JP2020107171A - Action pattern analysis support device, action pattern analysis support system, and action pattern analysis support method - Google Patents

Abstract

To allow a user to add observation data of a field to field data acquired from the field by an existing sensor and perform more detailed action pattern analysis.SOLUTION: An action pattern analysis support system 1 includes: a field data generation unit 20 that acquires and collects measured data from a sensor 10 to generate field data 110; a pattern analysis support unit 40 that expands and analyzes the field data 110; and a data storage unit 30 that stores the data generated by the field data generation unit 20 or the pattern analysis support unit 40. The pattern analysis support unit 40 then includes: an observation data input unit 41 that inputs observation data 140 based on observation of a field by a user; a data integration unit 42 that generates integrated data 120 by integrating the observation data 140 with the field data 110; and a data presentation unit 44 that presents the integrated data 120 to the user.SELECTED DRAWING: Figure 1

Description

The present invention relates to an action pattern analysis support device, an action pattern analysis support system, and an action pattern analysis support method, and an action pattern analysis support that assists in analyzing an action pattern in data analysis in space involving detection and tracking of a moving object. It is suitable for application to a device, an action pattern analysis support system, and an action pattern analysis support method.

Conventionally, as a method of detecting a moving object such as a person or a vehicle, a method of measuring using a surveillance camera or a sensor has been proposed. For example, when a surveillance camera is used, the position and trajectory of a moving object are calculated by image recognition, and this is utilized for security and monitoring of the state of a target area. An image monitoring system that has the function of performing recognition processing on images acquired from imaging devices such as cameras and detecting moving objects that appear in the monitoring area records only the images in which moving objects appear by using the detection results. It has a function of displaying, a function of presenting a warning icon on the display device, a function of outputting a warning such as a buzzer, and calling attention to the surveillance center. Such a video surveillance system not only helps to reduce the burden of surveillance work that used to always require confirmation work, but also records when criminal acts such as theft or fraudulent acts occur. It is also possible to use the images for post-crime investigations.

In addition, sensing technology is often used in various fields (spaces) along with recognition based on surveillance cameras as described above. Specifically, for example, it is used not only for crime prevention but also for marketing purposes and public It is also used to improve the convenience of space usage. In addition, the target area is not limited to buildings and public facilities, and its use is expanding to include city blocks. For example, in urban development, there is an increasing need for understanding the situation in the field.

However, in order to understand the situation of the field described above, it is essential to improve the detection accuracy and the like, and attempts are being made to deepen understanding of behavior in a specific area by utilizing a new sensor. For example, Patent Literature 1 discloses a behavior analysis system that analyzes behaviors of consumers and protects personal information of consumers. The behavior analysis system of Patent Literature 1 utilizes a characteristic sensor such as a three-dimensional sensor (described as a distance image sensor in Patent Literature 1) in order to detect detailed behavior without disposing a plurality of sensors. , Not only image data but also distance data is used to acquire the behavior history of the consumer. More specifically, the distance image sensor is used to capture an image of the consumer, and the result is output from the distance image sensor. Dimension of distance data included in the distance data is analyzed to recognize the consumer's area in front of the product shelf, and the recognized area is tracked to recognize the consumer's behavior in front of the product shelf. ..

JP, 2010-067079, A

However, the conventional technique disclosed in Patent Document 1 has the following problems when an attempt is made to improve understanding accuracy in a specific area (field). First, since there is a limit to improvement in sensor accuracy, the actions that can be detected are limited. In addition, there is a limit to the detection itself by the sensor that there is a situation in which the image is not captured by the sensor, or it is not possible to capture the internal situation of the subject or a minute change in behavior. Furthermore, in the above-described conventional technique, the situation and the behavior pattern may differ depending on the field to which the application is applied, so that it is necessary to adjust the parameters and the like each time according to the field to which the application is applied. As a result, the measurement was limited to the detectable range, and it was difficult to gain a deep understanding of the field.

The present invention has been made in consideration of the above points, and enables observation data of a field to be added by a user to the field data acquired from the field by an existing sensor so that a more precise behavior pattern analysis can be performed. It is intended to propose an action pattern analysis support device, an action pattern analysis support system, and an action pattern analysis support method capable of supporting.

In order to solve such a problem, the present invention provides the following action pattern analysis support device that supports the analysis of the action pattern of a moving object. This behavior pattern analysis support device includes a field data generation unit that acquires and collects measurement data from sensors to generate field data, and a pattern analysis support unit that expands and analyzes the field data. The section presents an observation data input section for inputting observation data based on field observation by a user, a data integration section for generating integrated data in which the observation data is integrated with the field data, and presents the integrated data to the user side. And a data presentation unit.

Further, in order to solve such a problem, the present invention provides the following action pattern analysis support system that supports the analysis of the action pattern of a moving object. This behavior pattern analysis support system includes a field data generation unit that acquires and aggregates measurement data from sensors to generate field data, a pattern analysis support unit that expands and analyzes the field data, the field data generation unit, or the field data generation unit. A data storage unit that stores the data generated by the pattern analysis support unit; and the pattern analysis support unit further includes an observation data input unit that inputs observation data based on a user's observation of a field, and the field data. And a data presenting unit that presents the integrated data to the user side.

