WO2013145530A1 - Système d'analyse - Google Patents

Système d'analyse Download PDF

Info

Publication number
WO2013145530A1
WO2013145530A1 PCT/JP2013/000797 JP2013000797W WO2013145530A1 WO 2013145530 A1 WO2013145530 A1 WO 2013145530A1 JP 2013000797 W JP2013000797 W JP 2013000797W WO 2013145530 A1 WO2013145530 A1 WO 2013145530A1
Authority
WO
WIPO (PCT)
Prior art keywords
analysis
unit
target data
processing
data acquisition
Prior art date
Application number
PCT/JP2013/000797
Other languages
English (en)
Japanese (ja)
Inventor
山田　洋志
Original Assignee
日本電気株式会社
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by 日本電気株式会社 filed Critical 日本電気株式会社
Publication of WO2013145530A1 publication Critical patent/WO2013145530A1/fr

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L67/00—Network arrangements or protocols for supporting network services or applications
    • H04L67/50—Network services
    • H04L67/535—Tracking the activity of the user
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N23/00—Cameras or camera modules comprising electronic image sensors; Control thereof
    • H04N23/60—Control of cameras or camera modules
    • H04N23/61—Control of cameras or camera modules based on recognised objects
    • H04N23/611—Control of cameras or camera modules based on recognised objects where the recognised objects include parts of the human body

