for each of the input fields V0 to VF corresponding to.

First, the compressed image data encoded by the above-described compression encoding method is input to the compression / decompression device 8 as a bit stream (step ST20 in FIG. 8). After being separated from the bit stream, the compressed image data is decoded in step ST21. That is, the compression data of the frames f1, f2,... Shown in FIG. At this time, in step ST21 in FIG. 8, the compressed data of the key frame f1 is compared with steps ST4 to ST4 in FIG.
Similarly to the processing procedure shown in ST7, decoding processing of intra-frame coding or inter-frame coding is performed in small block units of about 8 × 8 pixels or 16 × 16 pixels.
Here, when decoding the compressed data of the key frame f1 in FIG. 7, the reference frame f0 (not shown) stored in the key block memory 11B (FIG. 8) in each of the memories M0 to MF in advance is used. Key frame f decrypted here
1 is stored in the key frame memory 10B (FIG. 8) in the corresponding memories M0 to MF based on the camera number information 6 (see FIG. 10) superimposed on the video signal during the vertical synchronization period in the non-video signal. Stored separately.

In FIG. 7, the compressed data of the frames f2, f3,... Input to the compression / decompression device 8 following the compressed data of the key frame f1 are compared with the frames in steps ST12, ST14 to ST18 in FIG. The same processing as the decoding processing of the inner coding or the inter-frame coding is performed for each small block. When the decoding process of the inter-frame encoding is performed, the individual key frames f1 stored in the key frame memories 10B of the memories M0 to MF are separately used. Also, frames f2, f3,
Are decoded, if the small block that is the basic processing unit belongs to the key GOB, the small block is individually stored in each key block memory 11B stored in the key frame memory 10B of each of the memories M0 to MF. Stored. When keys GOB for one frame are accumulated, a reference frame A (see FIG. 2) composed of these keys GOB is synthesized. The synthesized reference frame A is sent to the compression / decompression device 8
This is used when decoding the compressed data of the key frame input to. For example, when the compressed data of the frames f2 to f5 shown in FIGS. 6A to 6D are input to the compression / expansion device 8, the compressed data of the blocks constituting each key GOB is subjected to intra-frame decoding. After that, based on the camera number information 6 (see FIG. 10) superimposed on the video signal during the vertical synchronization period in the non-video signal, it is sequentially and individually stored in the key block memories 11B in the corresponding memories M0 to MF. Then, the reference frame A is reconstructed.

By the way, as described above, step ST21
When the frames f1, f2,... Decoded in the above are displayed as a moving image as they are, the difference in the image quality of the moving image between the GOB coded in the frame and the GOB coded in the interframe by the compression / decompression device 8 is determined. In some cases, the key GOB, which is easy to appear and is particularly coded within a frame, can be clearly seen in a moving image. Considering this, in this embodiment, FIG.
In step ST22 shown in FIG.
A key GOB requantization process for decoding after decoding only the key GOB decoded in step 1 is executed.

FIG. 9 is a flowchart showing the procedure of the subroutine of the key GOB requantization process in step ST22. As shown in FIG. 9, first, a small block of about 8 × 8 pixels or 16 × 16 pixels is input (step ST30). Next, in step ST31, a condition determination is made as to whether or not the block belongs to the key GOB.

If the block does not belong to the key GOB, the block is not requantized and the key GOB is not requantized.
The B requantization process is completed, and step ST23 shown in FIG.
The processing shifts to.

On the other hand, if the block belongs to the key GOB, the process proceeds to step ST32, where the difference between the pixel value of the key frame stored in the key frame memory 10B and the block and the difference of the difference value are determined. Sum of absolute values S
Is calculated. In this case, based on the camera number information 6 (see FIG. 10) superimposed on the video signal during the vertical synchronization period in the non-video signal, the compression / decompression device 8 stores the key frame memories in the corresponding memories M0 to MF. For each 10B, a difference value and a sum S of difference absolute values are calculated.

Next, in step ST33, a condition judgment is made as to whether or not the difference absolute value sum S is equal to or smaller than a threshold value. If the difference absolute value sum S exceeds the threshold value, the block is not requantized and the key GOB The quantization process ends, and the process proceeds to step ST23 shown in FIG.

