WO2012140834A1 - Dispositif de traitement d'informations - Google Patents

Dispositif de traitement d'informations Download PDF

Info

Publication number
WO2012140834A1
WO2012140834A1 PCT/JP2012/001806 JP2012001806W WO2012140834A1 WO 2012140834 A1 WO2012140834 A1 WO 2012140834A1 JP 2012001806 W JP2012001806 W JP 2012001806W WO 2012140834 A1 WO2012140834 A1 WO 2012140834A1
Authority
WO
WIPO (PCT)
Prior art keywords
engine
analysis
information
time width
input data
Prior art date
Application number
PCT/JP2012/001806
Other languages
English (en)
Japanese (ja)
Inventor
小山　和也
恒久 河又
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
Priority to JP2013509749A priority Critical patent/JPWO2012140834A1/ja
Publication of WO2012140834A1 publication Critical patent/WO2012140834A1/fr

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04N—PICTORIAL COMMUNICATION, e.g. TELEVISION
    • H04N7/00—Television systems
    • H04N7/18—Closed-circuit television [CCTV] systems, i.e. systems in which the video signal is not broadcast
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V20/00—Scenes; Scene-specific elements
    • G06V20/50—Context or environment of the image
    • G06V20/52—Surveillance or monitoring of activities, e.g. for recognising suspicious objects