Definitions

  • the present invention relates to an analysis system, and more particularly to an analysis system in which a device that acquires data to be analyzed and a device that performs analysis processing are connected via a network.
  • the video monitoring system includes an object recognition unit, a recognition result analysis unit, a video processing control parameter determination unit, and a video processing control parameter setting unit.
  • the video surveillance system described in Patent Document 1 having the above-described configuration operates as follows. First, the object recognition unit recognizes an object from the video data of the camera, the recognition result analysis unit analyzes the recognition result, and the minimum size of the face in the video recognized by the object recognition unit from the received frame image, Find the moving speed. Then, the video processing control parameter determination unit determines a lower limit value of the frame rate and the resolution from the minimum face size and the moving speed. Thereafter, the determined frame rate and the lower limit value of the resolution are sent to the video processing control parameter setting unit to set the camera. As described above, in the technique described in Patent Document 1, the amount of data is reduced by reducing the frame rate of video data.
  • this video surveillance system is composed of a camera and a surveillance center, and the camera has a programmable internal module configuration.
  • the video surveillance system described in Patent Document 2 having the above-described configuration operates as follows. First, a camera and a monitoring center are connected via a network. Based on the video data to be monitored, the camera acquires, from the module management center, a program module that executes a recognition process for detecting an abnormality of the monitoring target and link information of the program module as necessary. As described above, in the technique described in Patent Document 2, the video data analysis processing is executed by the module on the camera side, and only the analysis processing result is transmitted to the monitoring center, thereby reducing the data amount. Yes.
  • the object of the present invention is the analysis system configured by connecting the data acquisition device that acquires the data to be analyzed and the analysis device that executes the analysis process, which is the above-described problem, via a network.
  • the object is to solve the problem that it is impossible to achieve both a reduction in the load on the network and a reduction in the performance of the device that acquires the analysis target data.
  • An analysis system is: An analysis target data acquisition device for acquiring analysis target data, connected to each other via a network, and an analysis processing device for executing a predetermined analysis process
  • the analysis target data acquisition device includes a first analysis unit that executes a predetermined analysis process on the acquired analysis target data, a data transmission unit that transmits data based on the acquired analysis target data to the analysis processing unit,
  • the analysis processing device includes second analysis means for executing a predetermined analysis process on the data transmitted from the analysis target data acquisition device,
  • the first analysis unit and the second analysis unit each include an analysis unit that executes an analysis process of equivalent content, Among the analysis units provided in the first analysis means and the second analysis means, respectively, provided with an analysis control means for controlling either one or both to operate,
  • the configuration is as follows.
  • the analysis object data acquisition device is An analysis target data acquisition device for acquiring analysis target data, connected to an analysis processing device that executes a predetermined analysis process, First analysis means for executing a predetermined analysis process on the acquired analysis target data; Data transmission means for transmitting data based on the acquired analysis target data to the analysis processing device,
  • the first analysis means includes another analysis unit that executes an analysis process having the same content as the analysis unit included in the analysis processing device, Among the analysis units provided in the first analysis means and the second analysis means, respectively, provided with an analysis control means for controlling either one or both to operate,
  • the configuration is as follows.
  • the program which is the other form of this invention is: To the analysis target data acquisition device that acquires the analysis target data connected to the analysis processing device that executes the predetermined analysis processing via the network, First analysis means for executing a predetermined analysis process on the acquired analysis target data; Data transmission means for transmitting data based on the acquired analysis target data to the analysis processing device; Analysis control means for controlling either one or both of the analysis units that perform analysis processing of equivalent contents respectively provided by the first analysis means and the second analysis means, It is a program for realizing.
  • an analysis processing method includes: In a communication system comprising an analysis target data acquisition device for acquiring analysis target data and an analysis processing device for executing a predetermined analysis process, which are connected to each other via a network,
  • the analysis target data acquisition device performs predetermined analysis processing on the acquired analysis target data, and transmits data based on the acquired analysis target data to the analysis processing device.
  • the analysis processing device performs a predetermined analysis process on the data transmitted from the analysis target data acquisition device,
  • the analysis target data acquisition device and the analysis processing device can each perform an analysis process of equivalent content, and either one or both of the analysis target data acquisition device and the analysis processing device analyze the equivalent content. Control to execute the process,
  • the configuration is as follows.
  • the present invention is configured as described above so that a data acquisition device that acquires data to be analyzed and an analysis device that executes analysis processing are connected via a network. It is possible to achieve both the reduction of the load and the performance reduction of the apparatus for acquiring the analysis target data.
  • Embodiment 1 of this invention It is a block diagram which shows the structure of the analysis system in Embodiment 1 of this invention. It is a flowchart which shows operation
  • FIG. 1 is a diagram for explaining the configuration of the analysis system in the present embodiment
  • FIGS. 2 to 3 are diagrams for explaining the operation of the analysis system.
  • the analysis system in this embodiment is a video surveillance system, and as shown in FIG. 1, a monitoring camera 1 (analysis target data acquisition device) that acquires video data to be analyzed and an analysis server 2 (analysis processing data acquisition device) Analysis processing device).
  • the monitoring camera 1 and the analysis server 2 are connected via a network 3 such as the Internet network or a LAN (Local Area Network).
  • the analysis system according to the present invention is not limited to the video surveillance system described below, and performs other analysis processing with an analysis target data acquisition device that acquires other analysis target data such as audio data.
  • the analysis processing apparatus may be configured to be connected via a network. Further, the network 3 may be any network regardless of wireless wired.
  • the monitoring camera 1 includes an arithmetic device and a storage device, and as shown in FIG. 1, a video acquisition unit 11, a first video analysis unit 12, a camera control unit 13, a first load determination unit 14, It has.
  • the first video analysis unit 12, the camera control unit 13, and the first load determination unit 14 are constructed by incorporating a program into the arithmetic device.
  • the video acquisition unit 11 includes a camera module that captures video, and also includes a processing unit that converts moving images and still images that are the captured video as video data. Then, the video acquisition unit 11 acquires video data by the operation being controlled by the camera control unit 13, and outputs the acquired video data to the first video analysis unit 12.
  • the first video analysis unit 12 executes analysis processing on the video data acquired by the video acquisition unit 11.
  • the first video analysis unit 12 includes one or more analysis modules (analysis units) that execute predetermined analysis processing, and executes all or only a part of the analysis modules.
  • the analysis module performs analysis processing such as processing for extracting a human face portion from video data and processing for extracting face feature data from a face image.
  • the first video analysis unit 12 may not perform any analysis processing by the analysis module.
