JP4554114B2 - Moving image recording system and moving image recording method - Google Patents

Description

[0001]
BACKGROUND OF THE INVENTION
The present invention relates to a moving image recording system for recording moving image data transmitted via a network, and more particularly to a moving image recording system suitable for a moving image having a small temporal change such as a monitoring video.
[0002]
[Prior art]
2. Description of the Related Art In recent years, digital video systems that handle digital video have become widespread as monitoring systems that transmit video data obtained by imaging a monitoring point, instead of analog monitoring systems that handle analog video. The advantages of the digital surveillance system are that it can perform digital image processing such as contrast adjustment, contrast adjustment, enlargement / reduction, etc., and individual recording conditions for each imaging camera placed at multiple surveillance points. For example, it can be easily set, and a large-scale system can be easily constructed on a computer network such as a LAN. Therefore, the digital monitoring system can collectively monitor a plurality of points, and can greatly contribute to reduction of management cost and monitoring cost. On the other hand, in analog video systems, analog signals are exchanged between devices, and there are many obstacles in terms of information transmission to remote locations, so CCTV (Closed Circuit Television) that is locally closed for monitoring inside buildings I had to become a system.
[0003]
In a digital surveillance system, moving image data obtained by imaging a monitoring point is usually compressed and encoded by a moving picture experts group (MPEG) method or a motion JPEG method based on the JPEG (Joint Photogrphic Experts Group) method. After being transmitted. In the MPEG system and the like, intra-frame coding that compresses and encodes only information in a frame (still image) to reduce redundancy in the spatial direction, and temporally the frame to reduce redundancy in the time axis direction. The moving image data is compressed by a combination of motion compensated prediction encoding that compresses and encodes difference information from a prediction frame predicted from an adjacent frame, and a high compression rate can be realized. On the other hand, in the motion JPEG method or the like, motion compensation predictive coding is not performed, and individual frames are compressed by intraframe coding. For this reason, the motion JPEG method or the like is disadvantageous in that the compression rate is lower than that of the MPEG method or the like, but is advantageous in that the processing load at the time of frame decimation, reverse playback, and processing editing is reduced.
[0004]
[Problems to be solved by the invention]
As the digital monitoring system as described above becomes larger, the required capacity of recording media for recording and storing video information tends to increase. As a recording medium, a hard disk is generally used in terms of recording speed, price, recording density and the like. However, its capacity is limited. In addition, for example, when monitoring stores that are spread over a wide area at a large number of locations, the recording cost of a huge amount of video data becomes enormous as the number of monitoring locations increases, and the amount of data transmission is predicted. Due to the difficulty, the management burden of recording media has increased.
[0005]
In view of the above circumstances, the present invention is to provide a moving image recording system capable of efficiently recording a monitoring video transmitted through a network.
[0006]
[Means for Solving the Problems]
In order to solve the above problem, the invention according to claim 1 is a moving image recording system comprising a monitoring server and an image recording apparatus for recording moving image data transmitted from the monitoring server via a network, The monitoring server generates and outputs compressed data obtained by compressing and encoding input moving image data; a transmission processing unit that converts the compressed data into transmission data and sends the transmission data to the image recording device; The image recording apparatus includes: a reception processing unit that receives the transmission data from the monitoring server and reconstructs the compressed data; an image storage unit that stores the compressed data; and a memory of the image storage unit Monitoring means for monitoring the free space and generating a warning signal when the free space is reduced to a predetermined amount; and storing in the image storage unit when the warning signal is received from the monitoring means. And having a memory management unit for controlling such thinned out in frame units in chronological order of the compressed data recording time.
[0007]
Then, in the invention according to claim 1, wherein the compression encoding unit of the monitoring server includes a reference frame obtained by the frame of the moving image data to be input periodically designated, and the moving image data and the reference frame and generating a difference frame composed of difference signals, that generates outputs compressed data compressed and encoded and the reference frame and the difference frame.
[0008]
Further, a plurality in the invention according to claim 1, wherein the memory management unit manages the frame group composed of said difference frame generated from the reference frame and the reference frame as a unit, and, which is set variably For each frame group in the compressed data stored in the image storage unit when receiving the warning signal from the monitoring means, the recording time of the frame group The reverse thinning rate is assigned so that the number of thinned out difference frames belonging to the frame group increases as the number of old frames increases.
