JPH05316465A - Digital recorder - Google Patents

Abstract

PURPOSE:To record a video with a few number of bits by dividing one frame into plural blocks, judging each block to be a moving area or a still area and executing optimum encoding at each area. CONSTITUTION:The error detector 81 of a moving area judgement device 43 accumulates the error of the unit of the picture element of inputted two blocks. A threshold value selection device 82 judges the block to be the moving area when the sum of error being the output of the error detector 81 is larger and judges the block to be the still area when the threshold value is larger. Then, concerning the block of the still area, namely an area which is judged to be a background in an actual video, only address information of the block and information of being the still area are needed to be recorded to that large scale compression can be expected in comparison with an original data volume. Besides, also concerning an area which is judged that an object such as a human body, etc., is in it, data obtained by high-efficient-encoding video data within the block is recorded and an information volume is sharply compressed.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a recording device for digitally recording a moving image obtained from a fixed image pickup tube such as a surveillance camera.

[0002]

2. Description of the Related Art Conventionally, an analog recording method using FM modulation has been often used as a means for continuously or intermittently recording images from a surveillance camera. On the other hand, DAT
(Digital Audio Tape Record
When a video signal is recorded on a digital recording medium such as a der tape, it is necessary to A / D convert the video signal and record it as digital data. In this case, for example, it is impossible to record the NTSC digital signal having the information amount of about 200 Mbps as it is from the viewpoint of the recording rate (1.5 Mbps) in the storage type recording medium such as DAT. Therefore, it becomes necessary to compress the digital video signal to 1/100 or less and record it.

As one of the high-efficiency coding methods for video signals, a video conference / telephone coding H.264 standard internationally standardized in CCITT SG XV is used. There is 261. Figure 1
H.3. 261 shows a block circuit diagram of a 261 interframe encoder. In FIG. 13, 21 is an input terminal for digital data, 22 and 23 are adders, 24 is a DCT arithmetic unit, 2
5 is a quantizer, 26 is an inverse quantizer, 27 is an inverse DCT operation device, 28 is a motion compensation variable delay frame memory, 29 is an in-loop filter, 30 is an encoding control device, 31 is an output terminal, and 46 is a multiplex device. It is a means of conversion.

Next, the operation will be described. Input terminal 21
The input image data is input to the adder 22 and the frame memory 28, respectively. The adder 22 generates prediction error data by subtracting the image data of one frame before considering the motion from the input image data. The prediction error data is DCT-transformed by the DCT arithmetic unit 24, further quantized by the quantizer 25, and input to the multiplexing means 46. The quantization characteristic in the quantizer 25 is controlled by the encoding control device 30,
The characteristic is also input to the multiplexing means 46.

The quantized DCT coefficient is converted to the inverse quantizer 2
6, the inverse DCT operation device 27 decodes the adder 2
Input to 3. In the adder 23, the same image data as the output of the decoder is obtained by adding the decoded prediction error and the image data of one frame before which the motion is taken into consideration. This data is input to the motion compensation variable delay frame memory 28. In the motion compensation variable delay frame memory 28, the same image data as the output of the decoder, which is the output of the adder 23, is stored, and the amount of motion is detected by comparing with the image data of the next frame. The image data with the compensation is output. The loop filter 29 is a filter for stabilizing the loop system.

In the above configuration, since the motion-compensated prediction error is utilized by utilizing the redundancy of the image in the time direction, the power of the image can be reduced to compress the information amount. Further, since the local decoder is formed in the encoder, the quantization noise generated on the encoder side is prevented from accumulating on the decoder side.

[0007]

Since the conventional digital recording apparatus is configured as described above, in the case of so-called intermittent recording in which only one frame is recorded per second, the moving distance of the moving body is reduced. If it is large, the search range for motion compensation is exceeded, and the compression efficiency by motion compensation frame prediction is not high. Furthermore, the background image behind the moving object had to be coded as a prediction error each time the moving object moved. In addition, there is a problem that the scale of the motion compensation frame prediction calculation device is large and the cost is high.

[0008]

A digital recording apparatus according to the present invention takes out only a moving area from a huge amount of video data, encodes and records the moving area, and stores a still area once in a frame memory. Only the code that calls the image is recorded.