Further, in order to solve such a problem, in the present invention, there is provided a behavior pattern analysis support method for supporting analysis of a behavior pattern of a moving object, which is field data generation for acquiring and collecting measurement data by a sensor to generate field data. A step, an observation data input step capable of inputting observation data based on observation of a field by a user, the field data generated in the field data generation step and the observation data input in the observation data input step are integrated. There is provided a behavior pattern analysis support method including a data integration step of generating integrated data by performing the above-described operation, and a data presentation step of presenting the integrated data generated in the data integration step to a user side.

According to the present invention, observation data of a field can be added by a user to field data acquired from a field by an existing sensor, and it is possible to support more sophisticated behavior pattern analysis.

It is a block diagram which shows the structural example of the action pattern analysis support system which concerns on one embodiment of this invention. It is a figure for demonstrating the whole process of action pattern analysis support. It is a figure which shows an example of the user interface for data input/output. It is a figure which shows an example of the screen which inputs observation information. It is a figure which shows another example of the screen which inputs observation information. It is a figure which shows an example of the display screen of presentation data. It is a figure which shows another example of the display screen of presentation data. It is a figure which shows an example of the input screen which changes the parameter of an analysis guideline.

Hereinafter, an embodiment of the present invention will be described in detail with reference to the drawings.

INDUSTRIAL APPLICABILITY The present invention integrates, in addition to measurement data, observation data of a person, thereby realizing an action pattern analysis support device, an action pattern analysis support system, and an action pattern analysis support device that realizes improvement in analysis efficiency and accuracy due to information that cannot be obtained by a sensor. As an embodiment of the analysis support method, in the following, in a space such as a shopping mall, data measured by a sensor such as a camera is used to efficiently and precisely acquire a person's action pattern. A detailed description will be given of a behavior pattern analysis support system that supports behavior pattern analysis for.

(1) Configuration of Action Pattern Analysis Support System FIG. 1 is a block diagram showing a configuration example of an action pattern analysis support system according to an embodiment of the present invention. The action pattern analysis support system 1 shown in FIG. 1 includes a field data generation unit 20 that generates field data based on imaged data obtained by a sensor 10 measuring a person's action, and a predetermined type of data used in the action pattern analysis support system. A data storage unit 30 that stores (field data, integrated data, analysis data), and a pattern analysis support unit 40 that expands the field data and corrects parameters related to the acquisition of the field data are configured.

The sensor 10 is a device that records the behavior of a person in the image capturing range in image capturing data (may be image data, image data, or data of another predetermined format), and various devices can be applied. For example, when the sensor 10 uses an image sensor, the sensor 10 includes a lens, a camera sensor, and the like, and the image sensor can acquire a predetermined image. The image sensor is a mechanism including an image pickup device (neither is shown) such as a CMOS (Complementary Metal Oxide Semiconductor) or a CCD (Charge Coupled Device). By using the imaging data acquired from such a sensor 10, the situation of a person or the like in the environment can be sensed. Further, as the sensor 10 in the present embodiment, not only an image sensor such as a camera but also various approaches such as utilization of a laser, a beacon sensor, and a wireless access point are possible, and the type of the sensor 10 can be changed according to the application. Field information can be obtained while using them properly. In this example, unless otherwise stated, it is assumed that the technology utilizing images and videos is used.

The field data generation unit 20 has a function of generating field data 110 by utilizing a known image recognition technique for the imaged data acquired from the sensor 10. Specifically, as shown in FIG. 1, the field data generation unit 20 includes a field data acquisition unit 21 and a data aggregation unit 22.

The field data acquisition unit 21 utilizes a known person detection technology or person tracking technology to acquire the movement trajectory of the person from the imaged data of the sensor 10 based on a predetermined analysis guideline. For example, if the positional correspondence between the sensor 10 and the imaged data can be recognized, the positional information of the person can be acquired. Furthermore, by acquiring and superimposing map information and the like, it is possible to acquire the behavior of a person in association with a field in the real world. These methods have been widely researched and developed so far, and general techniques thereof may be used. Then, the field data acquisition unit 21 collects the field data (also referred to as primary field data) indicating the person's action trajectory and the like obtained by performing the above-described technical processing on the imaged data of the sensor 10, the data aggregation unit 22. Send to.

The data aggregating unit 22 aggregates the primary field data transmitted from the field data acquiring unit 21 in a predetermined unit such as a certain time or a specific area to perform statistical processing on the field data (also referred to as secondary field data). Generated). Specifically, as secondary field data, for example, if primary fields are aggregated at regular time intervals and subjected to statistical processing, statistical information indicating the flow rate (personal flow) of a person at a specific time and the gender ratio can be obtained. Can be done. Also, for example, if the map is divided into specific grids and the primary field data is aggregated in grid units and subjected to statistical processing, the trajectory on the map can be traced, and in which areas many people have acted, etc. It is possible to obtain statistical information that can identify the As described above, the secondary field data generated by the data aggregating unit 22 can be acquired from any viewpoint by using the acquirable primary field data.