[0009]
The invention according to claim 2 is a moving image recording method for recording moving image data transmitted from a single or a plurality of monitoring points via a network, wherein (a) compressed data obtained by compressing and encoding moving image data is recorded. Generating, converting to transmission data and sending it to the network; (b) receiving the transmission data sent in step (a) and reconstructing the compressed data; and (c) the compressed data. (D) monitoring the free space of the memory of the image storage unit and issuing a warning signal when the free space is reduced to a predetermined amount; (e) the step ( wherein when the warning signal is issued in d), and controlling the compressed data stored in the image storage unit in the step (c) to thin out the oldest recording time in units of frames, Ru with a It is also of the
[0010]
The invention according to claim 2, before Symbol step (a), the said compressed data includes a reference frame obtained by the frame of the moving image data by periodically specified, the moving image data and the reference frame Is generated by compressing and encoding the reference frame and the difference frame.
[0011]
Further, in the invention according to claim 2, wherein step (e) includes the steps of managing as a unit a frame group constituted by the said difference frame generated from (e1) said reference frame and said reference frame, (E2) a step of preparing a plurality of inverse thinning rates that are variably set; and (e3) when the warning signal is issued in step (d), for each frame group in the compressed data And the step of assigning the reverse thinning rate so that the number of thinned out differential frames belonging to the frame group becomes larger as the recording time of the frame group becomes older.
[0012]
DETAILED DESCRIPTION OF THE INVENTION
FIG. 1 is a functional block diagram showing a schematic configuration of a moving image recording system according to an embodiment of the present invention. The moving picture recording system, a plurality of monitoring points 3 _1, 3 _2, ..., 3 captured disposed _n camera 2 _1, 2 _2, ..., from 2 _n and the monitoring server 1 captures an image signal, IP (Internet Protocol) It is composed of a network NW1 and an image recording apparatus 4 that can be interconnected with the monitoring server 1 via the IP network NW1. The imaging cameras 2 ₁ ,..., 2 _n are general ones equipped with a digital imaging device such as a CCD or a CMOS.
[0013]
The monitoring server 1 compresses and encodes an image input / output unit 10 that digitizes and outputs image signals input from the imaging cameras 2 ₁ to 2 _n and digital moving image data input from the image input / output unit 10. And a compression encoding unit 11 that outputs compressed data and a transmission processing unit 12 that converts the compressed data into a packet (datagram) according to the communication protocol of the IP network NW1 and transmits the packet. On the other hand, the image recording device 4 receives a packet sent from the monitoring server 1 and reconstructs the original compressed data, an image storage unit 14 for storing the compressed data, and this image storage. And a memory capacity control unit 15 that performs memory management of the unit 14. Note that the compression encoding unit 11 of the monitoring server 1 and the memory capacity control unit 15 of the image recording device 4 may be configured by either hardware or software.
[0014]
The image output unit 10 of the monitoring server 1, a plurality of monitoring points 3 _1, 3 _2, ..., 3 captured _n installed in the camera 2 _1, 2 _2, ..., a plurality of input is transmitted from the 2 _n The image signal is selectively captured, and the captured image signal is subjected to analog image processing such as gain adjustment or A / D conversion to generate digital moving image data, which is output to the compression encoding unit 11. The digital moving image data is composed of a plurality of time-series still images (frames).
[0015]
The compression encoding unit 11 performs compression encoding processing using DCT (Discrete Cosine Transform), DWT (Discrete Wavelet Transform), etc., on the frames sequentially input from the image input / output unit 10. To generate compressed data. Hereinafter, the procedure of the compression encoding process will be described with reference to FIGS. As shown in FIG. 2, the compression encoding unit 11 includes time-series frames f ₀ , f ₁ , f ₂ ,..., F _k , f _{k + 1} , f _m , f _{m + 1} along the elapsed time t. ,..., F _n , f _{n + 1} ,... Are sequentially input (where k, m, and n are integers satisfying the magnitude relationship of n>m> k).
[0016]
The compression encoding unit 11 periodically designates reference frames f ₀ , f _k , f _m , f _n ,... From sequentially input frames f _{0 ,.} Specifically, an input frame may be automatically designated from the input frames f ₀ , f ₁ ,... At intervals of several seconds, for example, and when 15 frames are input per second, for example, every 15 frames or 30 A reference frame may be set for each frame. It is also possible to automatically designate the input frame at the time when the information amount difference between the input frames exceeds the threshold as the reference frame.
