Score calculating device and score calculating method

Abstract

To provide a score calculating device and a score calculating method capable of preventing a trouble between a plurality of subjects belonging to a group.SOLUTION: A score calculating device includes: an isolated action score calculating unit configured to calculate, based on action information of subjects, an isolated action score indicating a degree of proactive isolated action in a predetermined group; a tendency score calculation unit configured to calculate a tendency score indicating a tendency of each subject regarding interaction with others; and a communication score calculating unit configured to calculate a communication score indicating a degree of communication with others in the group for each of the subjects during a target period based on the isolated action score and the tendency score.SELECTED DRAWING: Figure 1

Description

[Technical field]

[0001] The present invention relates to a score calculation device and a score calculation method. [Background technology]

[0002] There are concerns about bullying and other troubles in facilities such as schools, kindergartens, nurseries, etc. Patent documents 1 to 3 describe detection systems that detect bullying and other troubles using video information, audio information, emotional information, etc. of a target person. [Prior art documents] [Patent documents]

[0003] [Patent Document 1] Patent Publication No. 2021-096836 [Patent Document 2] JP 2020-123204 A [Patent Document 3] JP 2018-112831 A Summary of the Invention [Problem to be solved by the invention]

[0004] The systems described in Patent Documents 1 to 3 distinguish between children who are the victims of bullying and other troubles and children who are contributing to the troubles, and then detect subsequent events such as injuries, and direct events such as inappropriate words, etc. However, it is preferable to prevent bullying and other troubles in advance by taking measures such as reaching out to the target child and surrounding children before the troubles become apparent.

[0005] The present invention has been made in consideration of the above, and aims to provide a score calculation device and a score calculation method that can prevent trouble between multiple subjects belonging to a group in advance. [Means for solving the problem]

[0006] The score calculation device according to the present invention includes an individual behavior score calculation unit that calculates an individual behavior score indicating the degree to which a subject actively engages in individual behavior toward a specified group based on behavioral information of the subject, a propensity score calculation unit that calculates a propensity score that indicates a tendency for each of the subjects regarding interactions with others, and a communication score calculation unit that calculates a communication score that indicates the degree of communication with others in the group for each of the subjects during a target period based on the individual behavior score and the propensity score.

[0007] The score calculation method of the present invention includes calculating an individual behavior score indicating the degree to which a subject actively engages in individual behavior toward a specified group based on behavioral information of the subject, calculating a propensity score indicating each subject's tendency regarding interactions with others, and calculating a communication score indicating the degree of communication with others in the group for each subject based on the individual behavior score and the propensity score. Effect of the Invention

[0008] According to the present invention, it is possible to prevent troubles between multiple subjects belonging to a group in advance. [Brief description of the drawings]

[0009] [Figure 1] FIG. 1 is a functional block diagram showing an example of a score calculation device according to the present embodiment. [Diagram 2] FIG. 2 is a diagram showing an example of a detection result of position information of a subject. [Diagram 3] FIG. 3 is a diagram showing an example of a detection result of position information of a target person and a target object. [Figure 4] FIG. 4 is a flowchart showing the flow of operations of the score calculation device according to this embodiment. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0010] Hereinafter, embodiments of the score calculation device and score calculation method according to the present invention will be described with reference to the drawings. Note that the present invention is not limited to these embodiments. In addition, the components in the following embodiments include those that are replaceable and easy for a person skilled in the art, or those that are substantially the same.

[0011] Fig. 1 is a functional block diagram showing an example of a score calculation device 100 according to this embodiment. The score calculation device 100 shown in Fig. 1 calculates a communication score indicating the degree of communication with others in a predetermined group for each of multiple subjects belonging to the group. In this embodiment, the predetermined group may be, for example, a school, a kindergarten, a nursery school, etc. The multiple subjects may be children belonging to these schools, kindergartens, nursery schools, etc.

[0012] The score calculation device 100 includes a processing device such as a CPU (Central Processing Unit) and a storage device such as a RAM (Random Access Memory) or a ROM (Read Only Memory). As shown in FIG. 1, the score calculation device 100 includes an isolation score calculation unit 10, an individual behavior score calculation unit 20, a tendency score calculation unit 30, a communication score calculation unit 40, and a storage unit 50.