Note that the statistical processing in the aggregation of the primary field data by the data aggregating unit 22 is a simple and routine analysis by a predefined analysis process, and a detailed analysis by the data analyzing unit 43 (second The analysis can be referred to as the first analysis. Since the first analysis is a predefined analysis process, it has a feature that it can be processed in real time.

Then, the data aggregating unit 22 transmits the field data (secondary field data) obtained by aggregating as described above to the data storage unit 30, and the field data that is one of the storage areas inside the data storage unit 30. It is stored in the storage unit 31. In the following description, the secondary field data will be referred to as field data 110 unless otherwise specified. By using this notation, it can be said that the data aggregating unit 22 transmits the generated field data 110 to the data storage unit 30 and stores it in the field data storage unit 31. The data aggregating unit 22 also transmits the generated field data 110 to the pattern analysis supporting unit 40.

Further, as will be described later in detail in the pattern analysis support unit 40, the field data generation unit 20 outputs the “analysis guideline” used when generating the field data 110 from the parameter change unit 46 of the pattern analysis support unit 40. It can be corrected according to the correction parameter 150 that is set. This modifiable “analysis guideline” may be, for example, the “analysis guideline” used when the field data acquisition unit 21 generates the primary field data from the imaged data of the sensor 10, or The “aggregation guideline” may be used when the data aggregating unit 22 aggregates the primary field data to generate the secondary field data. From these things, the “analysis guideline” can be said to be a parameter for acquiring the field data 110.

The data storage unit 30 has a storage function of being connected to the field data generation unit 20 and the pattern analysis support unit 40 and storing predetermined data, and is realized by, for example, a server. In FIG. 1, the data storage unit 30 includes a field data storage unit 31 for storing field data 110, an integrated data storage unit 32 for storing integrated data 120, and an analysis data for storing analysis data 130, depending on the type of data to be stored. Although it is divided into the data storage unit 33, the storage device does not have to be actually divided.

In the behavior pattern analysis support system 1 according to the present embodiment, the field data generation unit 20 and the pattern analysis support unit 40 are configured by a computer (for example, a PC) that a CPU (Central Processing Unit) reads and executes program data. It is feasible, and the arithmetic processing device that realizes these can be regarded as the behavior pattern analysis support device according to the present embodiment. On the other hand, the data storage unit 30 may be realized by a server or the like that is communicably connected to the behavior pattern analysis support device, or may be realized by a storage device installed inside the behavior pattern analysis support device. May be.

The pattern analysis support unit 40 has a function of receiving an input of the observation data 140 from the analyst/user 50 and determining the analysis policy of the field data 110 in consideration of the input observation data 140. Specifically, as shown in FIG. 1, the pattern analysis support unit 40 includes an observation data input unit 41, a data integration unit 42, a data analysis unit 43, a data presentation unit 44, an analysis guideline input unit 45, and a parameter change unit. 46 is provided.

The observation data input unit 41 has a function of inputting, as the observation data 140, observation information obtained by visual observation of the site by an analyst.

The data integration unit 42 integrates the observation data 140 input to the observation data input unit 41 with respect to the field data 110 transmitted from the data aggregation unit 22, and the integrated data storage unit of the data storage unit 30 as integrated data 120. 32 has the function of saving.

The data analysis unit 43 performs detailed analysis on the integrated data 120 generated by the data integration unit 42 by utilizing the input observation data 140, and outputs the analysis result as analysis data 130. Have the function to The analysis data 130 is stored in the analysis data storage unit 33 of the data storage unit 30.

The detailed analysis (second analysis) by the data analysis unit 43 is assumed to be an analysis that can be performed after the quantitative data called the field data 110 is aggregated. Therefore, the first analysis by the data aggregation unit 22 is performed. Unlike, it is not suitable for real-time processing. Further, since the integrated data 120 is generated by reflecting the observation data 140 on the field data 110, the content may change from time to time, and the second analysis performed on the integrated data 120 as a target. However, the analysis content may change. Therefore, in other words, the second analysis is a mutable analysis.

The data presentation unit 44 has a function of presenting the integrated data 120 generated by the data integration unit 42 or the analysis data 130 generated by the data analysis unit 43 to the analyst/user 50. By observing these presentation data, the user can gain new awareness of the statistical and quantitative data by the field data 110 based on the observed information (observation data 140). ..

In the present embodiment, for the analyst/user 50, the observer who visually observes the field or the like and inputs the observation information (observation data 140) and the presentation data (integrated data 120, analysis data 130) are observed. The user who does this is not limited to the same person. Then, when another person makes each observation, insights that the related party (analyzer who observed the field) did not notice can be obtained by the related person (the user who observed the presented data). It can be expected to gain "new awareness" from the observed information.

The analysis guideline input unit 45 is used when the analyst/user 50 requests further analysis guideline (change or addition of analysis guideline) for the analysis guideline used when the field data generation unit 20 generates the field data 110. , Has a function of inputting the change request. The analyst/user 50 who formulates the analysis guideline may be the same as the person observing the field or the person observing the presentation data, or may be a different person.

The parameter changing unit 46 has a function of appropriately changing the parameter of the analysis guideline by the field data acquisition unit 21 based on the analysis guideline change request input to the analysis guideline input unit 45, and generating the correction parameter 150. Then, the parameter changing unit 46 transmits the generated correction parameter 150 to the field data generating unit 20.