Definitions

  • the present invention relates to an information processing apparatus, and more particularly to an information processing apparatus that manages analysis processing performed using an analysis engine.
  • analysis engines that perform analysis processing on various data have been developed.
  • various analysis engines exist such as generating position information for tracing a human flow line from moving image data, specifying a person from still image data, and generating text data from audio data.
  • the analysis engine as described above sets the analysis target as media data such as image data and audio data.
  • analysis processing such as monitoring and search for such media data can be performed by using a related database.
  • a related database Unlike structural data, it is difficult. For example, media data has a large gap with conditions requested by users, and it is difficult to describe conditions and process them.
  • analysis processing such as search for media data may be performed by a method of adding meta information to the media data and performing a search using conditions for the metadata.
  • media data is used as a search condition itself when searching for media data. For example, as shown in Patent Document 1, a search target similar image is searched from a video to be analyzed.
  • each analysis engine has different information such as features used in the analysis process depending on the requirements for the analysis target data. For this reason, in order to respond to various requests, it is necessary to save all input data to the analysis engine and perform analysis processing using necessary input data according to the request. However, if all of the input data is stored so as to correspond to any analysis engine, there arises a problem that the processing load and storage cost according to the request increase.
  • an object of the present invention is to provide an information processing apparatus that improves the versatility of the analysis engine used for the analysis processing, which is the above-described problem, and suppresses the data storage cost.
  • An information processing apparatus An engine information storage unit that stores engine information including time width information that is set in advance for each analysis engine and represents the time width of input data necessary for analysis processing by the analysis engine; An engine control unit that inputs input data to the analysis engine and acquires output data from the analysis engine according to an engine control rule that represents a processing procedure using the analysis engine that can output a result for a predetermined processing request When, Control that the engine control unit holds input data input to the analysis engine according to the engine control rule within a range of time width specified by the time width information included in the engine information of the analysis engine.
  • An accumulation management unit to It has the structure of having.
  • An information processing apparatus An engine information storage unit that stores engine information including time width information that represents a time width of input data necessary for analysis processing by the analysis engine that is set in advance for each analysis engine, An engine control unit that inputs input data to the analysis engine and acquires output data from the analysis engine according to an engine control rule that represents a processing procedure using the analysis engine that can output a result for a processing request.
  • An accumulation rule generating means for setting an accumulation rule that represents a time width of the input data that holds the input data input to the analysis engine is set based on the engine information.
  • the program which is the other form of this invention is: An information processing apparatus including an engine information storage unit that stores engine information including time width information that represents a time width of input data necessary for analysis processing by the analysis engine, which is set in advance for each analysis engine.
  • An engine control unit that inputs input data to the analysis engine and acquires output data from the analysis engine according to an engine control rule that represents a processing procedure using the analysis engine that can output a result for a predetermined processing request
  • Control that the engine control unit holds input data input to the analysis engine according to the engine control rule within a range of time width specified by the time width information included in the engine information of the analysis engine.
  • An accumulation management unit to It is a program for realizing.
  • the program which is the other form of this invention is: An information processing apparatus including an engine information storage unit that stores engine information including time width information that represents a time width of input data necessary for analysis processing by the analysis engine, which is set in advance for each analysis engine.
  • An engine control unit that inputs input data to the analysis engine and acquires output data from the analysis engine according to an engine control rule that represents a processing procedure using the analysis engine that can output a result for a processing request.
  • An accumulation rule generating means for setting an accumulation rule representing a time width of the input data to hold the input data input to the analysis engine based on the engine information; It is a program for realizing.
  • an information processing method includes: In an information processing apparatus including an engine information storage unit that stores engine information including time width information that represents a time width of input data necessary for analysis processing by the analysis engine, which is set in advance for each analysis engine. Input of input data to the analysis engine and acquisition of output data from the analysis engine according to an engine control rule representing a processing procedure using the analysis engine in which the engine control unit can output a result for a predetermined processing request.
  • the accumulation management unit has input data that the engine control unit inputs to the analysis engine in accordance with the engine control rule within a time width range specified by the time width information included in the engine information of the analysis engine. It takes the structure of controlling to hold.
  • an information processing method includes: In an information processing apparatus including an engine information storage unit that stores engine information including time width information that represents a time width of input data necessary for analysis processing by the analysis engine, which is set in advance for each analysis engine.
  • An engine control unit that inputs input data to the analysis engine and acquires output data from the analysis engine according to an engine control rule that represents a processing procedure using the analysis engine that can output a result for a processing request.
  • An accumulation rule that represents a time width of the input data that holds the input data input to the analysis engine is set based on the engine information.
  • the present invention which is configured as described above, can improve the versatility of the analysis engine used for the analysis processing and can suppress the data storage cost.