  • the first video analysis unit 12 outputs the analysis result and / or the data before the analysis process to the camera control unit 13. The operation of the first video analysis unit 12 is controlled by the camera control unit 13.
  • At least one of the analysis modules included in the first video analysis unit 12 is equivalent to the analysis module (second substitutable analysis module) included in the second video analysis unit 22 included in the analysis server 2 described later. It is the 1st substitutable analysis module which performs analysis processing of contents.
  • the first substitutable analysis module and the second substitutable analysis module may be the same analysis module, and the same analysis result can be obtained by executing the analysis process of the same content.
  • the analysis module may be used.
  • the first load determination unit 14 (determination means) measures the load in the monitoring camera 1 and determines the load level. For example, the processing load of the entire arithmetic device equipped in the monitoring camera 1 may be measured, or the processing load of the first video analysis unit 12 may be measured. Then, the measured load level is determined based on a preset criterion. For example, when the low level reference value and the high level reference value are set as the reference value, and the measured load is lower than the low level reference value, the load level is determined to be low and measured. When the load is higher than the high level reference value, it is determined that the load level is a high level, and when it is not any, it is determined that the load level is other level. Then, the first load determination unit 14 outputs the determined load level to the camera control unit 13.
  • the 1st load determination part 14 in this embodiment demonstrates the case where the processing load of the arithmetic unit in the monitoring camera 1 is measured as a load of the said monitoring camera 1, it measures the load of the monitoring camera 1 by another method. Then, the load level may be determined. For example, the amount of memory used in the surveillance camera 1, the delay status of processing executed by the surveillance camera 1, the amount of data input / output to / from the surveillance camera 1, the status of the network to which the surveillance camera 1 is connected, etc. You may measure the load of the said monitoring camera 1 from the information showing the operating condition of the monitoring camera 1, and may determine the level based on the preset reference
  • the camera control unit 13 (analysis control unit, data transmission unit) controls the operation of the camera control unit 12 based on the load level of the monitoring camera 1 output from the first load determination unit 14. At the same time, the camera control unit 13 controls data communication with the analysis server 2.
  • the camera control unit 13 first performs control so that the analysis process by the analysis module preset by the first video analysis unit 12 and the analysis process by the first substitutable analysis module are executed in the normal state.
  • the analysis result data by the first video analysis unit 12 that is, the analysis result data by the first substitutable analysis module is transmitted to the analysis server 2 via the network.
  • the camera control unit 13 checks the load level of the monitoring camera 1 output from the first load determination unit 14 while executing the analysis process and the transmission process, and the load level is not “high load level”. In this case, the control is performed as it is.
  • the camera control unit 13 operates the analysis module operating in the first video analysis unit 12. Among these, control is performed to stop the operation of the first substitutable analysis module described above. Accordingly, the camera control unit 13 transmits data to be analyzed by the first substitutable analysis module to the analysis server 2 via the network. Further, the camera control unit 13 indicates that the operation of the first substitutable analysis module has been stopped, that is, the analysis server 2 executes analysis processing having the same content as the analysis processing by the first substitutable analysis module. The instruction is transmitted to the analysis server 2.
  • the camera control unit 13 indicates that the load level of the monitoring camera 1 output from the first load determination unit 14 is “when the operation of the first substitutable analysis module is stopped by the first video analysis unit 12.
  • the first video analysis unit 12 is controlled to resume the operation of the first substitutable analysis module stopped. Accordingly, the camera control unit 13 transmits the data of the analysis result by the first video analysis unit 12 to the analysis server 2 via the network.
  • the camera control unit 13 indicates that the operation of the first substitutable analysis module has been resumed, that is, the analysis server 2 stops the analysis processing having the same content as the analysis processing by the first substitutable analysis module. The instruction is transmitted to the analysis server 2.
  • the analysis server 2 is a general information processing device (analysis processing device) including an arithmetic device and a storage device, and is configured by one or a plurality of information processing devices. And the analysis server 2 is provided with the result output part 21, the 2nd image
  • the result output unit 21 outputs an analysis processing result by a second video analysis unit 22 described later via an output device such as a monitor or a printer.
  • the second video analysis unit 22 executes analysis processing on data transmitted from the monitoring camera 1 via the network 3.
  • the second video analysis unit 22 includes one or more analysis modules (analysis units) that execute predetermined analysis processing, and executes all or only a part of the analysis modules.
  • the analysis module performs analysis processing such as processing for extracting face feature data from a face image and processing for matching a specific face from face feature data.
  • the second video analysis unit 22 outputs the analysis result to the result output unit 21.
  • the operation of the second video analysis unit 22 is controlled by the server control unit 23.
  • At least one of the analysis modules included in the second video analysis unit 22 is equivalent to the analysis module (first substitutable analysis module) included in the first video analysis unit 12 included in the monitoring camera 1 described above. It is a 2nd substitutable analysis module which performs analysis processing of contents.
  • the second substitutable analysis module and the first substitutable analysis module may be the same analysis module, and an analysis process having the same content is executed to obtain an equivalent analysis result. It may be another analysis module that can be obtained.
  • the correspondence storage unit 41 stores information representing the correspondence between the second substitutable analysis module and the first substitutable analysis module, and is constructed in a storage device connected to the analysis server 2. ing. Specifically, the correspondence relationship storage unit 41 stores information for identifying the first substitutable analysis module and the second substitutable analysis module corresponding to each other in association with each other. Execution methods such as parameters set when executing the possibility analysis module are also stored. In addition, although the example of FIG. 1 has shown the case where the correspondence storage part 2 is equipped in the apparatus connected to the analysis server 2, it may be equipped in the analysis server 2, and other It may be equipped at the place.
  • the server control unit 23 (analysis control unit) controls the operation of the second video analysis unit 22 based on the information transmitted from the monitoring camera 1 and the information in the correspondence storage unit 41. At the same time, the server control unit 23 controls data communication with the monitoring camera 1.
  • the server control unit 23 first, in a normal state, analyzes data set in advance by the second video analysis unit 22 with respect to data transmitted from the monitoring camera 1, and a second alternative is possible.
  • the analysis process by the analysis module is controlled to be executed, and the analysis result is output to the result output unit 21.
  • the server control unit 23 performs the control as it is.
  • the server control unit 23 analyzes the same content as the analysis process from the monitoring camera 1 by the first substitutable analysis module as described above. Since an instruction to execute the process in the analysis server 2 is transmitted, control is performed so that the analysis process by the second substitutable analysis module is executed in accordance with the instruction. At this time, the server control unit 23 receives information specifying the first substitutable analysis module that has stopped the analysis processing on the monitoring camera 1 side from the monitoring camera 1 and corresponds to the first substitutable analysis processing module. The second substitutable analysis processing module is specified from the information in the correspondence storage unit 41 and controlled to be executed.