[0013] The isolation score calculation unit 10 calculates an isolation score based on the history of position information of multiple subjects belonging to a predetermined group. The isolation score is a score indicating the degree of isolation from the group for each subject. The isolation score calculation unit 10 can calculate the isolation score based on, for example, the average value of the distance between the subject and other subjects during a target period, and the number of events indicating the number of times the subject comes into close proximity with other subjects for a first time or more during the target period.

[0014] The isolation score calculation unit 10 detects position information for each subject as coordinates. The position information is calculated based on, for example, image data captured by the image capture device 60. The image data captured by the image capture device 60 can be stored in the storage unit 50. The image capture device 60 can be placed in a location where the subject is active, such as a school classroom. In this case, the isolation score calculation unit 10 can calculate the position information as coordinates in a coordinate plane based on, for example, a predetermined position in the classroom.

[0015] The isolation score calculation unit 10 distinguishes and extracts individual subjects based on the photographed data, for example, by image processing. The isolation score calculation unit 10 calculates position information for each extracted subject for each frame.

[0016] The isolation score calculation unit 10 calculates, for each subject, an average value of the distance between the subject and other subjects during a predetermined period based on the calculated position information. The predetermined period can be set, for example, as a break time, the time from arriving at school to going home from school, etc. The isolation score calculation unit 10 can calculate the distance between the subject and other subjects for each frame, for example, and calculate the average value based on the number of frames.

[0017] The isolation score calculation unit 10 calculates the average value (A), for example, as shown in the following formula: Note that the following formula is for the case where attention is focused on one subject. A=Σ0~M-2Σ0~N-1(Distancem,n) / (N·(M-1)) Here, Distance indicates the distance between the subject and other subjects in one frame. M indicates the total number of subjects, including the subject and other subjects, in the frame. m is a number indicating the other subjects. The first other subject is indicated by m=0. The last other subject is indicated by m=M-2. n indicates the frame number. The first frame is indicated by n=0. Also, the last frame is indicated by n=N-1.

[0018] The isolation score calculation unit 10 detects, for each target, whether or not the target and another target have come into close proximity with each other based on the calculated position information. When the target and another target have come into close proximity with each other, it can be estimated that an event for communication such as conversation has occurred between the target and the other target. Therefore, when the isolation score calculation unit 10 detects, for each target, that the target and the other target have come into close proximity with each other, it determines that an event has occurred between the target and the other target, and counts the number of events as one.

[0019] The isolation score calculation unit 10 can divide the distance into ranges such as personal distance (0.45 m or more and less than 1.2 m), close distance (0 m or more and less than 0.45 m), etc. based on the concept of personal space, and set a different time for each range as the first time, which is the time for detecting proximity. For example, the first time corresponding to the personal distance can be set to 15 seconds, and the first time corresponding to the close distance can be set to 5 seconds. In this case, the isolation score calculation unit 10 detects proximity and determines that an event has occurred when the subject and another subject are at personal distance for 15 seconds or more during a predetermined period, and when the subject and another subject are at close distance for 5 seconds or more, and counts the number of events as one.

[0020] The isolation score calculation unit 10 calculates the isolation score (S1) based on the calculated average value (A) and the number of events (B) as shown in the following formula. The isolation score calculation unit 10 stores the calculated isolation score S1 in the storage unit 50. S1 = α A + (1-α) (Max-B) Here, α is a number indicating the weight and can be set to 0≦α≦1. Also, Max indicates the maximum value of the event count B. The value of the event count B will not increase any more once it reaches Max.

[0021] The higher the isolation score S1, the higher the subject's degree of isolation, and the lower the score, the lower the subject's degree of isolation. The isolation score S1 increases as the average distance from other subjects increases and the number of events decreases. Conversely, the isolation score S1 decreases as the average distance from other subjects decreases and the number of events increases.

[0022] The individual behavior score calculation unit 20 calculates an individual behavior score based on the behavior information of the subject. The individual behavior score is a score indicating the degree to which the subject actively performs individual behavior in relation to a group. The behavior information of the subject on which the individual behavior score is calculated is information on the proximity situation of the subject to an object for performing the individual behavior, and when the individual behavior score calculation unit 20 detects that the subject has approached an object for performing the individual behavior during the target period, it calculates the individual behavior score based on the type of object for each detection. Examples of the above objects include toys, textbooks, notebooks, pencils, books, etc. When the subject is in proximity to such objects, it can be estimated that the subject is actively performing individual behavior using the object.