When reproducing, the data in the frame memory is called for the static area, the recorded data is decoded for the moving area, and the respective data are combined.

The determination of the moving area and the stationary area is
One image frame is divided into several blocks, and the sum of the errors of the corresponding blocks of the input image and the image in the frame memory is used as a reference.

Further, the address of the block itself is transmitted to specify the moving area. Alternatively, the designated code is transmitted while being compressed in a certain scan order. Alternatively, the two means are adaptively switched according to the size of the moving area.

The background data in the frame memory is
It is refreshed at regular intervals or by some control means.

Further, a block regarded as a still area adjacent to a block regarded as a moving area is processed as a moving area.

[0014]

In the digital recording apparatus according to the present invention, the redundant background portion is completely encoded by the moving area detecting means with several bits of the address bits and the judgment bits per block.

[0015]

EXAMPLES Example 1. Hereinafter, a first embodiment of the present invention will be described with reference to the drawings. FIG. 1 is a block circuit diagram of an encoder according to the first embodiment, in which 11 is an image pickup device and 12 is an image pickup device 11.
A / D converter arranged in the latter part of the A / D converter 47a is a first frame memory arranged in the latter part of the A / D converter 12,
48 is a rear stage of the A / D converter 12 and is a refresh means arranged in parallel with the frame memory 47a, 47b is a second frame memory arranged after the refresh means 48, and 42a and 42b are first frame memories respectively. 47a, a first block dividing means and a second block dividing means arranged in the subsequent stage of the second frame memory 47b, and 43 is a first and second block dividing means 42a,
A moving area determination means for receiving the output data of 42b as an input, 45
Is an encoding means arranged in parallel with the moving area determination means 43 in the subsequent stage of the first block dividing means 42a, and 44 is an output of the encoding means 45 and an output of the second block dividing means 42b. Selection means for selecting by the output of the area determination means 43, 41 is a first block division means 42a and a second block division means 42a.
Address control means for controlling the block address of the block dividing means 42b, 46 is a multiplexing means for multiplexing the outputs of the selecting means 44, the moving area judging means 43, the second frame memory 47b and the address controlling means 41, and 13 is, for example, D
A recording means such as an AT.

Next, the operation will be described. Imaging device 11
The video signal output from the second frame memory 4 is filtered, sampled, and quantized by the A / D converter 12 to be converted into digital data, and the second frame memory 4 is passed through the first frame memory 47a and the refresh means 48.
7b is input. The refreshing means 48 is composed of, for example, a timer and a switch, and updates the contents of the second frame memory 47b at certain intervals. The reference data, which is the content of the second frame memory 47b, is input to the multiplexing means 46 every time the content of the second frame memory 47b is updated.

The block dividing means 42a and 42b take out the data of the block at the address controlled by the address control means 41 from the one frame of video data stored in the frame memories 47a and 47b, respectively.
The block size is, for example, 16 pixels × 16 lines.

The moving area judging means 43 is composed of, for example, an error detecting means 81 and a threshold selecting means 82 as shown in FIG. The error detection means 81 accumulates the error in pixel units of the two input blocks. The threshold value selecting means 82 compares the sum of the errors output from the error detecting means 81 with a predetermined threshold value. If the sum of the errors is larger, the block is determined to be a moving area, and the threshold value is set. Is larger, it is determined that the block is a static area.

The encoding means 45 is, for example, a transform encoder such as a discrete cosine transform, a predictive encoder such as a differential PCM, a quantizer, and an entropy encoder such as a Huffman encoder and an arithmetic encoder. The video signal information in one block is efficiently encoded according to the visual characteristics while reducing the information amount.

The selecting means 44 selects the data to be recorded according to the output result of the moving area judging means 43, and the multiplexing means 4
Send to 6. For example, the block determined to be the moving region by the moving region determination unit 43 does not send the output data of the encoding unit 45, and the block determined to be the still region does not send anything.

The multiplexing means 46 multiplexes the outputs of the selection means 44, the moving area determination means 43, the second frame memory 47b and the address control means 41. Of these, the output of the selecting means 44 is multiplexed only when the moving area determining means 43 determines that the moving area is a moving area. Further, the second frame memory 47b is multiplexed by one frame only when the refresh means 48 updates the reference data which is the content of the second frame memory 47b. The data multiplexed by the multiplexing means 46 is input to the digital recording device 13 and recorded on a magnetic tape or the like.