  • FIG. 1 is a diagram illustrating a configuration of a data management apparatus
  • FIG. 2 is a diagram illustrating an example of engine information
  • FIG. 3 is a diagram illustrating an example of the engine control rule
  • FIG. 4 is a diagram illustrating the operation of the data management device.
  • FIG. 5 is a diagram illustrating another configuration of the data management apparatus.
  • the information processing system includes a plurality of analysis engines 31, 32, 33, an application 4 installed in the information processing apparatus, and data connected to the plurality of analysis engines and the information processing apparatus. And a management device 1.
  • the analysis engines 31, 32, and 33 are information processing apparatuses that perform analysis processing on analysis target data. For example, media data such as moving image data, still image data, and audio data is analyzed as analysis target data. I do. For example, a person extraction engine that extracts a person image showing a person from image data, a face recognition engine that recognizes a person's face from image data or person image data, and a position where human flow lines are traced from image data There is a flow line extraction engine that generates information.
  • the analysis engines 31, 32, and 33 are not limited to those that perform the above-described analysis processing, and the number thereof is not limited to three as shown in FIG.
  • the application 4 is implemented in an information processing apparatus connected to the data management apparatus 1, and has a function of requesting the data management apparatus 1 for an analysis result using the analysis processing by the analysis engines 31, 32, and 33 described above. Have.
  • the processing request requested from the application 4 to the data management apparatus 1 is a monitoring request for searching or monitoring a specific person from video data that is analysis target data, for example.
  • the application 4 acquires the results analyzed by the analysis engines 31, 32, and 33 from the data management device 1 in response to the processing request described above.
  • the data management device 1 is an information processing device including an arithmetic device and a storage device. As shown in FIG. 1, the data management device 1 is constructed by incorporating a program into an arithmetic device, and includes a monitoring request reception unit 11, an engine control unit 12, an engine control rule management unit 13, and an analysis data management unit. 14, an analysis data monitoring unit 15, a similarity determination rule management unit 16, a monitoring result transmission unit 17, a media data management unit 18, and a rule generation unit 20.
  • the rule generation unit 20 includes an engine information management unit 21, an engine control rule generation unit 22, an accumulation rule generation unit 23, and a monitoring rule generation unit 24.
  • each configuration will be described in detail.
  • the engine information management unit 21 stores and manages engine information preset for each analysis engine in a storage device.
  • the engine information includes an input data format (characteristic information) that represents the data format of input data that can be input to the analysis engine, an output data format (characteristic information) that represents the data format of output data that is an analysis result by the analysis engine, and And time width information representing the time width of the input data required for the analysis processing by the analysis engine.
  • the “input data format” is, for example, information indicating the data type of the input data to be analyzed by the analysis engine, and has values such as “video” and “person video” as an example.
  • the “output data type” is information representing the data type of output data which is an analysis result by the analysis engine, for example.
  • the analysis engine analyzes the gender of the person shown in the video data, the value is “gender”, or the person that the analysis engine shows in the video data In the case of analyzing the feature amount of the face, the value is “face feature amount”.
  • time width information is a value that represents the amount of data along the time series of input data necessary for performing analysis processing by the analysis engine.
  • this time width information when the analysis target data is video data, it is “number of frames” or “video time”.
  • Fig. 2 shows an example of engine information.
  • the analysis engine is an “image face recognition engine” for recognizing a human face from image data
  • the input data format is “image”
  • the output data format is “face feature amount”
  • the input data time width is “1 frame”.
  • the analysis engine is a “video face recognition engine” for recognizing a human face from video data
  • the input data format is “human video”
  • the output data format is “face feature value”
  • the input data time width is “3”.
  • the value is “frame”.
  • the analysis engine is a “person extraction engine” that extracts a person from video data
  • the input data format is “video”
  • the output data format is “person video”
  • the input data time width is “5 frames”. Value.
  • the analysis engine is a “flow line extraction engine” that recognizes the flow line of a person from video data
  • the input data format is “human video”
  • the output data format is “human flow line”
  • the input data time width is “ The value is “
  • the engine information described above may include data representing other characteristics of the analysis engine for each analysis engine, and may be information that can be referred to when generating an engine control rule, as will be described later.
  • the engine control rule generation unit 22 (engine control rule generation means) performs an analysis process corresponding to the process request received from the application 4 and outputs a result corresponding to the process request.
  • An engine control rule representing a processing procedure using 32 and 33 is generated.
  • the engine control rule generation unit 22 firstly, from the content of the processing request, necessary input data, output data representing the requested result itself, or output data necessary for deriving the requested result, Is identified. Then, an analysis engine in which the data format of the specified input data and output data is set is selected with reference to the engine information described above.
  • the engine control rule generation unit 22 selects one analysis engine that can input the input data specified from the processing request, and another analysis engine that uses the output data of this one analysis engine as input data. Is selected.
  • the output data from the other analysis engine only needs to be output data required in response to the processing request. If not, the output data from the other analysis engine is further input to the input data.
  • Another analysis engine is selected. Then, a combination of a plurality of analysis engines that can finally obtain output data corresponding to the processing request is selected, and an engine control rule that is a processing procedure connecting these plurality of analysis engines is generated.
  • the monitoring request which is a processing request from the application 4
  • the monitoring request has the content that “if a person with a face similar to the input photo appears in the monitoring video, the action history of that person is notified”.