Then, the field data generation unit 20 changes the parameters of its own analysis guideline in accordance with the modified parameter 150 transmitted from the parameter change unit 46, and reacquires the field data 110 using the analysis guideline after the parameter change. .. As a result, the field data generation unit 20 can acquire effective data by the action pattern analysis.

In the behavior pattern analysis support system 1 according to the present embodiment, the above-mentioned correction parameter 150 may be used for the purpose of improving the data acquisition after the above-mentioned parameter change (offline data analysis), or In the detailed analysis of the data analysis unit 43 for the already acquired field data 110, it may be utilized for the purpose of adjusting the parameters for the analysis (online data analysis). Further, these off-line/on-line data analyzes may be appropriately switched and used.

Further, in the present embodiment, the input means for inputting information from the analyst/user 50 to the pattern analysis support unit 40, and the output or display of information by the pattern analysis support unit 40 to the analyst/user 50. As a means, it is assumed that a user interface (UI) including an application running on a PC or the like is provided. Specifically, the observation data input UI 410 is provided as an input unit to the observation data input unit 41, the data presentation UI 440 is provided as an output unit by the data presentation unit 44, and the analysis guideline is provided as an input unit to the analysis guideline input unit 45. An input UI 450 is provided. Then, in the following description, the input/output between the pattern analysis support unit 40 and the analyst/user 50 will be described by using each of the above UIs.

(2) Analysis Process Next, the operation of the action pattern analysis support by the action pattern analysis support system 1 according to the present embodiment will be described together with the movement of the entire analysis process with reference to FIG.

FIG. 2 is a diagram for explaining the entire process of supporting action pattern analysis. The analysis process by the action pattern analysis support system 1 (steps S11 to S15) is shown in the upper part of FIG. 2, and the analysis process by the analyst/user 50 (steps S21 to S24) is shown in the lower part of FIG. ing. In the middle part of FIG. 2, the flow of processing by each component of the behavior pattern analysis support system 1 is shown centering on data input/output and movement, and these upper, middle and lower parts are arranged in chronological order. It is written.

In the analysis process by the action pattern analysis support system 1, first, preparation for data acquisition by the sensor 10 is performed (step S11). This data acquisition is acquisition of quantitative data, and for example, the field data generation unit 20 of the behavior pattern analysis support system 1 prepares a data acquisition plan 200. Specifically, the data acquisition plan 200 defines basic operations by the field data generation unit 20, such as how the sensors 10 are arranged and what items are acquired by imaging.

Next, the field data generation unit 20 acquires and aggregates the field data 110 from the imaged data of the sensor 10 obtained based on the data acquisition plan 200 (step S12).

Next, the data integration unit 42 integrates the field data 110 acquired/collected in step S12 and the observation data 140 input to the observation data input unit 41 to generate integrated data 120 (step S13).

Next, the data analysis unit 43 analyzes the integrated data 120 again to generate the analysis data 130. Further, the data presentation unit 44 presents the integrated data 120 and the analysis data 130 to the analyst/user 50 (step S14).

When the analysis guideline input unit 45 receives a request for changing the analysis guideline by the analyst/user 50, the parameter changing unit 46 generates the correction parameter 150 according to the intention of the analyst/user 50, The field data generation unit 20 changes the parameter for acquiring the field data 110 based on the modified parameter 150, and tries to acquire more data with the changed analysis guideline (step S15).

On the other hand, in the analysis process performed by the analyst/user 50, first, field observation information is acquired in the target area where the sensor 10 is arranged (step S21). The process of step S21 is preferably performed in the same time zone as the time (step S12) for obtaining the field data 110 in step S12, but the observation data is obtained in another time zone or viewpoint after the field data is obtained. It may be one.

Next, the observation information (observation data) acquired in step S21 is input to the observation data input unit 41 by the observation data input UI 410. The observation data 140 input here is used for integration with the field data 110 in step S13, so that the field data 110 is expanded (step S22).

Then, the data presentation unit 44 confirms the presentation data (integrated data 120, analysis data 130) output to the data presentation UI 440, and formulates the analysis guideline for the next cycle. Then, according to the established analysis guideline, the parameter to be changed is input to the analysis guideline input unit 45 by the analysis guideline input UI 450 (step S23).

Furthermore, by confirming the analysis data 130, the observation point of the analyst is changed, or the observation policy of the analyst is changed (step S24). Then, the process proceeds to step S21 of the next cycle, and the observation data of the field is acquired according to the changed observation policy.

As described above, in the present embodiment, while interactively repeating the analysis process by the action pattern analysis support system 1 and the analysis process by the analyst/user 50, the parameters of the analysis guideline of the field data 110 and the field observation policy. By modifying the, it is possible to make the data analysis of the behavior pattern analysis skillful.

(3) Specific Example of Data Input/Output As described with reference to FIGS. 1 and 2, in the behavior pattern analysis support system 1 according to the present embodiment, data input/output for the analyst/user 50 is performed. An observation data input UI 410, a data presentation UI 440, and an analysis guideline input UI 450 are provided as user interfaces for this. Hereinafter, regarding such data input/output, a case where a common user interface (common UI 400) is used in the entire behavior pattern analysis support system 1 will be described as an example while exemplifying a specific display screen.