By the above operation, for the block of the still area, that is, the area judged to be the background in the actual video, only the address information of the block and the information indicating the still area need be recorded. Significant compression can be expected compared to the amount of data. In addition, regarding the block of the moving area, that is, the area in which it is determined that an object such as a person is reflected in the actual video, the address information of the block, the information that the area is the moving area, and the video data in the block High efficiency coded data is recorded. This high-efficiency coded data has an information amount of about 1/2 to 1/1 as compared with the original video data by utilizing the visual characteristics.
It can be compressed to about 0.

For example, when the input data is as shown in FIG. 4A and the reference data is as shown in FIG. 4B, the respective data are block dividing means 42a, 42.
5 (a) and 5 (b) respectively, the corresponding blocks are compared with each other, and when the sum of the errors exceeds a certain threshold value, it is regarded as a moving area, Is encoded with. If the sum of the errors is within the threshold value, it is regarded as a static area, and the block copies the reference data. The state of the determination is shown in FIG. In FIG. 6, * represents a block in which the reference data is copied, which is determined to be a still area, and # represents an area which is determined to be a moving area and is encoded.

7 (a) and 7 (b) show an example of a data format coded in the moving area and in the stationary area, respectively. If it is determined to be a moving area in FIG. 7A, the address information of the block is transmitted first, and then the area designation information, the additional information, and the block end code (EOB) are transmitted. Further, when it is determined to be the still area in FIG. 7B, the address information of the block is transmitted first, and then the area designation information is transmitted. Of these, the address information is fixed bit information,
The position of a block in one frame is shown. The area designation information is 1-bit information that designates whether the block indicated by the address information is a moving area or a still area. The additional information is high-efficiency coded video data in the block.

In this way, the subsequent frames are also coded. When recording with this method for a long time, the background changes due to changes in brightness due to sunlight. Therefore, the content of the second frame memory 47b is updated by the refreshing unit 48 every certain fixed time or when the number of blocks in the moving area within a fixed time exceeds a certain value. This is the end of the description of the first embodiment.

Example 2. The second embodiment according to the present invention will be described below. Since the configuration of the encoder of the second embodiment is the same as that of the first embodiment, the description thereof will be omitted. The difference from the first embodiment is that, when the address information, the moving area / still area designating information, and the additional information are multiplexed by the multiplexing means 46, as shown in FIG. , Additional information of the address (high-efficiency coded video data), EOB, and frame end code (EO
F) only is sent, nothing is sent for the static area. In this format, since only EOF needs to be sent as one frame of information for a complete still image, a high compression rate is expected.

Further, as a modification of the second embodiment, as shown in FIG. 9, the area designation information, the additional information only on the blocks of the moving area, and the EOB can be transmitted. in this case,
Since the block scan order is determined in advance and the block information is transmitted in that scan order, address information is not required. This is the end of the description of the second embodiment.

Example 3. The third embodiment of the present invention will be described below. In the third embodiment, the configuration of the encoder is the same as that of the first and second embodiments, and the description thereof will be omitted. The difference from the first embodiment is that in the multiplexing means 46,
When the address information, the moving area / still area specifying information, and the additional information are multiplexed, as shown in FIG. 10, the moving area / still area specifying information for one frame is sent first, and then,
It is about to send additional information and EOB for the blocks in the moving area. In this case, the scan order of the blocks is determined in advance, and the block information is transmitted in the scan order. The moving area of the image in one frame often exists in a certain group (cluster), and by transmitting the moving area / still area specifying information collectively, entropy coding such as run length coding can be performed efficiently. Can be used. This is the end of the description of the third embodiment.

Example 4. The fourth embodiment of the present invention will be described below. In the fourth embodiment, the configuration of the encoder is the same as that of the first to third embodiments, and the description thereof will be omitted. The difference from the first embodiment is that when the moving area determination means 43 selects blocks of a still area and a moving area,
As in 1 (a) or (b), 4 and 4
That is, a block in a still area that is in contact with a connection or an eight connection is regarded as a moving area. In FIGS. 11A and 11B, * is a block in which reference data is copied and is determined to be a still region, # is a block in which input data is encoded and is determined to be a moving region, and * is determined to be a still region. Shows the block that encodes the input data.