On the other hand, if it is determined in step ST33 that the sum of absolute differences S is smaller than the threshold value,
The process shifts after T34. First, in step ST34, a difference signal between the block and the key frame is transform-coded, and then, in step ST35, the transform coefficient is quantized. The processing of these steps ST33 to ST35 includes the determination processing (ST16) of the encoding method (interframe encoding, intraframe encoding) based on the sum of absolute differences S performed by the compression / decompression device 8, and the orthogonal transformation such as DCT. And the quantization process (ST17).

Thereafter, in step ST36, the quantized coefficient is inversely quantized, and then in step ST37, decoding of the transform coding (inverse orthogonal transform such as inverse DCT) in step ST34 is executed. As a result, step ST34 is performed.
ＳＴST37, the compression / expansion device 8 sets the key G
An irreversible difference signal including a quantization error is obtained in the same manner as when the encoded signal is decoded by the compression / expansion device 8 after the block area other than the OB is inter-frame encoded.

Next, in step ST38, each memory M0
Each block is reconstructed and output from the difference signal using the key frame individually stored in the key frame memory 10B for each of the .about.MF.

The block subjected to the key GOB requantization processing is output after being synthesized into a frame (decoded image) in step ST23 shown in FIG.

The above key GOB requantization process will be described with reference to frames f2 to f5 shown in FIG. 6 as an example.
As schematically shown in FIG. 3, the keys GOB of the frames f2 to f5 decrypted in the above step ST21 are stored in each memory M
The difference from the key frame stored in the key frame memory 10B for each of 0 to MF is obtained.

Next, in step ST32, it is determined whether or not the sum of absolute differences S of the difference values is equal to or smaller than a threshold value, that is, whether or not to perform inter-frame coding. Is subjected to inter-frame encoding (transformation encoding and quantization) for the key GOB, and then to decoding (inverse quantization and inverse transform decoding) of the inter-frame encoding, thereby obtaining the above-mentioned frame. Decoded images F1 to F5 corresponding to f2 to f5 are generated.

In this manner, each of the cameras CAM0-CAM
In the same procedure as in the compression encoding process (ST15 to ST17), if the difference between the key GOB and the key frame is small, the moving image captured by the MF and the compression encoding is performed. After performing compression encoding on the differential signal of
Is reconstructed, the inter-frame encoding is performed on the key GOB in the same manner as the case where the compression / expansion device 8 inter-frame encodes the block area other than the key GOB and then decodes the encoded signal using the compression / expansion device 8. And errors associated with the decoding. Therefore, there is an effect that the key GOB does not stand out in the moving image when displaying the decoded moving image, and the viewer of the moving image does not feel uncomfortable.

Since the moving images are separately decompressed using the plurality of memories M0 to MF respectively associated with the cameras CAM0 to CAMF, the input fields V0 to VF are compressed using the frame switcher unit 1. The encoded moving image can be easily decoded (decompressed).

It should be noted that the compression encoding and decoding methods can be applied to any compression encoding method other than the above, for example, a method using a difference on the time axis such as MPEG. Absent.

Further, in the above embodiment, the moving image which is switched by the frame switcher unit 1 for each of the input fields V0 to VF and further compression-encoded by the compression / decompression device 8 is always transmitted to the moving image compression digital image recording device 9. Although the recording is performed, the present invention may be applied to a communication system for transmitting a moving image compressed and coded to another station using a predetermined communication line, for example.

Further, in the embodiment described with reference to FIG.
Although 16 cameras CAM0-CAMF were connected to the frame switcher unit 1, the number of dice of this camera is determined by the memory size for storing the reference data, and therefore the number of cameras is limited to 16 Instead, two, three or more may be used.

[0069]

According to the first, fourth and seventh aspects of the present invention, a plurality of moving images individually picked up by a plurality of cameras are switched by a frame switcher unit and output, and the frame switcher unit is provided. When compressing and encoding the moving image switched in at least partly using the difference information between frames, a key frame to be referred to when generating difference information between frames is separately set for each of a plurality of cameras. A frame switcher unit is used because it is stored for each of a plurality of associated memories, and difference information for each key frame in each memory is separately generated in a state of being associated with each of a plurality of cameras. Even if the input field is compressed and coded, the difference information between frames in the moving image captured by each camera is It is possible to perform moving picture compression using.

According to the second, fifth, and seventh aspects of the present invention, camera number information uniquely associated with a camera is stored in a non-video signal of a moving image. Each key frame is stored in each memory based on the number information, and difference information is separately generated based on the camera number information in a state of being associated with each of a plurality of cameras. Each moving image can be individually compression-encoded in the attached state.