[0017]
Next, each reference frame f ₀ , f _k , f _m , f _n ,... Is subjected to intra-frame coding for compressing and coding only the information in the frame, and then output to the transmission processing unit 12. The The processing contents of the intra-frame encoding are as follows. The reference frame is subjected to preprocessing such as color space conversion from RGB signals to YCbCr signals, thinning processing, and the like, and then converted into conversion coefficients by DCT, DWT, or the like. When DCT is performed, the reference frame is divided into 8 × 8 pixel region or 16 × 16 pixel region macroblocks at the pre-processing stage, and DCT is performed for each macroblock unit. On the other hand, when DWT is performed, the execution memory capacity is taken into consideration in units of frames, or the frame is divided into a plurality of areas called tiles and executed in units of tiles. Next, transform coefficients generated by DCT, DWT, etc. are quantized with a predetermined number of quantization bits and converted to quantized coefficients, and subjected to entropy coding such as Huffman coding to be converted to coded data. The A bit stream is generated by multiplexing the encoded data and additional information such as the resolution information of the reference frame, which is the original signal, the number of quantization bits, and the Huffman table, and is output to the transmission processing unit 12. .
[0018]
Further, the compression encoding unit 11 inputs the input frames f ₁ , f ₂ ,..., F _k−1 , f _{k + 1} , f _m− between the reference frames f ₀ , f _k , f _m , f _{n ,. 1, f m + 1, ...} , f n-1, f n + 1, ... to generate a difference frame composed of difference signal between the input frame and the reference frame, inter-frame coding for compressing and encoding this Execute the conversion process. FIG. 3 is a schematic diagram showing reference frames f ₀ , f _k , f _m , f _n ,... And difference frames df ₁ , df ₂ ,. In the figure, in the frame group G ₀ , the difference frames df _{1 to} df _k-1 are generated based on the reference frame f ₀ , and in the frame group G _k , the difference frames df _{k + 1 to} df _m-1 are the reference frames. is generated based on f _k, the difference frame df m _{+ 1} ~df _n-1 in the frame group G _m is one that is generated based on the reference frame f _m. The compression encoding unit 11 compresses and encodes these difference frames in the same manner as the above intra-frame encoding, and outputs the result to the transmission processing unit 12.
[0019]
The compressed data of the reference frame and the difference frame generated by the above procedure is transmitted to a communication protocol such as TCP / IP (Transmission Control Protocol / Internet Protocol) or UDP / IP (User Datagram Protocol / Internet Protocol) in the transmission processing unit 12. And is sent to the image recording apparatus 4 via the IP network NW1. Note that the transmission order of the difference frames does not necessarily follow the time series of the input frames. For example, it is possible to perform an optimization process in which the order of the difference frame df ₁ and the difference frame df ₂ belonging to the same frame group G ₀ is switched and compressed, and the order is returned to the original order by the image recording device 4 on the receiving side. .
[0020]
Next, each part of the image recording apparatus 4 will be described. When receiving the packet transmitted from the monitoring server 1, the reception processing unit 13 of the image recording device 4 reconstructs the compressed data of the reference frame and the difference frame from the received packet and outputs the reconstructed data to the image storage unit 14. The image storage unit 14 sequentially stores the compressed data. Here, the monitoring servers 1, 1,... Having the same configuration as the monitoring server 1 are distributed on the IP network NW1, and the image recording apparatus 4 is a moving image sent from a large number of monitoring servers 1, 1,. Data can be received and stored in the image storage unit 14.
[0021]
The memory capacity control unit 15 constantly monitors the free space (remaining amount of memory capacity) of the memory of the image storage unit 14, and when the free space falls below a predetermined threshold, the compression stored in the image storage unit 14 It is a functional block for executing a thinning process on data according to a procedure described later. The memory capacity control unit 15 includes monitoring means 16 and memory management means 17. Note that the memory capacity control unit 15 may be configured by either hardware or software.
[0022]
The monitoring unit 16 monitors the free space in the image storage unit 14 and, when the free space is below a predetermined threshold, notifies the memory management unit 17 of a warning signal indicating that until the free space is below the threshold. It has the function to do. Further, when receiving the notification to that effect, the memory management means 17 has a function of controlling so that the compressed data stored in the image storage unit 14 is thinned out in order from the oldest recording time.