According to the fourth embodiment, the moving area determining means 4
By the threshold selection means 82 in No. 3, the slowly moving object is regarded as the stationary area for a while after entering the block which was the stationary area until now, and the movement of the object is jerky. It can be expected that it will not be lost or that the edge part of the animal body is lost. This is the end of the description of the fourth embodiment.

Example 5. Hereinafter, a fifth embodiment according to the present invention will be described with reference to the drawings. FIG. 2 is a block circuit diagram of the decoder of the fifth embodiment, and 61 is the digital recording device 13.
, 62 is a code data detecting means, 64 is an address detecting means, 65 is a reference data detecting means, 63 is a decoding means arranged in the latter stage of the code data detecting means 62, Reference numeral 66 denotes a frame memory arranged at the subsequent stage of the reference data detection means 65,
67 is a block dividing means for generating block data at an address determined by the output of the address detecting means 64 from the output data of the frame memory 66, and 68 is a decoding means 6.
Selecting means 6 for selecting the output of 3 and the block dividing means 67 by the output signal of the moving area designation code detecting means 61;
Reference numeral 9 is a framing means for reconstructing one frame of video data from the blocked data output from the selection means 68 and the output signal from the address detection means 64, and 14 is a D / A converter arranged after the framing means 69. 1
Reference numeral 5 denotes an image receiving device, such as a monitor, which is arranged in the subsequent stage of the D / A converter 14.

Next, the operation will be described. The digital data recorded on the magnetic tape or the like is reproduced by the recording means 13.

The address of the current block of the reproduced digital data is detected by the address detecting means 64.

Further, the moving area designation code detecting means 61 detects whether the block is judged to be a moving area or a still area.

If the moving area designation code detecting means 61 determines that the block is a moving area, the coded data detecting means 62 detects the encoded video data. The coded data detected by the coded data detection means 62 is input to the decoding means 63. The decoding unit 63 has a configuration reverse to that of the encoding unit 45 of the first embodiment, and decompresses the compressed data to restore the original video data (for example, 16 pixels × 16 lines).

Further, when there is one frame of reference data at the beginning of the frame, the reference data detection means 65 detects the reference data,
It is stored in the frame memory 66. Block dividing means 6
7 has the same configuration as the block dividing means 42b of the first embodiment, and a block of an address designated by the output of the address detecting means 64 from the reference data stored in the frame memory 66 (for example, 16 pixels × 16 lines). Take out.

The selecting means 68 selects the output data of the decoding means 63 and the output data of the block dividing means 67 based on the output of the moving area designation code detecting means 61. That is, when the output of the moving area specifying code detecting means 61 specifies the moving area, the data of the output of the decoding means 63 is selected, and the output of the moving area specifying code detecting means 61 specifies the stationary area. At this time, reference data such as the background of the output of the block dividing means 67 is selected.

The data of the block selected by the selection means 68 is input to the framing means 69. The framing means 69 is composed of a frame memory or the like, and stores the data of the block selected by the selection means 68 at the address specified by the address detection means 64, and
The complete video data of the frame is constructed, and further information such as a sync signal is added.

The video signal data created by the framing means 69 is converted into an analog signal by the D / A converter 14 and output to the output device 15 such as a monitor.

Example 6. In the fifth embodiment, the moving area designation code detecting means 61, the code data detecting means 62, the address detecting means 64, and the reference data detecting means 65 are arranged in parallel after the recording means 13 as shown in FIG. It may be arranged in series as shown in FIG. In FIG. 12, each constituent element and operation are the same as those in the fifth embodiment, and therefore description thereof will be omitted. This is the end of the description of the sixth embodiment.

[0041]

According to the present invention, one frame is divided into several blocks, a moving area and a still area are determined for each block, and optimal coding is performed in each area. The effect that video can be recorded with the number of bits is obtained.

[Brief description of drawings]

FIG. 1 is a block circuit diagram showing a configuration of an encoder of a digital recording apparatus that is Embodiment 1 of the present invention.