  • the server control unit 23 is configured as described above. As described above, the instruction to stop the second substitutable analysis module is received from the monitoring camera 1, and the analysis processing by the second substitutable analysis module is controlled to stop.
  • the first video analysis unit 12 of the analysis camera 1 performs analysis processing by the first substitutable analysis module, and the second substitutable analysis that performs analysis processing equivalent to this is performed.
  • the analysis process by the module is not executed by the second video analysis unit 22 of the analysis server 2.
  • the first load determination unit 14 determines the load of the monitoring camera 1 (step S1).
  • the determination timing may be set based on system requirements such as a fixed interval, a specific time, and a start / end timing of processing.
  • the camera control unit 13 is executed by the first video analysis unit 12.
  • the analysis process that can be stopped is selected (step S3), and the analysis server 2 is requested to execute the equivalent process (step S4).
  • the camera control unit 13 selects that the analysis process by the first substitutable analysis module executed in the first video analysis unit 12 can be stopped, and stops the first substitutable analysis module. Is sent to the analysis server 2.
  • the camera control unit 13 receives a notification from the analysis server 2 that analysis processing equivalent to that of the first substitutable analysis module can be executed (“OK” in step S5), the first substitutable analysis is performed.
  • the analysis process by the module is stopped (step S6).
  • the camera control unit 13 sends the data to be analyzed by the first substitutable analysis processing module to the analysis server 2 when the analysis processing is stopped by the first substitutable analysis module. Send to.
  • step S11 When the server control unit 23 receives a processing request from the monitoring camera 1 (step S11), the server control unit 23 performs processing according to the type (step 12, S13, S15). Specifically, as in step S4 in FIG. 2, when a request for execution of analysis processing equivalent to that of the first substitutable analysis module is received from the monitoring camera 1 ("execution of processing" in step S11), first, A second substitutable analysis module that is an analysis module that executes analysis processing equivalent to that of the first substitutable analysis module is specified from within the relationship storage unit 41.
  • step S12 it is determined whether or not the analysis process by the specified second substitutable analysis module can be executed by the analysis server 2 according to the situation of the analysis server 2 (step S12).
  • the analysis processing by the second substitutable analysis module is executed (step S13).
  • the analysis server 2 outputs the result of the analysis process by the 2nd image
  • step S2 an analysis process that can be executed is selected from the analysis processes stopped by the first video analysis unit 12 (step S7).
  • the camera control unit 13 selects that the analysis process by the first substitutable analysis module stopped by the first video analysis unit 12 can be started, and a process equivalent to the first substitutable analysis module. Is requested to the analysis server 2 to stop (step S8).
  • the camera control unit 13 receives a notification from the analysis server 2 that analysis processing equivalent to that of the first substitutable analysis module can be stopped (“OK” in step S9), the first substitutable analysis is performed.
  • the analysis process by the module is resumed (step S10).
  • the server control unit 23 receives a request to stop the analysis process equivalent to the first substitutable analysis module from the monitoring camera 1 (“process stop” in step S11), the server control unit 23 receives the analysis module executed by the analysis server 2 Among these, the second substitutable analysis module, which is an analysis module that executes an analysis process equivalent to that of the first substitutable analysis module, is specified from the correspondence storage unit 41. Then, the server control unit 23 stops the analysis process by the specified second substitutable analysis module (step S15). The analysis server 2 outputs the result of the analysis processing by the second video analysis unit 22 including the second substitutable analysis module from the result output unit 21 (step S14).
  • the analysis processing module that executes the analysis processing of the equivalent content according to the determination result of the load of the monitoring camera 1 by the first load determination unit 14 of the monitoring camera 1.
  • the specific analysis process according to is operated by either the monitoring camera 1 or the analysis server 2. For this reason, normally, when a specific analysis process is executed on the surveillance camera 1, only the result of the specific analysis process is transmitted to the analysis server 2 via the network N. It is reduced. On the other hand, when the load on the monitoring camera 1 is increased, the execution of the specific analysis process is switched from the monitoring camera 1 to the analysis server 2, and therefore the specific analysis process can be continued as the entire analysis system. . Accordingly, it is possible to reduce the performance of the monitoring camera 1 while reducing the load on the network 3, and to construct an analysis system that achieves both improvement in performance and reduction in equipment costs.
  • the camera control unit 13 and the server control unit 23 may control such that both perform analysis processing simultaneously. As a result, part of the data to be analyzed is analyzed by the monitoring camera 1 and the rest is analyzed by the analysis server 2, and load distribution in the entire system can be achieved.
  • FIG. 4 is a block diagram showing the configuration of the video analysis system in the present embodiment
  • FIG. 5 is a diagram showing an example of data stored in the video analysis system
  • FIG. 6 is a diagram for explaining the state of analysis processing by the video analysis system.
  • This embodiment shows an example of the specific configuration and operation of the video analysis system described in the first embodiment. For example, a person who has registered a video monitoring system in the video of the surveillance camera 1 is shown in the video. In this example, the system is applied to a registrant notification system in which an alarm is issued when an error occurs.
  • the video analysis system first performs face extraction processing 72 for extracting a human face portion on the video data 71 acquired by the monitoring camera 1, and outputs face image data 73. Subsequently, the video analysis system performs a facial feature extraction process 74 that extracts facial features such as the positions of eyes and noses on the facial image data 73, and outputs facial feature data 75. Further, the video analysis system collates the face feature data 75 with the data in the face database, performs face collation processing 76 for determining whether the person is registered in the face database, and outputs a collation result 77.
  • the first video analysis unit 12 of the monitoring camera 1 includes a face extraction module 12a that performs the above-described face extraction processing 72 as shown in FIG. And a facial feature extraction module 12b that performs 74 extraction processes.
  • the second video analysis unit 22 of the analysis server 2 includes a face feature extraction module 22a that performs the above-described face feature extraction processing 74 and a face matching module 22b that performs the above-described face matching processing 76.
  • the facial feature extraction module 12b included in the first video analysis unit 12 and the facial feature extraction module 22a included in the second video analysis unit 22 are analysis modules that perform analysis processing of equivalent contents, and are functional. Are the same.
  • the facial feature extraction modules 12b and 22a may be completely the same module, but are not necessarily the same.
  • the OS Operating System
  • the modules may correspond to the respective OSs or may be completely different modules.
  • the correspondence relationship between the face feature modules 12b and 22a is stored as data in the correspondence storage unit 41 shown in FIG.
  • the data in the correspondence storage unit 41 shown in FIG. 5 includes, for each monitoring camera 1 (for each camera ID), an analysis module (first substitutable analysis module) mounted on the monitoring camera 1. And an analysis module (second substitutable analysis module) mounted on the analysis server 2 that performs an analysis process having the same content as that in FIG. For example, in the second registered item from the top in FIG. 5 (see reference numeral 41a in FIG.