FIG. 3 is a diagram showing an example of a user interface for data input/output. The data display screen 500 shown in FIG. 3 is an example of an upper layer display screen displayed by the common UI 400 for data input/output when the behavior pattern analysis is performed using the behavior pattern analysis support system 1.

Although a specific display mode of each will be described later, the common UI 400 enables data display from various viewpoints required by the analyst/user 50, and displays, for example, an action pattern on the map of the target area at a specified time. It is possible to display the amount of traffic on the floor within a predetermined time range, and to display the ratio to a specific attribute in a pie chart. Hereinafter, an operation example of the common UI 400 will be described using the data display screen 500.

First, when the user selects a date and time in the date and time column 501 and a wide area (building in this example) in the area column 502, a plurality of floors of the building are displayed in the display area 503. Further, when the user selects a specific area from the plurality of floors displayed in the display area 503, the information of the specific area is displayed and output in the display area 504. In the case of FIG. 3, the display area 504 displays the information of the floor with “4F” as the specific area in the wide area “ABC Shopping Center” at date and time “November 11, 2018, 15:00”.

Then, in the display area 504, the user can select the field data to be confirmed. Specifically, for example, the field data can be selected based on a category such as a male flow (for example, data A) and a retained person (for example, data C). The selectable field data is limited to that acquired as the field data 110, but the user can flexibly select the display target within the range.

Further, in the data display screen 500 of FIG. 3, the display areas 505 and 506 in the lower part can be displayed for analysis related to the field data, so that the data acquired as the field data 110 can be displayed from various viewpoints. It is possible to try the analysis while comparing. Specifically, in the case of FIG. 3, in the display area 505, the traffic volume of the area specified in the display area 503 (4F floor in this example) is graphically displayed by time zone. In the display area 506, the ratio of the field data selected in the display area 504 to the attributes A and B is displayed in a pie chart. Note that the data types for analysis that can be displayed in the display areas 505 and 506 are not limited to the example shown in the figure, and are obtained as field data 110 using a drop-down menu or the like. The data type to be displayed can be changed.

Further, buttons 507 and 508 are provided in the lower right portion of the data display screen 500 of FIG. When the button 507 is clicked, an input screen for inputting observation information (observation data) by the analyst/user 50 is displayed in a pop-up display. Further, when the button 508 is clicked, the current display content (may be the selection content) on the data display screen 500 can be saved in a file or the like.

The input of the observation information (observation data) will be described below with a specific example.

FIG. 4 is a diagram showing an example of a screen for inputting observation information. The observation data input screen 510 shown in FIG. 4 is used when the observation information (observation data) obtained by visual observation of the site by the analyst is input to the pattern analysis support unit 40 (observation data input unit 41). 6 is an example of a display screen displayed by a data input UI 410. The observation data input screen 510 of FIG. 4 is displayed in a pop-up, for example, when the button 507 is clicked on the data display screen 500 shown in FIG.

Here, the observation information that can be input in the present embodiment is not limited to specific information. For example, in the case of FIG. 4, when “Flow mode” is selected in the mode field 511 of the drop-down menu, the observation data input screen 510 becomes a screen for inputting observation information focusing on “people flow”. There is.

In the case of the observation data input screen 510, information of the area selected on the data display screen 500 is displayed and output in the display area 512 as in the display area 504 of FIG. In the case of 4, the sensor field of view 513 of the camera is also displayed. It is assumed that an analyst can write environmental information in the display area 512 or the like by using a dedicated input device.

At this time, as input of specific observation information to the observation data input screen 510, for example, by describing a flow line while tracking and observing a person moving on the floor (trajectory 514), the visual field of the camera (sensor visual field 513). ) You can also enter the behavior of a specific person who is out of. Furthermore, regarding the attribute of the person, information that is difficult to obtain with a camera or the like can be input as observation information (basic information column 516). Further, it is assumed that the information indicating the staying time in the predetermined area, the information indicating the content of conversation, and the like can be input in the trajectory 514 of the person (detailed information 515).

In addition to this, the remarks column 517 can additionally describe observation information that cannot be entered in the display area 512, the basic information column 516, or the like. Then, when the button 518 is clicked, the observation information input to the observation data input screen 510 is saved, and the observation data input unit 41 outputs the input observation information to the data integration unit 42 as the observation data 140. To do. Then, the data integration unit 42 creates and outputs the integrated data 120 by integrating the observation data 140 with the field data 110.

FIG. 5 is a diagram showing another example of a screen for inputting observation information. The observation data input screen 520 shown in FIG. 5 is an example of a display screen by the observation data input UI 410 when inputting the observation information (observation data) obtained by the analyst, like the observation data input screen 510 of FIG. However, it is possible to input observation information from a different viewpoint from the observation data input screen 510 of FIG. Specifically, in the case of FIG. 5, when the “Area mode” is selected in the mode field 521 of the drop-down menu, the observation data input screen 520 is a screen for inputting the observation information focusing on the “area”. Has become.