FIG. 2 is a block circuit diagram showing a configuration of a decoder that is Embodiment 5 of the present invention.

FIG. 3 is a block circuit diagram showing a configuration of a moving area detecting unit according to the first exemplary embodiment.

FIG. 4 is a diagram showing an example of an input image and a reference image thereof for explaining the operation of the first embodiment.

FIG. 5 is a diagram for explaining the operation of the block dividing means according to the first embodiment.

FIG. 6 is a diagram showing a moving area and a still area of the input image of FIG. 4 in the first embodiment.

FIG. 7 is a diagram showing an example of a recording format according to the first embodiment.

FIG. 8 is a diagram showing an example of a recording format according to a second embodiment of the present invention.

FIG. 9 is a diagram showing an example of a recording format of a digital recording device according to a modification of the second embodiment.

FIG. 10 is a diagram showing an example of a recording format according to a third embodiment of the present invention.

FIG. 11 is a diagram showing an example of a moving area and a stationary area according to a fourth embodiment of the present invention.

FIG. 12 is a block circuit diagram showing a configuration of a decoder of a digital recording device according to a sixth embodiment of the present invention.

FIG. 13 is a block circuit diagram showing a configuration of an encoder of a conventional digital recording device.

[Explanation of symbols]

 11 image pickup apparatus 12 A / D converter 13 recording means 14 D / A converter 15 image receiving apparatus 41 address control means 42 block dividing means 43 moving area determining means 44 selecting means 45 encoding means 46 multiplexing means 47 frame memory 48 refreshing means 61 Moving area designation code detecting means 62 Code data detecting means 63 Decoding means 64 Address detecting means 65 Reference data detecting means 66 Frame memory 67 Block dividing means 68 Selecting means 69 Framing means

Claims (6)

[Claims] 1. An imaging device for photoelectrically converting image information into an electric signal, and an A / D for converting the electric signal into a digital signal.
A converter, a first frame memory for storing the digital signal, a second frame memory for storing the digital signal, and an update of the contents of the second frame memory connected to the preceding stage of the second frame memory Refreshing means for controlling the above, and two systems of block dividing means respectively connected to the latter stages of the two systems of frame memories;
Control means for controlling the address of the block dividing means of the system, moving area detecting means for judging whether the block of the address is the moving area or the still area according to the data of the two-system block division, and the first An encoding means connected to a subsequent stage of the block dividing means connected to the frame memory; a selecting means for selecting whether to output the output of the encoding means according to the output result of the moving area detecting means; and the selecting means. The moving area detecting means, the address control means, a multiplexing means for time-division multiplexing the contents of the second frame memory, etc., and a digital recording means for recording the digital signal multiplexed by the multiplexing means. Digital recording device. 2. A digital recording means, a moving area designation code detecting means, a code data detecting means, an address detecting means, and a reference data detecting means, which are arranged in the latter stage of the recording means, and in the latter stage of the code data detecting means. Decoding means provided, a frame memory provided after the reference data detecting means, and block dividing means for generating block data at an address determined by the output of the address detecting means from output data of the frame memory. Selecting means for selecting the output of the block dividing means and the decoding means according to the output signal of the moving area designation code detecting means, the blocked data output from the selecting means and the output of the address detecting means Framing means for reconstructing one frame of video data from a signal,
A digital recording device comprising a D / A converter arranged in the latter stage of the framing means, and an output device arranged in the latter stage of the D / A converter. 3. A means for determining a moving area and a stationary area by comparing a sum of errors of input signal data input to the moving area detecting means and reference data with a certain threshold value. The digital recording device according to claim 1. 4. A multiplexing means for multiplexing moving area designation information and additional data such as output of the encoding means in order of address when multiplexing the output result of the moving area detecting means. 1. The digital recording device according to 1. 5. When multiplexing the output result of the moving area detecting means, there is provided a multiplexing means for multiplexing the address of the block judged as the moving area and the additional data such as the output of the encoding means. The digital recording device according to claim 1. 6. When selecting the output of the encoding means and the reference data according to the output result of the moving area detecting means, a selecting means for selecting a still area in the vicinity of a group of blocks of the moving area as a moving area is provided. The digital recording apparatus according to claim 1, wherein the digital recording apparatus is a digital recording apparatus.