  • the “face feature extraction” module installed in the surveillance camera 1 (C1) is “ -mode 1 ”is executed with the parameter“ 20 ”, and the case where the“ face inspection ”module installed in the analysis server 2 is executed with the parameter“ 20 ”is registered in association with the analysis processing of the same content. Yes. That is, an equivalent analysis result can be obtained by performing any of the analysis processes associated with each other.
  • the correspondence storage unit 41 is installed in the analysis server 2, but may be installed at any position.
  • the face feature extraction process 74 shown in FIG. 6 can be executed by either the monitoring camera 1 or the analysis server 2.
  • the facial feature extraction process 74 is executed by the monitoring camera 1
  • the facial feature data 75 is transmitted from the monitoring camera 1 to the analysis server 2, but when the facial feature extraction process 74 is executed by the analysis server 2.
  • the face image data 73 is transmitted from the monitoring camera 1 to the analysis server 2.
  • the amount of data output from the monitoring camera 1 is smaller in the face feature data 75 than in the face image data 73, so that the face feature extraction process 74 is executed by the monitoring camera 1.
  • the load on 3 is reduced.
  • the facial feature extraction process 74 may be unable to be processed by the surveillance camera 1 when a large number of people are reflected because the load varies depending on the number of people shown in the video. Therefore, in the normal state, the facial feature extraction processing 74 is executed by the monitoring camera 1 to reduce the load on the network 3, and when the monitoring camera 1 reaches a high load level by the first load determination unit 14, A part or all of the feature extraction processing 74 can be requested to reduce the load on the monitoring camera 1.
  • the alarm display application 5 provides information to the user.
  • FIG. 7 is a block diagram showing the configuration of the video analysis system in the present embodiment
  • FIG. 8 is a diagram for explaining the state of analysis processing by the video analysis system.
  • this embodiment shows an example of a specific configuration and operation of the video analysis system described in the first embodiment.
  • the video monitoring system can be used to identify a person shown in the video data of the monitoring camera 1 by gender.
  • This example shows an application to a human video search system in which a search is performed by designating clothes and clothes.
  • the video analysis system performs face extraction processing 82 for extracting a human face portion on the video data 81 acquired by the monitoring camera 1, and outputs face image data 83. Subsequently, the video analysis system performs gender determination processing 84 for determining gender based on the face image data 83, outputs gender data 85, and performs information registration processing 86. In parallel with the above processing, the video analysis system performs a person extraction process 87 for extracting the whole body part of the person from the video data 81 acquired by the monitoring camera 1 and outputs the person image data 88. Subsequently, the video analysis system performs a clothing extraction process 89 for extracting information such as the type and color of clothes from the person image data 88 and performs an information registration process 86.
  • the first video analysis unit 12 of the surveillance camera 1 in this embodiment includes a face extraction module 12c for performing the face extraction processing 82 and the gender determination processing 84 as shown in FIG.
  • a gender determination module 12d that performs the above-described process
  • a person extraction module 12e that performs the person extraction process 87
  • a clothing extraction module 12f that performs the clothes extraction process 89.
  • the second video analysis unit 22 of the analysis server 2 includes a gender determination module 22 c that performs the gender determination process 84 and a clothing extraction module 22 d that performs the clothes extraction process 89.
  • the gender determination module 12d included in the first video analysis unit 12 and the gender determination module 22c included in the second video analysis unit 22 are analysis modules that perform analysis processing of equivalent contents, and are functionally Are the same.
  • the clothing extraction module 12f included in the first video analysis unit 12 and the clothing extraction determination module 22d included in the second video analysis unit 22 are analysis modules that perform analysis processing of equivalent contents, and are functionally Are the same.
  • the gender determination process 84 and the clothing extraction process 89 shown in FIG. 8 can be executed by either the monitoring camera 1 or the analysis server 2. Since the above two processes (gender determination process 84 and clothes extraction process 89) are independent, either one or both may be monitored camera 1 or analysis server 2 depending on the load state of monitoring camera 1 or analysis server 2. It may be executed with. Thereby, the load of the network 3 can be reduced and the performance of the surveillance camera 1 can be reduced.
  • the gender and clothes information which is the analysis result by the video analysis system, is output to the information registration / search application 6 and registered in the database. Thereby, the information registered in the database can be searched from the information registration / search application.
  • FIG. 9 is a block diagram showing the configuration of the video analysis system in the present embodiment.
  • the video analysis system according to the present embodiment has substantially the same configuration as the video analysis system according to the first embodiment described above.
  • the analysis server 2 includes a second load determination unit 24 (determination means).
  • the second load determination unit 24 measures the preset load in the analysis server 2 and determines the load level, similarly to the first load determination unit 14 described above. For example, the processing load of the entire arithmetic device provided in the analysis server 2 may be measured, or the processing load by the second video analysis unit 22 may be measured. Then, the measured load level is determined based on a preset criterion. Then, the second load determination unit 24 outputs the determined load level to the server control unit 23.
  • the second load determination unit 24 may measure the load of the analysis server 2 by another method and determine the load level. For example, the amount of memory used in the analysis server 2, the delay state of processing executed in the analysis server 2, the amount of data input / output to the analysis server 2, the state of the network to which the analysis server 2 is connected, etc. You may measure the load of the said analysis server 2 from the information showing the operating condition of the analysis server 2, and may determine the level based on the preset reference
  • the server control unit 23 in the present embodiment controls the operation of the second video analysis unit 22 based on the load level of the analysis server 2 output from the second load determination unit 24.
  • the server control unit 23 is a case where the camera control unit 13 of the monitoring camera 1 is instructed to execute the second substitutable analysis module when the monitoring camera 1 is at a high load level.
  • the monitoring camera 1 may be answered that the second substitutable analysis module cannot be executed.
  • the server control unit 23 can execute the second substitutable analysis module for the monitoring camera 1 when it is determined that the analysis server 2 is at the low load level, and the first alternative The monitoring camera 1 may be notified that the possibility analysis module may be stopped.
  • the server control unit 23 and / or the camera control unit 13 controls the execution of the second substitutable module in the second video analysis unit 22, One or both of the one substitutable analysis module and the second substitutable analysis module may be controlled to be executed.
  • the server control unit 23 and / or the camera control unit 13 determines the first substitutable analysis module and the second substitutable analysis in accordance with the determination results of the second load determination unit 24 and the first load determination unit 14. One or both of the modules may be controlled to be executed.
  • the analysis server 2 is equipped with the second determination unit 24 and performs analysis processing of equivalent contents in each of the video analysis units 12 and 22 in consideration of operating conditions such as the load of the analysis server 2.
  • the method for controlling the module is also applicable to the second and third embodiments.
  • FIG. 9 is a block diagram showing the configuration of the video analysis system in the present embodiment.
  • the video analysis system in the present embodiment has substantially the same configuration as the video analysis system in the fourth embodiment described above.
  • the first load determination unit 14 and the second load determination unit 24 use different methods from the above-described embodiment to check the operation status of the monitoring camera 1 and the analysis server 2. Measure and judge.