In the case of the observation data input screen 520, the information of the area selected on the data display screen 500 is displayed and output in the display area 522 similarly to the display area 512 of FIG. 4, and the sensor 10 (see FIG. In the case of 5, the sensor field of view 523 of the camera) is displayed. Then, it is assumed that the analyst can write environmental information in the display area 522 and the like by using a dedicated input device.

At this time, as an input of specific observation information to the observation data input screen 520, for example, an analyst displays information summarizing a specific area (target area 524) in a specific time zone, condition, etc. (basic information 526). It is possible to input with a memo in text format (detailed information 525). As shown in FIG. 5, the detailed information 525 can include information that is difficult to obtain by the sensor 10 such as a camera, such as the status of congestion and the usage status of elevators. In addition to this, in the remarks column 527, observation information that does not have a place to be entered in the display area 522, the basic information 526, or the like can be additionally described.

Further, in the observation data input screen 520 of FIG. 4, in the additional input screen 529 displayed in a pop-up, it is possible to further input the observation information focusing on the area. In the case of FIG. 4, the additional input screen 529 summarizes the area coordinates, pedestrian flow, average residence time, and density for each area (Area1 to Area3), and describes quantitative information of the fields acquired by the sensor 10. ing. Then, in this additional input screen 529, in addition to the above-mentioned quantitative information, observation information can be additionally input in a blank space (hatched portion) with a certain scale. Here, the certain scale can express a predetermined degree, and specifically means, for example, to describe “crowd: large” in a blank column of the flow of people. In the present embodiment, the input of the observation information based on the above-described fixed scale may be directly input by the analyst, or the conversion processing based on the input content such as the detailed information 525 is performed. You may make it input automatically. The pop-up display of the additional input screen 529 is performed by a predetermined operation as a trigger, and is displayed, for example, when the button 528 is clicked.

As described above with reference to FIG. 4 and FIG. 5, in the present embodiment, the situation of the area other than the acquisition range of the field data 110 by the sensor 10 (that is, outside the sensor target) is input by the observation information. Can be recorded in detail.

Next, the output of the presentation data (integrated data 120, analysis data 130) will be described with a specific example.

As described above, the pattern analysis support unit 40 inputs the observation information focusing on the person behavior (for example, the flow of people) or the area on the observation data input UI 410, and confirms it by the analyst (observer)'s consideration or visual observation. The characteristic information (observation data 140) to be obtained can be acquired. Then, the data analysis unit 43 performs a predetermined process on the integrated data 120 obtained by integrating the observation data 140 into the field data 110 and aggregates the integrated data 120, so that a more detailed second analysis can be performed. The unit 44 can output the analysis information as presentation data.

FIG. 6 is a diagram illustrating an example of a display screen of presentation data. The presentation data display screen 530 shown in FIG. 6 is a display screen displayed by the data presentation UI 440 when the data presentation unit 44 presents the analysis information (analysis data 130) of the integrated data 120 to the analyst/user 50. This is an example. The presentation data display screen 530 of FIG. 6 has the same screen configuration as the data display screen 500 illustrated in FIG. 3, and the description of the common screen configuration will be omitted. It may be considered that the presentation data display screen 530 is displayed by appropriately transitioning from the data display screen 500 according to the progress of processing by the pattern analysis support unit 40.

On the presentation data display screen 530, icons 532 and 534 are displayed in the attention area where the observation information is input and the information has a certain level or more in the display area 531 in which the information of the area selected by the user (or the analyst) is displayed. Has been done.

When the icon 532 is selected, notices estimated from the document (detailed information 525, remarks column 517, 527, etc.) described in the memo by the person who entered the observation information regarding the attention area is displayed, for example, in order of frequency. (Feature information 533).

Here, the above-mentioned "awareness" can be displayed in the second analysis by the data analysis unit 43 by decomposing the text into word levels and aggregating it as statistical data for a predetermined period or condition. As another example, when general (all time period) data is aggregated as statistical data, the notice common to all time zones can be displayed, and when only the holiday data is separately aggregated, Can also collect the notices that are unique to holidays.

It is also possible to aggregate traffic volume, attribute information, etc. in a rather wide area as statistical data. As a specific example of the case where such aggregation is performed, for example, when the icon 534 is selected, the traffic volume (attribute A) and other attribute information (attribute B) in the extended area 535 are displayed in the attention area information 536 as a graph. Can be displayed. As another method, by selecting a desired area such as the extended area 535, data of only the area of interest can be displayed in the area of interest information 536.

Next, another example of data presentation will be described. FIG. 7 is a diagram showing another example of the display screen of the presentation data. On the presentation data display screen 540 shown in FIG. 7, movement information when attention is paid to an attribute (personal attribute information) of a person or the like is displayed.

Here, "personal attribute information" is attribute information of a person that can be understood only by observation information, and specifically, for example, nationality that cannot be visually recognized, or a unique persona defined for each user. Examples include classification according to categories. In addition to this, a specific category such as a person who used a certain tenant if the target field is a store may be used. Then, as a result of analyzing the movement information such as the movement line of the person having the above-mentioned predetermined attribute, for example, the frequency of movement, the detailed integrated data 120 including the observation data 140 is aggregated and analyzed. Is displayed as presentation data.