  • the similarity determination between the face of the person in the monitoring video and the face of the person in the input photograph is performed by the analysis data monitoring unit 15 without using an analysis engine as will be described later.
  • the engine control rule corresponding to the above (A) specifies “image” as the input data format because the input data is “input photograph”, and as the output data format because the output data is “person's face”.
  • the “face feature amount” is specified.
  • the analysis engine in which the input data format and the output data format are set is selected from the engine information as shown in FIG. Thereby, an engine control rule of a processing procedure using the “image face recognition engine” as shown in FIG.
  • the engine control rule corresponding to the above (B) specifies “video” as the input data format because the input data is “monitoring video”.
  • the output data is “person's face” and “person's flow line”
  • two output data formats, “face feature value” and “person flow line” are specified.
  • the analysis engine in which the input data format and the output data format are set is selected from the engine information as shown in FIG. 2, and from the “video” as the input data format to the “face feature value” as the output data format.
  • the engine control rule described above is not necessarily generated by the data management device 1 (engine control rule generation unit 22).
  • An engine control rule that is manually generated in advance may be registered in the data management device 1.
  • the registered engine control rule is selected by the engine control unit 12 and the engine control rule management unit 13 as will be described later, and the process for the analysis engine 3 is performed according to the engine control rule.
  • the accumulation rule generation unit 23 sets an accumulation rule that represents the time width of input data required when each analysis engine described above performs analysis processing. Specifically, the accumulation rule generation unit 23 inputs the time width specified by the time width information set for each analysis engine included in the engine information to the analysis engine set in the engine control rule. Set as time span. At this time, when one input data is input to a plurality of analysis engines based on the engine control rule described above, the accumulation rule generation unit 23 sets each time width set for each analysis engine based on the engine information. Among these, the value of the longest time width is set as the accumulation rule.
  • the analysis engine to be used is one of “image face recognition engine”.
  • “1 frame”, which is the time width set in this analysis engine is set as the accumulation rule.
  • the analysis engine to be used is one of the “person extraction engine”, and therefore, with reference to the engine information of FIG. A certain “5 frames” is set as an accumulation rule.
  • image face recognition engine and “flow line extraction engine”
  • each of these analysis engines is referred to with reference to the engine information of FIG. “10 frames”, which is the maximum value of “3 frames” and “10 frames”, which are the time widths set in FIG.
  • the monitoring rule generation unit 24 generates a rule for outputting a result corresponding to the monitoring request that is a processing request from the application 4 from the data analyzed by the analysis engine. For example, in the case where the monitoring request has the above-described content, “If a person with a face similar to the input photo appears in the monitoring video, the person's action history is notified”, the analysis shown in FIG. A face similar to the “human face in the input photo” that is the analysis processing result shown in FIG. 3A is detected from the “human face in the monitoring video” that is the processing result, and the above-described figure of the detected person A monitoring rule for notifying “person flow line”, which is the analysis processing result shown in FIG.
  • the monitoring request receiving unit 11 receives a monitoring request that is a processing request input from the application 4 and passes it to the rule generating unit 20.
  • the engine control unit 12 and the engine control rule management unit 13 input input data and analysis results to the analysis engines 31, 32, and 33 in accordance with the engine control rules generated by the engine control rule generation unit 22 described above. Is obtained and output to the analysis data management unit 14 or the like.
  • engine control rules are registered in advance, the registered engine control rules are selected, and processing for each analysis engine 3 is performed according to the engine control rules.
  • the analysis data management unit 14 (accumulation management unit) manages input data input to each analysis engine 31, 32, 33 and output output data via the engine control 12 described above.
  • the analysis data management unit 14 inputs the input data input to each analysis engine 31, 32, 33 within the time range set by the accumulation rule generated by the accumulation rule generation unit 23 described above.
  • control is performed so that the data is stored in the storage area installed in the data management device 1.
  • the accumulation of input data will be described with a specific example.
  • the accumulation rule for input data input to the analysis engine is set to “3 frames”.
  • image data d0, d1, d2, and d3 to be input to the analysis engine arrive in the order of times t0, t1, t2, and t3.
  • the analysis processing is executed in three frames of the image data d0, d1, d2 together with the stored image data d0, d1, and the processed image data d2 is Accumulated according to the accumulation rule.
  • the image data d2 is added, the number of accumulated data (the number of frames) reaches “3”, so the old image data d0 is discarded (accumulated data: d1, d2).
  • recognition is performed on the three frames of the image data d1, d2, d3 together with the stored image data d1, d2, and the processed image data d3 is stored. Accumulated by rules.
  • the image data d3 is added, the number of accumulated data (number of frames) reaches “3”, so the old image data d1 is discarded (accumulated data: d2, d3).
  • the analysis data management unit 14 accumulates input data to the analysis engine based on the accumulation rule generated by the accumulation rule generation unit 23, but analyzes the engine information as shown in FIG.
  • the data management unit 14 may directly refer to the information on the time width set for each analysis engine, and the input data may be accumulated as described above based on the time width.
  • the analysis data monitoring unit 15 monitors the output data from each analysis engine 31, 32, 33 in accordance with the monitoring rules generated by the monitoring rule generation unit 24 described above. For example, in the above-described example, a monitoring process is performed in which a face similar to “a person's face in the input photo” is detected from “a person's face in the monitoring video” and a “person flow line” of the detected person is notified. . At this time, the process of detecting a face similar to the “face of the person in the input photo” from the “face of the person in the monitoring video” compares each “face feature amount” output from each analysis engine and determines the similarity. It is determined whether or not they are similar according to a similarity determination rule set in advance in the rule management unit 16, and a face similar to the face of the input photo is detected from the monitoring video.