  • a method for measuring the load on the monitoring camera 1 by the first load determination unit 14 of the monitoring camera 1 will be described, but the same method can also be applied to the second load determination unit 24 of the analysis server 2. It is.
  • the first load determination unit 14 (operation status measurement means) in the present embodiment stores the history of the operation status of the monitoring camera 1 measured so far in the storage device as operation history information. For example, the processing load of the arithmetic device equipped in the monitoring camera 1 is stored together with time information. Then, the first load determination unit 14 predicts the processing load that is the current operation state in the monitoring camera 1 from the past operation history information in the monitoring camera 1, and the level of the predicted processing load is set in advance. Judge based on criteria. More specifically, the first load determination unit 14 stores a record of load fluctuations, predicts the degree of increase or decrease, predicts the current load, and performs periodic fluctuations such as date and time. You can learn and predict the current load. The load level is determined from the predicted load of the monitoring camera 1 in this way, and the operation of the monitoring camera 1 and / or the analysis server 1 is controlled as described above.
  • the first load determination unit 14 can measure the load that is the operating state of the monitoring camera 1 by another method. For example, the first load determination unit 14 measures a situation such as a load on the network 3 connected to the monitoring camera 1, predicts the load on the monitoring camera 1 from the measurement result, and determines the predicted load level. Also good. As an example, when the load on the network 3 is lower than a preset reference value, it is considered that the amount of data transmitted from the monitoring camera 1 is small, and therefore it can be determined that the processing load on the monitoring camera 1 is low. Therefore, the current load of the monitoring camera 1 can be predicted based on the current load value of the network 3.
  • the subsequent process load can be predicted from the previous process result. is there.
  • the load of the face feature extraction process will increase.
  • the mutual relationship between the positions of the surveillance cameras can be used. For example, in the case of surveillance cameras arranged on a passage, it can be predicted that similar additional fluctuations occur in the order of arrangement according to the movement of a person.
  • load measurement by the first load determination unit 14 information from an external system can be used. For example, in a station monitoring camera, it is conceivable that the number of people will increase when a train arrives. Therefore, it is possible to predict the load by interlocking with operation information.
  • the analysis system in this invention is the above-mentioned. It is not limited to having the configuration.
  • the analysis system executes each analysis module of the first video analysis unit 12 and the second video analysis unit 22 based on other determination criteria without determining the operation status such as the load of the monitoring camera 1 or the analysis server 2. Or stop may be controlled. That is, the analysis system according to the present invention includes the first substitutable analysis module and the second substitutable analysis that execute the analysis processing of the same content without determining the operation status such as the load of the monitoring camera 1 or the analysis server 2. One or both of the modules may be controlled to operate.
  • An analysis target data acquisition device 100 that acquires analysis target data, and an analysis processing device 200 that executes a predetermined analysis process, connected to each other via a network 300;
  • the analysis target data acquisition apparatus 100 includes a first analysis unit 110 that executes a predetermined analysis process on the acquired analysis target data, and a data transmission unit that transmits data based on the acquired analysis target data to the analysis processing apparatus.
  • the analysis processing device 200 includes a second analysis unit 210 that performs a predetermined analysis process on the data transmitted from the analysis target data acquisition device 100,
  • the first analysis unit 110 and the second analysis unit 210 are respectively provided with analysis units 111 and 211 that execute analysis processing of equivalent contents,
  • An analysis control means 130 for controlling one or both of the analysis units 111 and 211 provided in the first analysis means 110 and the second analysis means 210 to operate is provided. Analysis system.
  • (Appendix 2) The analysis system according to attachment 1, wherein A determination unit that determines an operation state of the analysis target data acquisition device based on a preset criterion; The analysis control unit controls, based on the determination result by the determination unit, either one or both of the analysis units included in the first analysis unit and the second analysis unit, respectively. Analysis system.
  • the analysis control unit may execute the analysis process by the analysis unit of the first analysis unit when the determination unit determines that the processing load of the analysis target data acquisition apparatus is higher than a preset reference. Without controlling the analysis processing by the analysis unit of the second analysis means, Analysis system.
  • the analysis control unit determines that the processing load of the analysis target data acquisition device is higher than a preset reference by the determination unit when the analysis process by the analysis unit of the first analysis unit is being performed. If it is, control to stop the execution of the analysis process by the analysis unit of the first analysis means, and to execute the analysis process by the analysis unit of the second analysis means, Analysis system.
  • Appendix 5 The analysis system according to appendix 3 or 4, When the analysis control unit determines that the processing load of the analysis target data acquisition apparatus is higher than a preset reference by the determination unit, the data transmission unit is changed from the first analysis unit to the analysis unit. Control to transmit data based on the analysis target data to be analyzed to the second analysis means; Analysis system.
  • Appendix 7 The analysis system according to any one of appendices 2 to 5, An operation state measuring means for measuring the operation state of the analysis object data acquisition device and storing it as operation history information; The determination means predicts the operating status of the analysis target data acquisition device based on the operating history information, and determines the predicted operating status based on a preset criterion. Analysis system.
  • the analysis system according to appendices 2 to 5, The determination unit measures the network status, predicts the operating status of the analysis target data acquisition device based on the network status, and determines the predicted operating status based on a preset criterion. Analysis system.
  • An analysis target data acquisition device for acquiring analysis target data, connected to an analysis processing device that executes a predetermined analysis process, First analysis means for executing a predetermined analysis process on the acquired analysis target data; Data transmission means for transmitting data based on the acquired analysis target data to the analysis processing device,
  • the first analysis means includes another analysis unit that executes an analysis process having the same content as the analysis unit included in the analysis processing device, Among the analysis units provided in the first analysis means and the second analysis means, respectively, provided with an analysis control means for controlling either one or both to operate, Analysis target data acquisition device.
  • a communication system comprising an analysis target data acquisition device for acquiring analysis target data and an analysis processing device for executing a predetermined analysis process, which are connected to each other via a network
  • the analysis target data acquisition device performs predetermined analysis processing on the acquired analysis target data, and transmits data based on the acquired analysis target data to the analysis processing device.
  • the analysis processing device performs a predetermined analysis process on the data transmitted from the analysis target data acquisition device,
  • the analysis target data acquisition device and the analysis processing device can each perform an analysis process of equivalent content, and either one or both of the analysis target data acquisition device and the analysis processing device analyze the equivalent content. Control to execute the process, Analysis processing method.
  • the above-described program is stored in a storage device or recorded on a computer-readable recording medium.
  • the recording medium is a portable medium such as a flexible disk, an optical disk, a magneto-optical disk, and a semiconductor memory.