[0023]
FIG. 4 is a flowchart illustrating an example of a processing method in the memory capacity control unit 15. The memory capacity control unit 15 executes the process shown in FIG. 4 every time compressed data is written in the image storage unit 14. First, in step S1, the monitoring unit 16 obtains free space information from the image storage unit 14, and in subsequent step S2, determines whether or not the free space is equal to or less than a predetermined threshold value. If the determination result indicates that the free space exceeds the threshold value, the above process ends. On the other hand, when the determination result indicates that the free space is equal to or less than the threshold value, the monitoring unit 16 issues a warning signal ALM to the memory management unit 17.
[0024]
In the next step S3, the memory management means 17 that has received the warning signal ALM searches the compressed data stored in the image storage unit 14 for the compressed data with the old recording time (compressed frame group), and continues. In step S4, the image storage unit 14 is controlled so that the compressed data is thinned out at a predetermined reverse thinning rate α (0 <α <1) in order of oldest recording time. For example, compressed as if the frame group G ₀ shown in FIG. 3 becomes thinned out, the difference frame df ₁ of the frame group _{G 0, ..., df k-} 1 frame number C _N is compressed to α times Data (compressed difference frame) is thinned out. At this time, the number of frames after compression is expressed by [C _N × M / N]. However, [x] is a symbol representing the maximum integer that does not exceed the number x, and [x] is an integer y that satisfies the relational expression y ≦ x <y + 1. Accordingly, in the number C _N 9 difference frame difference frame group G _0, if the inverse thinning rate α is 4/5, the number of difference frames after thinning, [9 × 4/5] = [7.2 ] = 7 sheets. Since the reference frame is a frame that determines the image quality of the moving image, the compressed data is not thinned out in principle.
[0025]
After the above step S4 ends, the above memory capacity control process ends. Then, the memory capacity control process is repeatedly executed until the accumulation of the compressed data in the image storage unit 14 is completed. If the process is repeatedly executed, the number of reference frames stored in the image storage unit 14 increases with the elapsed time in order to thin out only the difference frames in principle. Therefore, the memory management unit 17 has a function of controlling to delete a predetermined number of past reference frames from the image storage unit 14 automatically or manually when the number of stored reference frames exceeds a predetermined threshold. .
[0026]
As described above, according to the above moving image recording system, the free space of the image storage unit 14 can always be maintained at a constant level. For this reason, even when an unexpectedly large amount of moving image data is transmitted from a large number of monitoring servers 1 on the IP network NW1, it is possible to reliably avoid a situation where the free capacity is insufficient. In addition, compressed data with an old recording time is not completely deleted, but is compressed at a predetermined inverse thinning rate, so that an image with an old recording time can also be reproduced. Therefore, it is possible to record moving image data for a long time.
[0027]
Next, a second processing method in the memory capacity control unit 15 will be described below with reference to FIGS. As shown in FIG. 5, the image storage unit 14 sequentially stores compressed data of a series of frames f ₀ , df, df,..., F ₂ , df, ..., f _n , df,. . Here, the frame f _i (i: integer) indicates the i-th reference frame, and the frame df indicates the difference frame. The number of difference frames between the reference frames f _i and f _{i + 1} is K ₀ (K ₀ : integer of 2 or more). The i-th reference frame f _i and the difference frames df,..., Df generated from the reference frame f _i constitute an i-th frame group. As will be described later, the memory management unit 17 performs a thinning process with a frame group as a unit.