  • the analysis data monitoring unit 15 outputs a result corresponding to the monitoring request from the application 4 to the monitoring result transmission unit 17. Accordingly, the monitoring result transmission unit 17 transmits the monitoring result output from the analysis data monitoring unit 15 to the application 4.
  • the media data management unit 18 performs management such as inputting a monitoring image to be monitored and an input image input together with a monitoring request from the application 4 to each analysis engine via the engine control unit 12.
  • step S ⁇ b> 1 the data management apparatus 1 receives it at the monitoring request reception unit 11 (step S ⁇ b> 1) and passes it to the rule generation unit 20.
  • the rule generation unit 20 generates an engine control rule that is a processing procedure capable of outputting a result corresponding to the received monitoring request with reference to the engine information in the engine control rule generation unit 22 (step S2). .
  • the accumulation rule generation unit 23 refers to the engine information, and generates an accumulation rule for input data to each analysis engine included in the generated engine control rule (step S3).
  • the monitoring rule generation unit 24 generates a monitoring rule that outputs a monitoring result corresponding to the received monitoring request, using the analysis result of the analysis engine (step S4).
  • the data management apparatus 1 performs analysis processing on the monitoring video according to the engine control rule and the accumulation rule, and performs monitoring processing on the analysis result according to the monitoring rule (step S4). That is, the data management device 1 inputs input data to a specific analysis engine according to the processing procedure represented by the engine control rule, and receives the analysis result. At this time, the data management device 1 controls the amount of input data retained for each analysis engine according to the accumulation rule, and does not retain input data exceeding the range set in the accumulation rule. Then, the monitoring result output by the analysis and monitoring process is output to the application 4.
  • the data management device 1 includes the engine control unit 12, the analysis data management unit 14 and the like, and performs analysis processing and monitoring processing on the monitoring video according to the generated engine control rules, accumulation rules, and the like.
  • Functions for performing analysis processing and management processing, such as the engine control unit 12 and the analysis data management unit 14, may be provided in another information processing apparatus. That is, the data management device 1 may have only the rule generation unit 20 and may be equipped with only a function for generating engine control rules, accumulation rules, and monitoring rules.
  • the application 4 does not need to understand how to use each analysis engine, and can obtain a desired monitoring result only by transmitting a processing request. It is possible to improve the general versatility of the system using and the convenience of the user. At this time, the storage of input data suitable for each analysis engine to be used is realized, so that the data storage cost can be suppressed.
  • the data management device 1 according to the present invention may be configured as shown in FIG.
  • the data management device 1 shown in FIG. 5 includes a monitoring state management unit 19 constructed by reading a program into the equipped arithmetic device, and other configurations are the same as those in the first embodiment.
  • the monitoring state management unit 19 changes the engine control rule, the accumulation rule, and the monitoring rule to be used according to the monitoring result by the analysis data monitoring unit 15. Thereby, the contents to be monitored can be changed according to the change in the status of the monitoring video.
  • the monitoring request is the above-mentioned content “if a person with a face similar to the input photo appears in the monitoring video, the subsequent action history of the person is notified” (1) “person in the input photo”
  • the “person extraction engine” analyzes the “monitoring video” and outputs the “person video”.
  • the “person video” is input to the “video face recognition engine” and the “facial features”
  • An analysis process is performed according to an engine control rule that “amount” is output, and a specific person is detected.
  • the “person image extraction engine” analyzes the “monitoring video” as the second stage and outputs “person video”.
  • the analysis process is performed by changing the engine control rule that “video” is input to “flow line extraction engine” and “human flow line” is output, and the “human flow line” that is the analysis result is notified.
  • the analysis engine to be used is different in accordance with the change of the engine control rule, it is changed to the accumulation rule and the monitoring rule corresponding to each.
  • the present invention can cope with analysis processing expressed by various engine control rules, and can improve versatility.
  • An engine information storage unit 101 that stores engine information including time width information that represents a time width of input data necessary for analysis processing by the analysis engine, which is set in advance for each analysis engine;
  • An engine control unit that inputs input data to the analysis engine and acquires output data from the analysis engine according to an engine control rule that represents a processing procedure using the analysis engine that can output a result for a predetermined processing request 102, Control that the engine control unit holds input data input to the analysis engine according to the engine control rule within a range of time width specified by the time width information included in the engine information of the analysis engine.
  • a storage management unit 103 for An information processing apparatus 100 comprising:
  • (Appendix 2) An information processing apparatus according to attachment 1, wherein When the engine control rule includes a processing procedure in which output data from one analysis engine is input data to another analysis engine, the accumulation management unit outputs the output data from the one analysis engine to the other analysis. As input data of the engine, control is performed so as to be held within the range of the time width specified by the time width information included in the engine information of the other analysis engine. Information processing device.
  • Appendix 3 An information processing apparatus according to appendix 1 or 2,
  • the accumulation management unit when the engine control rule includes a processing procedure for inputting the same input data to the plurality of analysis engines, each time included in each engine information corresponding to each of the plurality of analysis engines. Controlling to hold input data to be input to the plurality of analysis engines within the range of the longest time width among the time widths specified by the width information, Information processing device.