Landscapes

  • Engineering & Computer Science (AREA)
  • Signal Processing (AREA)
  • Computer Hardware Design (AREA)
  • General Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Multimedia (AREA)
  • Image Analysis (AREA)
  • Closed-Circuit Television Systems (AREA)
  • Studio Devices (AREA)
  • Communication Control (AREA)

Abstract

La présente invention concerne un système d'analyse qui comporte un dispositif d'acquisition de données pour une analyse et un dispositif de traitement d'analyse qui sont connectés par l'intermédiaire d'un réseau. Le dispositif d'acquisition de données pour une analyse comporte : un premier moyen d'analyse qui exécute un traitement d'analyse spécifié sur des données pour une analyse acquises ; et un moyen d'envoi de données qui, au dispositif de traitement d'analyse, envoie des données sur la base des données pour une analyse acquises. Le dispositif de traitement d'analyse comporte un second moyen d'analyse qui exécute un traitement d'analyse spécifié sur les données envoyées à partir du dispositif d'acquisition de données pour une analyse. Le premier moyen d'analyse et le second moyen d'analyse : comportent chacun une unité d'analyse, lesdites unités d'analyse exécutant des traitements d'analyse qui sont mutuellement équivalents en nature ; et comportent un moyen de commande d'analyse qui effectue une commande de telle sorte que l'unité d'analyse du premier moyen d'analyse et/ou l'unité d'analyse du second moyen d'analyse fonctionnent.
PCT/JP2013/000797 2012-03-28 2013-02-14 Système d'analyse WO2013145530A1 (fr)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
JP2012073350 2012-03-28
JP2012-073350 2012-03-28

Publications (1)

Publication Number Publication Date
WO2013145530A1 true WO2013145530A1 (fr) 2013-10-03

Family

ID=49258857

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/JP2013/000797 WO2013145530A1 (fr) 2012-03-28 2013-02-14 Système d'analyse

Country Status (2)

Country Link
JP (1) JPWO2013145530A1 (fr)
WO (1) WO2013145530A1 (fr)

Cited By (7)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2015151449A1 (fr) * 2014-03-31 2015-10-08 日本電気株式会社 Système de traitement d'image, dispositif de traitement d'image, procédé de traitement d'image et support d'enregistrement
JP2017045319A (ja) * 2015-08-27 2017-03-02 沖電気工業株式会社 情報処理装置、情報処理方法、及び情報処理システム
WO2018097058A1 (fr) * 2016-11-22 2018-05-31 日本電気株式会社 Nœud d'analyse, procédé de gestion de ressources, et support d'enregistrement de programme
WO2018150649A1 (fr) * 2017-02-14 2018-08-23 ソニーセミコンダクタソリューションズ株式会社 Appareil électronique, procédé de demande de prise en charge de traitement pour appareil électronique, et système de traitement d'informations
JP6628948B1 (ja) * 2019-01-30 2020-01-15 三菱電機株式会社 モータ駆動装置
JP2020035086A (ja) * 2018-08-28 2020-03-05 富士ゼロックス株式会社 情報処理システム、情報処理装置およびプログラム
JP2021099690A (ja) * 2019-12-23 2021-07-01 横河電機株式会社 装置、システム、方法およびプログラム

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2002165210A (ja) * 2000-08-04 2002-06-07 Matsushita Electric Ind Co Ltd データ送信端末及びデータ送受信装置
JP2010028594A (ja) * 2008-07-23 2010-02-04 Hitachi Ltd 移動物体監視装置
WO2011099320A1 (fr) * 2010-02-12 2011-08-18 株式会社日立製作所 Dispositif de traitement d'informations, et procédé de traitement d'informations sur le dispositif de traitement d'informations