According to the third, sixth and seventh aspects of the present invention, based on the difference information for each key frame in each memory, decoding is performed separately in a state where each key frame is associated with a plurality of cameras. Therefore, each moving image can be individually decoded in a state where it is associated with each camera.

[Brief description of the drawings]

FIG. 1 is a block diagram showing an image processing system according to an embodiment of the present invention.

FIG. 2 is a schematic diagram for explaining a compression encoding method according to an embodiment of the present invention.

FIG. 3 is a flowchart for realizing a compression encoding method according to an embodiment of the present invention.

FIG. 4 is a flowchart for realizing a compression encoding method according to an embodiment of the present invention.

FIG. 5 is a flowchart for implementing a compression encoding method according to an embodiment of the present invention.

FIG. 6 is an explanatory diagram showing each frame divided into four block areas.

FIG. 7 is a schematic diagram for explaining a decoding method according to an embodiment of the present invention.

FIG. 8 is a flowchart for implementing a decoding method according to an embodiment of the present invention.

FIG. 9 is a flowchart illustrating a requantization process.

FIG. 10 is a schematic diagram showing an image recording operation of a conventional image processing system.

FIG. 11 is a schematic diagram showing an image reproducing operation of a conventional image processing system.

FIG. 12 is a diagram illustrating an operation in which a moving image from a conventional image processing system is switched by a frame switcher unit.

[Explanation of symbols]

 1 Frame switcher unit 6 Camera number information 8 Compression / expansion device 9 Video compression digital image recording device CAM0-CAMF camera M0-MF memory

 ──────────────────────────────────────────────────続 き Continued on front page (72) Inventor Hirohito Mase 4-6-1, Miyahara, Yodogawa-ku, Osaka-shi Inside Mega Chips Co., Ltd. (72) Hidekazu Tanaka 4-6-1, Miyahara, Yodogawa-ku, Osaka-shi Stock Company 5C054 CH08 EG01 GA01 GB01 HA18 5C059 MA04 MA05 MA23 MA24 MC11 ME02 PP15 PP16 SS14 UA02 UA05 UA33 UA35 UA39

Claims (7)

[Claims] 1. A camera comprising: a plurality of cameras; a frame switcher unit for switching and outputting a plurality of moving images individually captured by the plurality of cameras; and a moving image switched by the frame switcher unit at least in part. Control means for performing compression encoding using difference information between frames, wherein the control means associates a key frame to be referred to when generating the difference information between frames with each of the plurality of cameras. Stored for each of the multiple memories
An image processing system, wherein the difference information for each key frame in each memory is separately generated in a state where the difference information is associated with each of the plurality of cameras. 2. The image processing system according to claim 1, wherein camera number information uniquely associated with the camera is stored in a non-video signal of the moving image. An image processing system, wherein the difference information is separately generated for each key frame in each memory based on camera number information in a state where the difference information is associated with each of the plurality of cameras. 3. The image processing system according to claim 1, wherein the control unit controls each of the plurality of cameras based on the difference information for each of the key frames in each of the memories. An image processing system wherein decoding is performed separately in an associated state. 4. A plurality of moving images individually captured by a plurality of cameras are switched by a frame switcher unit and output, and the moving images switched by the frame switcher unit are used as at least a part of difference information between frames. A compression encoding method of a moving image to be compression-encoded using a key frame to be referred to when generating difference information between frames, for each of a plurality of memories respectively associated with the plurality of cameras. And a second step of separately generating the difference information for each key frame in each of the memories in a state in which the difference information is associated with each of the plurality of cameras. Method. 5. The compression encoding method for a moving image according to claim 4, wherein camera number information uniquely associated with the camera is stored in a non-video signal of the moving image, A first step of storing each key frame in each memory based on the camera number information; and a second step of transmitting the difference information to the plurality of cameras based on the camera number information. A compression encoding method for a moving image, which is separately generated in an associated state. 6. A moving picture decoding method for decoding a moving picture that has been compression-encoded by the moving picture compression-encoding method according to claim 4 or 5, wherein: A decoding method for a moving image, wherein decoding is performed separately in a state in which each key frame is associated with a plurality of cameras based on the difference information. 7. A program for causing a computer to execute each of the steps in order to realize the method according to claim 4 on a computer.