  • Appendix 4 An information processing apparatus according to any one of appendices 1 to 3, Based on the engine information, comprising an accumulation rule generating means for setting an accumulation rule that represents a time width of input data required for the analysis processing performed by the analysis engine included in the processing procedure represented by the engine control rule, The accumulation management unit controls to hold input data to be input to the analysis engine within a time range set by the accumulation rule. Information processing device.
  • An information processing apparatus comprising an accumulation rule generating means for setting an accumulation rule that represents a time width of input data required for the analysis processing performed by the analysis engine included in the processing procedure represented by the engine control rule,
  • the accumulation rule generating means when the engine control rule includes a processing procedure for inputting the same input data to the plurality of analysis engines, each of the engine information corresponding to each of the plurality of analysis engines Of each time width specified by the time width information, the longest time width is set as the accumulation rule,
  • the accumulation management unit controls to hold input data to be input to the analysis engine within a time range set by the accumulation rule.
  • Information processing device comprising an accumulation rule generating means for setting an accumulation rule that represents a time width of input data required for the analysis processing performed by the analysis engine included in the processing procedure represented by the engine control rule.
  • the engine information stored in the engine information storage unit includes input data characteristic information that can be input to the analysis engine, and output data characteristic information that is an analysis result by the analysis engine, Based on the engine information, the engine control rule generating means for selecting the analysis engine capable of outputting the result corresponding to the processing request and generating the engine control rule representing the processing procedure using the analysis engine.
  • Information processing device includes input data characteristic information that can be input to the analysis engine, and output data characteristic information that is an analysis result by the analysis engine.
  • the engine control unit changes the engine control rule to another engine control rule based on output data acquired from the analysis engine according to the engine control rule, and inputs data to the analysis engine according to the other engine control rule. Input and acquisition of output data from the analysis engine,
  • the accumulation management unit has input data to be input to the analysis engine according to the changed other engine control rule within a time width range specified by time width information included in the engine information of the analysis engine. Control to hold in, Information processing device.
  • the engine information storage unit 201 stores engine information including time width information that is set in advance for each analysis engine and represents the time width of input data necessary for analysis processing by the analysis engine.
  • An engine control unit that inputs input data to the analysis engine and acquires output data from the analysis engine according to an engine control rule that represents a processing procedure using the analysis engine that can output a result for a processing request.
  • An accumulation rule generating unit 202 configured to set an accumulation rule representing a time width of the input data holding the input data input to the analysis engine based on the engine information; Information processing apparatus 200.
  • Appendix 9 An information processing apparatus according to appendix 8, wherein The accumulation rule generating means, when the engine control rule includes a processing procedure for inputting the same input data to the plurality of analysis engines, each of the engine information corresponding to each of the plurality of analysis engines Of each time width specified by the time width information, the longest time width is set as the accumulation rule.
  • Information processing device when the engine control rule includes a processing procedure for inputting the same input data to the plurality of analysis engines, each of the engine information corresponding to each of the plurality of analysis engines Of each time width specified by the time width information, the longest time width is set as the accumulation rule.
  • An information processing apparatus including an engine information storage unit that stores engine information including time width information that represents a time width of input data necessary for analysis processing by the analysis engine, which is set in advance for each analysis engine.
  • An engine control unit that inputs input data to the analysis engine and acquires output data from the analysis engine according to an engine control rule that represents a processing procedure using the analysis engine that can output a result for a predetermined processing request When, Control that the engine control unit holds input data input to the analysis engine according to the engine control rule within a range of time width specified by the time width information included in the engine information of the analysis engine.
  • An information processing apparatus including an engine information storage unit that stores engine information including time width information that represents a time width of input data necessary for analysis processing by the analysis engine, which is set in advance for each analysis engine.
  • An engine control unit that inputs input data to the analysis engine and acquires output data from the analysis engine according to an engine control rule that represents a processing procedure using the analysis engine that can output a result for a processing request.
  • An accumulation rule generating means for setting an accumulation rule representing a time width of the input data to hold the input data input to the analysis engine based on the engine information;
  • an information processing apparatus including an engine information storage unit that stores engine information including time width information that represents a time width of input data necessary for analysis processing by the analysis engine, which is set in advance for each analysis engine.
  • the accumulation management unit has input data that the engine control unit inputs to the analysis engine in accordance with the engine control rule within a time width range specified by the time width information included in the engine information of the analysis engine.
  • an information processing apparatus including an engine information storage unit that stores engine information including time width information that represents a time width of input data necessary for analysis processing by the analysis engine, which is set in advance for each analysis engine.
  • An engine control unit that inputs input data to the analysis engine and acquires output data from the analysis engine according to an engine control rule that represents a processing procedure using the analysis engine that can output a result for a processing request.
  • An accumulation rule that represents a time width of the input data that holds the input data input to the analysis engine is set based on the engine information.
  • the 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)
  • Multimedia (AREA)
  • Signal Processing (AREA)
  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)
  • Image Analysis (AREA)