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2002165210A (ja) * 2000-08-04 2002-06-07 Matsushita Electric Ind Co Ltd データ送信端末及びデータ送受信装置
JP2010028594A (ja) * 2008-07-23 2010-02-04 Hitachi Ltd 移動物体監視装置
WO2011099320A1 (fr) * 2010-02-12 2011-08-18 株式会社日立製作所 Dispositif de traitement d'informations, et procédé de traitement d'informations sur le dispositif de traitement d'informations

Cited By (15)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US11100691B2 (en) 2014-03-31 2021-08-24 Nec Corporation Image processing system, image processing method and program, and device
JPWO2015151449A1 (ja) * 2014-03-31 2017-04-13 日本電気株式会社 画像処理システム、画像処理装置、画像処理方法および記録媒体
WO2015151449A1 (fr) * 2014-03-31 2015-10-08 日本電気株式会社 Système de traitement d'image, dispositif de traitement d'image, procédé de traitement d'image et support d'enregistrement
US11798211B2 (en) 2014-03-31 2023-10-24 Nec Corporation Image processing system, image processing method and program, and device
JP2017045319A (ja) * 2015-08-27 2017-03-02 沖電気工業株式会社 情報処理装置、情報処理方法、及び情報処理システム
WO2018097058A1 (fr) * 2016-11-22 2018-05-31 日本電気株式会社 Nœud d'analyse, procédé de gestion de ressources, et support d'enregistrement de programme
JPWO2018097058A1 (ja) * 2016-11-22 2019-10-17 日本電気株式会社 解析ノード、リソース管理方法およびプログラム記録媒体
WO2018150649A1 (fr) * 2017-02-14 2018-08-23 ソニーセミコンダクタソリューションズ株式会社 Appareil électronique, procédé de demande de prise en charge de traitement pour appareil électronique, et système de traitement d'informations
US11561908B2 (en) 2017-02-14 2023-01-24 Sony Semiconductor Solutions Corporation Electronic device, processing proxy requesting method of electronic device, and information processing system
JP2020035086A (ja) * 2018-08-28 2020-03-05 富士ゼロックス株式会社 情報処理システム、情報処理装置およびプログラム
JP6628948B1 (ja) * 2019-01-30 2020-01-15 三菱電機株式会社 モータ駆動装置
WO2020157865A1 (fr) * 2019-01-30 2020-08-06 三菱電機株式会社 Dispositif d'entraînement de moteur
US11373496B2 (en) 2019-12-23 2022-06-28 Yokogawa Electric Corporation Security system having an alert signal generated by sensors or analytics on image data
JP2021099690A (ja) * 2019-12-23 2021-07-01 横河電機株式会社 装置、システム、方法およびプログラム
JP7211353B2 (ja) 2019-12-23 2023-01-24 横河電機株式会社 装置、システム、方法およびプログラム

Also Published As

Publication number Publication date
JPWO2013145530A1 (ja) 2015-12-10

Similar Documents

Publication Publication Date Title
WO2013145530A1 (fr) Système d'analyse
JP6270433B2 (ja) 情報処理装置、情報処理方法、情報処理システム
KR101709715B1 (ko) 카메라를 이용한 사물인터넷 기반의 제어 및 감시장치
US11727578B2 (en) Crowd type classification system, crowd type classification method and storage medium for storing crowd type classification program
EP3249919B1 (fr) Appareil de traitement d'images, procédé de traitement d'images et programme
CN111581436B (zh) 目标识别方法、装置、计算机设备和存储介质
JPWO2013157265A1 (ja) 画像処理システム、受信装置、画像送信装置、画像処理方法、受信側画像処理方法、及び、画像送信方法
KR102054930B1 (ko) 영상 공유 방법 및 이를 위한 장치
KR20210155655A (ko) 이상 온도를 나타내는 객체를 식별하는 방법 및 장치
AU2021104400A4 (en) An intelligent system for detecting behavioral cyber attack on industrial iot using ai federated learning algorithm
US12010470B2 (en) Data processing system, data processing method, sensor apparatus, and receiving apparatus
WO2023045084A1 (fr) Procédé et appareil de commande de dispositifs de station de métro, dispositif électronique et support de stockage
US20240211480A1 (en) Information aggregation in a multi-modal entity-feature graph for intervention prediction
KR102290857B1 (ko) 채널상태정보를 이용한 인공지능 기반의 스마트 사용자 검출 방법 및 장치
JP2022185634A (ja) 端末装置、情報処理装置、情報処理方法、情報処理プログラムおよび情報処理システム
KH et al. Smart CCTV surveillance system for intrusion detection with live streaming
US11043091B2 (en) Method for controlling an image capturing device that captures an image to be displayed or stored by a terminal device from among a plurality of image capturing devices
US11442129B1 (en) Systemic certainty of event convergence
US20240161492A1 (en) Image collection sensor device, unmanned counting edge computing system and method using the same
US20230267741A1 (en) Analysis system and control method for analysis system
US20230169125A1 (en) Information processing system, information processing method, information processing device, and computer program product
US11005996B2 (en) Contact center network and method for establishing a communication session in a contact center network
US20230394836A1 (en) Server device, method for controlling server device, and storage medium
JP7040207B2 (ja) 情報処理装置、通信システム、通信制御方法及びプログラム
JP6525937B2 (ja) 表示制御装置及びプログラム

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 13768187

Country of ref document: EP

Kind code of ref document: A1

ENP Entry into the national phase

Ref document number: 2014507361

Country of ref document: JP

Kind code of ref document: A

NENP Non-entry into the national phase

Ref country code: DE

122 Ep: pct application non-entry in european phase

Ref document number: 13768187

Country of ref document: EP

Kind code of ref document: A1