Specifically, in the case of the presentation data display screen 540 shown in FIG. 7, the movement information of the person having the selected attribute (attribute A) is indicated by arrows 542 to 544 in the display area 541. The thickness of the arrows 542 to 544 represents the frequency. Note that the expression of the frequency is not limited to the thickness of the arrow, and the expression method can be determined by the user setting, such as the expression with a color such as a heat map or the intensity of the brightness.

As described above with reference to FIGS. 6 and 7, in the present embodiment, various settings are possible for the period and conditions for collecting statistical data in the second analysis, and the setting options are set on the system. By setting in advance, it is possible to design according to the problem for each observation target. Then, the result of the second analysis performed according to these designs is displayed on the presentation data display screen 530, so that the analyst/user 50 can display the information (“notice” or “ It is possible to recognize the characteristics of a person's action pattern and to examine whether or not the analysis pattern of the action pattern is appropriate based on "movement information" and the like).

Next, the correction input of the analysis guide (change of parameters of the analysis guide) will be described in detail.

FIG. 8 is a diagram showing an example of an input screen for changing the parameters of the analysis guideline. The parameter change screen 550 shown in FIG. 8 uses the analysis data 130 and the integrated data 120 displayed as the presentation data to correct the parameters of the analysis guide regarding the acquisition of the field data 110 by the sensor 10. 6 is an example of a display screen displayed by the input UI 450. The parameter change screen 550 of FIG. 8 is displayed, for example, when a predetermined setting button is clicked on the data display screen 500 shown in FIG.

Specifically, in the parameter change screen 550, in the display area 551, the information of the presentation data is displayed similarly to the display area 531 of FIG. 6 and the display area 541 of FIG. 7, and while viewing these presentation data, The analyst/user 50 writes the corrected value of the desired parameter in the parameter change field 552.

Then, when the button 553 is clicked, the analysis guideline input unit 45 sends the correction value (analysis guideline change request) of the parameter entered in the parameter change column 552 to the parameter change unit 46, and the parameter change unit 46 The correction parameter 150 is generated based on the correction value. Based on this correction parameter 150, the field data generation unit 20 can correct the analysis guideline used when generating the field data 110, and this is reflected from the subsequent cycles.

(4) Summary As described above, according to the behavior pattern analysis support system 1 according to the present embodiment, the observation data 140 can be input to the field data 110 acquired from the sensor 10 and aggregated. By performing detailed analysis on the integrated data 120 obtained by integrating the input observation data 140 into the field data 110, and changing (correcting) the viewpoint of field data acquisition based on the analysis result. The analysis guideline for the space that is the target of the behavior pattern analysis can be developed to be more suitable.

Thus, according to the behavior pattern analysis support system 1 according to the present embodiment, the measured data (field data 110) is utilized for further understanding of the situation of a person or the like who uses the field, and further observes in the field. By integrating the data (observation data 140) acquired by the above with the above-mentioned data, it is possible to gain deep insight into the data analysis. Further, by interactively inputting a guideline for data analysis by a person, the data measurement by the existing sensor 10 is advanced, and it can be expected to further improve the accuracy of data analysis and the comprehensiveness of analysis items.

That is, according to the behavior pattern analysis support system 1 according to the present embodiment, it is possible to interactively input a guideline for data analysis (behavior pattern analysis) by the user, and to the field data 110 acquired by the existing sensor 10. , It can support to be able to perform more precise behavioral pattern analysis.

In addition, in the present embodiment, the input by the analyst/user 50 such as the input of the observation information and the input of the changed value of the parameter has been described by using the process performed by the screen input on the PC. The input process by 50 is not limited to this, and may be another process. Specifically, for example, it may be possible to perform real-time local input with a tablet terminal, or a process of converting the page into data by scanning based on the entry on the page.

As can be seen from the embodiment of the present invention described in detail above, the present invention is an ethnographic survey in analysis of field data that is quantitative information such as trajectory data and statistical data of a person acquired by a sensor. By making effective use of information from such field observations and human behavior observations, it is possible to gain more precise and deep insights into field data. Further, according to the present invention, the measuring means by the sensor can be changed according to the situation, and further, the method of site observation can be improved based on the field data acquired after the change. As described above, the present invention interactively conducts the data analysis by the action pattern analysis support device (or the action pattern analysis support system) and the insight by the observer (analyzer/user) regarding the guidelines and methods of the field data analysis. By repeating this process, it is possible to enrich the understanding of the field and obtain more effective information for behavior pattern analysis.

It should be noted that the present invention is not limited to the above-described embodiment, and various modifications are included. For example, the above-described embodiments have been described in detail in order to explain the present invention in an easy-to-understand manner, and are not necessarily limited to those having all the configurations described. Further, it is possible to add/delete/replace other configurations with respect to a part of the configurations of the embodiments.

Further, each of the above-mentioned configurations, functions, processing units, processing means, etc. may be realized in hardware by designing a part or all of them with, for example, an integrated circuit. Further, each of the above-described configurations, functions, and the like may be realized by software by a processor interpreting and executing a program that realizes each function. Information such as a program, a table, and a file that realizes each function can be stored in a memory, a recording device such as a hard disk and an SSD (Solid State Drive), or a recording medium such as an IC card, an SD card, and a DVD.