Abstract

Le présent dispositif de traitement d'informations comporte : une unité de mémoire d'informations de moteur servant à mémoriser les informations de moteur pour les moteurs d'analyse individuels, les informations de moteur comprenant des informations de largeur temporelle représentant la largeur temporelle de données d'entrée nécessaires pour un processus d'analyse par le moteur d'analyse ; un contrôleur de moteur servant à entrer des données d'entrée dans le moteur d'analyse et à acquérir des données de sortie du moteur d'analyse en fonction d'une règle de contrôle de moteur représentant une procédure de traitement dans laquelle un moteur d'analyse capable de produire un résultat par rapport à une requête de traitement prédéterminée est utilisé ; et une unité de gestion d'accumulation servant à contrôler les données d'entrée, qui sont entrées dans le moteur d'analyse par le contrôleur de moteur en fonction de la règle de contrôle de moteur, de manière à ce qu'elles soient maintenues dans une plage de largeur temporelle spécifiée par les informations de largeur temporelle comprises dans les informations de moteur pour le moteur d'analyse.
PCT/JP2012/001806 2011-04-11 2012-03-15 Dispositif de traitement d'informations WO2012140834A1 (fr)

Priority Applications (1)

Application Number Priority Date Filing Date Title
JP2013509749A JPWO2012140834A1 (ja) 2011-04-11 2012-03-15 情報処理装置

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
JP2011-086940 2011-04-11
JP2011086940 2011-04-11

Publications (1)

Publication Number Publication Date
WO2012140834A1 true WO2012140834A1 (fr) 2012-10-18

Family

ID=47009029

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/JP2012/001806 WO2012140834A1 (fr) 2011-04-11 2012-03-15 Dispositif de traitement d'informations

Country Status (2)

Country Link
JP (1) JPWO2012140834A1 (fr)
WO (1) WO2012140834A1 (fr)

Cited By (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2019103087A1 (fr) * 2017-11-27 2019-05-31 日本電気株式会社 Dispositif de traitement d'informations, procédé de traitement d'informations et support d'enregistrement

Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2006053167A (ja) * 2004-07-14 2006-02-23 Rarugo:Kk 電子楽器装置および仮想楽器の演奏方法並びに仮想楽器の演奏処理のプログラムおよびそのプログラムを記録した記録媒体
JP2006133937A (ja) * 2004-11-04 2006-05-25 Fuji Xerox Co Ltd 動作識別装置
JP2007124526A (ja) * 2005-10-31 2007-05-17 Hitachi Ltd 画像監視記録装置及び画像監視記録システム
JP2011070576A (ja) * 2009-09-28 2011-04-07 Toshiba Corp 画像処理装置、及び画像処理方法

Patent Citations (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2006053167A (ja) * 2004-07-14 2006-02-23 Rarugo:Kk 電子楽器装置および仮想楽器の演奏方法並びに仮想楽器の演奏処理のプログラムおよびそのプログラムを記録した記録媒体
JP2006133937A (ja) * 2004-11-04 2006-05-25 Fuji Xerox Co Ltd 動作識別装置
JP2007124526A (ja) * 2005-10-31 2007-05-17 Hitachi Ltd 画像監視記録装置及び画像監視記録システム
JP2011070576A (ja) * 2009-09-28 2011-04-07 Toshiba Corp 画像処理装置、及び画像処理方法

Cited By (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
WO2019103087A1 (fr) * 2017-11-27 2019-05-31 日本電気株式会社 Dispositif de traitement d'informations, procédé de traitement d'informations et support d'enregistrement
JPWO2019103087A1 (ja) * 2017-11-27 2020-09-24 日本電気株式会社 情報処理装置、情報処理方法およびプログラム
US11195025B2 (en) 2017-11-27 2021-12-07 Nec Corporation Information processing device, information processing method, and storage medium for temporally dividing time-series data for analysis

Also Published As

Publication number Publication date
JPWO2012140834A1 (ja) 2014-07-28

Similar Documents

Publication Publication Date Title
US9384197B2 (en) Automatic discovery of metadata
RU2701508C1 (ru) Способ и система рекомендаций контента на основании информации поведения пользователя
JP2016076073A (ja) データ処理装置、データ処理方法、及び、コンピュータ・プログラム
US20140379346A1 (en) Video analysis based language model adaptation
US9380257B2 (en) Portable electronic equipment and method of recording media using a portable electronic equipment
WO2019137391A1 (fr) Procédé et appareil pour réaliser une mise en correspondance catégorisée de vidéos, et moteur de sélection
JP2009536390A (ja) コンテンツに注釈付けする装置及び方法
US11537636B2 (en) System and method for using multimedia content as search queries
JP6142533B2 (ja) 情報処理装置
US20130191368A1 (en) System and method for using multimedia content as search queries
KR20100018070A (ko) 멀티미디어 파일의 개요들을 자동으로 생성하기 위한 방법 및 장치
US20150030260A1 (en) Analysis system
JP6075287B2 (ja) 解析エンジン制御装置
WO2012140834A1 (fr) Dispositif de traitement d'informations
TW201435627A (zh) 搜索優化系統及方法
US11640426B1 (en) Background audio identification for query disambiguation
US20220188304A1 (en) Method and system for handling query in iot network
KR20190031829A (ko) 전자 장치 및 그 제어 방법
JP5672307B2 (ja) 情報処理装置
US20150066840A1 (en) Analysis system
KR102568143B1 (ko) 맞춤형 서비스 모드를 제공하는 방법 및 단말
JP2016535341A (ja) メディアコンテンツ順序付けシステムおよびメディアコンテンツを順序付けするための方法
JP2014002336A (ja) コンテンツ処理装置、コンテンツ処理方法、およびコンピュータプログラム
KR20190076621A (ko) 전자 장치 및 전자 장치에서 방송 콘텐트와 관련된 서비스 정보 제공 방법
US20150036880A1 (en) Analysis system

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: 12771607

Country of ref document: EP

Kind code of ref document: A1

ENP Entry into the national phase

Ref document number: 2013509749

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: 12771607

Country of ref document: EP

Kind code of ref document: A1