[0028]
Further, the memory management means 17 uses a series {α, α ² ,..., Α ^m−1 , α ^m } of inverse thinning rates α ^k (0 <α <1; k = 1, ² ,. (M is a natural number of 2 or more), and the reverse thinning rate α ^k belonging to this sequence is applied to the frame group to be thinned, and the number of difference frames belonging to each frame group is compressed to α ^k times Control to do. Here, the initial value α can be set variably, and a sequence of decimation rates {α, α ² ,..., Α ^m−1 , α ^m } is generated using the initial value α as a generation source. Each frame group is assigned a reverse decimation rate having a smaller value from the oldest recording time of the frame group. That is, the reverse thinning rate α ^k is assigned so that the number of thinned out difference frames belonging to the frame group increases as the recording time of the frame group decreases. Specifically, as shown in FIG. 6, the reverse decimation rate α ⁿ for the first frame group, the reverse decimation rate α ^{n-1 for} the second frame group, and the reverse decimation rate for the n-1 frame group. The rate α ² is assigned, and the inverse thinning rate α is assigned to the nth frame group. As a result of assigning the reverse thinning rate to each frame group, as shown in FIG. 7, the number of difference frames in the first frame group is [K ₀ × α ⁿ ] (= K ₁ ), and the difference in the second frame group The number of frames is [K ₀ × α ⁿ⁻¹ ] (= K ₂ ), the number of difference frames in the n−1th frame group is [K ₀ × α ² ] (= K _n−1 ), The number of difference frames in the nth frame group is [K ₀ × α ¹ ] (= K _n ) and is compressed. In the example shown in FIG. 7, the numbers K ₁ and K ₂ of the first frame group and the second frame group are both zero.
[0029]
FIG. 8 is a flowchart showing a second processing method of the memory capacity control unit 15 having the above functions. In FIG. 6, blocks with the same step numbers as the step numbers shown in FIG.
[0030]
The processing in steps S1 and S2 in the figure is as described above. In step S2, when the monitoring unit 16 determines that the free space in the image storage unit 14 is equal to or less than the threshold value and issues a warning signal ALM to the memory management unit 17, the memory management unit 17 in the next step S10, A frame group in which the number of difference frames is 1 or more and the oldest recording time is searched from among the frame groups recorded in the image storage unit 14. In the example shown in FIG. 6, the frame group with the oldest recording time is the first frame group. In subsequent step S11, the memory management means 17 assigns a reverse thinning rate α ^k to each frame group in order from the first frame group as shown in FIG. In the next step S12, as shown in FIG. 7, the memory management means 17 performs control so as to thin out the compressed data of the difference frames of each frame group to which the reverse thinning rate α ^k is assigned.
[0031]
After step S12 ends, the above memory capacity control process ends. Subsequently, when the next compressed data is written in the image storage unit 14, the memory capacity control process is started again. That is, after the above-described steps S1 and S2 are executed, in step S10, as shown in FIG. 7, the third frame group in which the number of difference frames is one or more and the oldest recording time is searched. In the next step S11, the reverse thinning rate α ^k is assigned to each frame group in order from the third frame group. Therefore, the reverse decimation rate α ⁿ for the third frame group, the reverse decimation rate α ^n-1 for the fourth frame group, the reverse decimation rate α ² for the n + 1 frame group, and the reverse for the n + 2 frame group. A thinning rate α is assigned to each. In the next step S12, the memory management means 17 uses the compressed data of the difference frames of each frame group to which the inverse thinning rate α ^k is assigned as the number of the difference frames multiplied by the inverse thinning rate α ^k. Control the thinning out. When the memory capacity control process is repeatedly executed as described above, the number of reference frames stored in the image storage unit 14 increases with the elapsed time. Therefore, in step S12, when the number of reference frames accumulated exceeds a predetermined threshold, the memory management unit 17 automatically or manually selects a reference for a group of frames having no difference frame from the image storage unit 14. It can be controlled to thin out the frame.
[0032]
As described above, in the second processing method, since the frame group is assigned with a reverse thinning rate that is low in the order of the recording time, the difference frames are thinned out in the descending order of the recording time, and the image quality of the moving image is gradually deteriorated. It becomes possible to make it.
[0033]
【The invention's effect】
As described above, according to the moving image recording system according to the first aspect of the present invention and the moving image recording method according to the second aspect of the present invention, the free capacity of the image storage unit can always be maintained at a constant level. For this reason, even when an unexpectedly large amount of moving image data is transmitted from the monitoring server on the network, it is possible to reliably avoid a situation where the free space of the image storage unit is insufficient. In addition, since compressed data with an old recording time is not completely deleted but is thinned out in units of frames, moving image data can be recorded over a long period of time, and a moving image with an old recording time can also be reproduced. .
[0034]
In addition , according to the first and second aspects, since the compressed data of the reference frame and the difference frame is stored in the image storage unit, the memory management unit easily performs the thinning-out control for each frame. be able to.