Further, in each drawing, control lines and information lines which are considered necessary for explanation are shown, and not all control lines and information lines are shown on the product. In practice, it may be considered that almost all configurations are connected to each other.

1 Behavior pattern analysis support system 10 Sensor 20 Field data generation unit 21 Field data acquisition unit 22 Data aggregation unit 30 Data storage unit 31 Field data storage unit 32 Integrated data storage unit 33 Analysis data storage unit 40 Pattern analysis support unit 41 Observation data input Part 42 data integration part 43 data analysis part 44 data presentation part 45 analysis guideline input part 46 parameter change part 50 analyst/user 110 field data 120 integrated data 130 analysis data 140 observation data 150 modified parameters 200 data acquisition plan 400 common UI
410 Observation data input UI
440 Data presentation UI
450 Analysis guideline input UI

Claims (13)

A behavior pattern analysis support device for supporting analysis of a motion pattern of a moving object,
A field data generation unit that acquires and collects measurement data from the sensor to generate field data,
A pattern analysis support unit for expanding and analyzing the field data,
Equipped with
The pattern analysis support unit,
An observation data input section for inputting observation data based on observation of a field by a user,
A data integration unit that generates integrated data by integrating the observation data with the field data,
A data presentation unit that presents the integrated data to the user side,
An action pattern analysis support device characterized by having: The field data generation unit can change the analysis guideline used when generating the field data, and when the change is made, generates the next field data using the changed analysis guideline. The action pattern analysis support device according to claim 1, wherein: The pattern analysis support unit,
An analysis guideline input unit for inputting a change request of the analysis guideline by a user,
In response to the change request input to the analysis guideline input unit, a parameter changing unit that generates a correction parameter of the analysis guideline,
Further has
The action pattern analysis support device according to claim 2, wherein the field data generation unit changes the analysis guideline based on the correction parameter generated by the parameter change unit. The action pattern analysis support device according to claim 3, wherein the processing by the field data generation unit and the processing by the pattern analysis support unit are alternately repeated. The observation data input to the observation data input unit is the first observation data obtained based on an observation focusing on a person's behavior indicated by the field data, or identification within the target area from which the field data is acquired. The behavior pattern analysis support device according to claim 1, wherein at least one of the second observation data obtained based on the observation focused on the area is selectable. The pattern analysis support unit further includes a data analysis unit that analyzes the integrated data and generates analysis data based on an analysis result,
The field data generation unit executes a predetermined first analysis process on the aggregated measurement data when generating the field data,
The data analysis unit performs a second analysis process on the integrated data, the analysis content being changeable based on the observation data,
The action pattern analysis support device according to claim 1, wherein the data presentation unit presents the integrated data or the analysis data to a user side. A behavior pattern analysis support system for supporting analysis of behavior patterns of a moving object,
A field data generation unit that acquires and collects measurement data from the sensor to generate field data,
A pattern analysis support unit for expanding and analyzing the field data,
A data storage unit that stores the data generated by the field data generation unit or the pattern analysis support unit;
Equipped with
The pattern analysis support unit,
An observation data input section for inputting observation data based on observation of a field by a user,
A data integration unit that generates integrated data by integrating the observation data with the field data,
A data presentation unit that presents the integrated data to the user side,
A behavior pattern analysis support system characterized by having. A behavior pattern analysis support method for supporting analysis of a motion pattern of a moving object, comprising:
A field data generating step of generating and collecting the measurement data by the sensor and collecting the field data,
An observation data input step capable of inputting observation data based on observation of a field by a user,
A data integration step of integrating the field data generated in the field data generation step and the observation data input in the observation data input step to generate integrated data;
A data presentation step of presenting the integrated data generated in the data integration step to the user side;
A method for supporting behavior pattern analysis, comprising: It is possible to change the analysis guideline used when generating the field data in the field data generating step,
The action pattern analysis support method according to claim 8, wherein when the analysis guideline is changed, the field data is generated using the changed analysis guideline in the field data generating step. After the data presentation step,
An analysis guideline input step capable of inputting a change request of the analysis guideline by a user,
In response to the change request input in the analysis guideline input step, a parameter changing step of generating a correction parameter of the analysis guideline and changing the analysis guideline based on the correction parameter,
The action pattern analysis support method according to claim 9, further comprising: The processing in the field data generation step and the processing in the observation data input step, the data integration step, the data presentation step, the analysis guideline input step, and the parameter changing step are alternately repeated. The behavior pattern analysis support method according to claim 10. The observation data that can be input in the observation data input step is the first observation data obtained based on an observation focusing on a person's behavior indicated by the field data, or identification within the target area from which the field data is acquired. 9. The behavior pattern analysis support method according to claim 8, wherein at least one of the second observation data obtained based on the observation focused on the area is selectable. After the data integration step,
Further comprising a data analysis step of analyzing the integrated data and generating analysis data based on the analysis result,
In the field data generating step, when the field data is generated, a predetermined first analysis process is executed on the aggregated measurement data,
In the data analysis step, a second analysis process capable of changing the analysis content based on the observation data is executed on the integrated data,
The action pattern analysis support method according to claim 8, wherein in the data presentation step, the integrated data or the analysis data is presented to a user side.

Images