[0035]
Furthermore, according to claim 1 and claim 2 , since the number of thinned-out difference frames belonging to the frame group increases as the recording time of each frame group increases, the level of recording increases as the recording time increases. Therefore, it is possible to degrade the image quality of the moving image.
[Brief description of the drawings]
FIG. 1 is a functional block diagram showing a schematic configuration of a moving image recording system according to an embodiment of the present invention.
FIG. 2 is a schematic diagram for explaining compression encoding processing in a compression encoding unit according to the embodiment.
FIG. 3 is a schematic diagram showing a reference frame and a difference frame arranged in time series.
FIG. 4 is a flowchart illustrating a processing method in a memory capacity control unit of the image recording apparatus according to the embodiment.
FIG. 5 is a schematic diagram showing a series of frames stored in an image storage unit of the image recording apparatus according to the embodiment.
FIG. 6 is a schematic diagram for explaining a second processing method in the memory capacity control unit of the image recording apparatus according to the embodiment.
FIG. 7 is a schematic diagram for explaining a second processing method in the memory capacity control unit of the image recording apparatus according to the embodiment.
FIG. 8 is a flowchart showing a second processing method in the memory capacity control unit of the image recording apparatus according to the embodiment.
[Explanation of symbols]
DESCRIPTION OF SYMBOLS 1 Monitoring server 2 Imaging camera 3 _{1-3 n} Monitoring point 4 Image recording apparatus 10 Image input / output part 11 Compression encoding part 12 Transmission processing part 13 Reception processing part 14 Image storage part 15 Memory capacity control part 16 Monitoring means 17 Memory management means

Claims (2)

A moving image recording system comprising a monitoring server and an image recording apparatus for recording moving image data transmitted from the monitoring server via a network,
The monitoring server is
A compression encoding unit that generates and outputs compressed data obtained by compression encoding input video data;
A transmission processing unit that converts the compressed data into transmission data and sends it to the image recording device,
The image recording apparatus includes:
A reception processing unit that receives the transmission data from the monitoring server and reconstructs the compressed data;
An image storage unit for storing the compressed data;
Monitoring means for monitoring the free space of the memory of the image storage unit and issuing a warning signal when the free space is reduced to a predetermined amount;
Memory management means for controlling the compressed data stored in the image storage unit when the warning signal is received from the monitoring means so as to be thinned out in frame units in order of oldest recording time ;
The compression encoding unit of the monitoring server generates a reference frame obtained by periodically specifying a frame of input moving image data, and a difference frame including a difference signal between the reference frame and the moving image data Generating and outputting compressed data obtained by compressing and encoding the reference frame and the difference frame,
The memory management means includes
Managing a group of frames consisting of the reference frame and the difference frame generated from the reference frame as a unit, and having a plurality of inverse thinning rates set variably,
For each frame group in the compressed data stored in the image storage unit when the warning signal is received from the monitoring means, the recording time of the frame group belongs to the frame group as the recording time is older. Assigning the reverse thinning rate so that the number of thinned out difference frames is large;
Video recording system. A moving image recording method for recording moving image data transmitted via a network from one or a plurality of monitoring points,
(A) generating compressed data obtained by compressing and encoding moving image data, converting it into transmission data, and sending it to the network;
(B) receiving the transmission data sent in the step (a) and reconstructing the compressed data;
(C) storing the compressed data in an image storage unit;
(D) monitoring the free space of the memory of the image storage unit and issuing a warning signal when the free space is reduced to a predetermined amount;
(E) When the warning signal is issued in the step (d), the compressed data stored in the image storage unit in the step (c) is controlled to be thinned out in units of frames in the order of oldest recording time. A process,
In the step (a), the compressed data generates a reference frame obtained by periodically specifying a frame of the moving image data, and a difference frame including a difference signal between the reference frame and the moving image data. Generated by compressing and encoding the reference frame and the difference frame,
The step (e)
(E1) managing a group of frames composed of the reference frame and the difference frame generated from the reference frame as a unit;
(E2) preparing a plurality of reverse thinning rates set variably;
(E3) When the warning signal is issued in the step (d), for each of the frame groups in the compressed data, the difference belonging to the frame group as the recording time of the frame group becomes older Assigning the reverse thinning rate so that the number of thinned frames increases,
Consist of,
Video recording method .

Images