[0023] The individual behavior score calculation unit 20 can detect the type of object and the position information of the object by performing image processing or the like based on the photographing data of the photographing device 60. The individual behavior score calculation unit 20 calculates the distance between the subject and the object based on the position information of the subject stored in the storage unit 50 and the calculated position information of the object.

[0024] When the individual behavior score calculation unit 20 detects that the individual subject and the object have come close to each other during a predetermined period, the individual behavior score calculation unit 20 determines that the individual subject has performed an individual behavior. For example, when the individual subject and the object are in close proximity (0 m or more and less than 0.45 m) or the individual subject and the object are in contact with each other for a second time or more, the individual behavior score calculation unit 20 can detect that the individual subject and the object are in close proximity to each other.

[0025] When the individual behavior score calculation unit 20 detects that the subject has been in close proximity to an object for performing an individual behavior for a second time or more, the individual behavior score calculation unit 20 calculates an individual behavior score (S2) based on the type of the object for each detection as shown in the following formula. The individual behavior score calculation unit 20 stores the calculated individual behavior score S2 in the storage unit 50. S2 = Σ0~H-1(Kg,h) Here, K is a value indicating the weight for each detection, and g is a value indicating the type of object. For example, the values ​​corresponding to toy A are set in advance as g=1, K1, h=10, the values ​​corresponding to toy B as g=2, K2, h=8, the values ​​corresponding to textbooks as g=3, K3, h=5, etc., and the setting results can be stored as information such as a table in the storage unit 50. h is a value indicating the number of detections, and H is the total number of detections.

[0026] The higher the individual behavior score S2, the higher the degree to which the subject is actively performing individual behavior, and the lower the score, the lower the degree to which the subject is actively performing individual behavior. The individual behavior score S2 is 0 when proximity to an object is not detected. In addition, the individual behavior score S2 becomes higher when an object with a higher weighting value is used.

[0027] The propensity score calculation unit 30 calculates a propensity score. The propensity score is a score that indicates the tendency of each subject regarding interactions with others. The propensity score calculation unit 30 can calculate the propensity score based on the communication score history of the subject and correction information that is set in advance for the subject.

[0028] The propensity score calculation unit 30 calculates the average value of the communication scores calculated for the subject in a predetermined period in the past (for example, the past year, etc.) as the typical value. In this case, the propensity score calculation unit 30 can use the communication scores stored in the storage unit 50. The correction information can be, for example, a value evaluated by the subject's teacher, guardian, etc. on the degree to which the subject tends to perform individual behavior. The correction information can be set, for example, by a questionnaire or the like. If the subject is a new kindergartener, for example, and the past communication scores for the subject are not stored in the storage unit 50, the average value of the communication scores of general children of the same age can be used as the typical value.

[0029] The propensity score calculation unit 30 calculates the propensity score (S3) based on the typical value (C) and the correction information (D) as shown in the following formula. The propensity score calculation unit 30 stores the calculated propensity score S3 in the storage unit 50. S3=C+D The communication score calculation unit 40 calculates a communication score based on the isolation score, the individual behavior score, and the tendency score. The communication score is a score indicating the degree of communication between each subject and others in a group.

[0030] The communication score calculation unit 40 calculates the communication score (S4) as shown in the following formula: The communication score calculation unit 40 stores the calculated communication score S4 in the storage unit 50. S4 = β / S1 + (1-β) · S2 · S3 Here, β is a number indicating a weight and can be set as 0≦β≦1.

[0031] The communication score becomes higher as the isolation score is lower and the personal behavior score and tendency score are higher.

[0032] The storage unit 50 stores various information. The storage unit 50 has storage such as a hard disk drive or a solid state drive. An external storage medium such as a removable disk may be used as the storage unit 50. The storage unit 50 stores data and programs for executing processes in the isolation score calculation unit 10, the individual behavior score calculation unit 20, the tendency score calculation unit 30, and the communication score calculation unit 40.

[0033] The storage unit 50 stores the photographed data photographed by the photographing device 60. The storage unit 50 stores the calculation results of the isolation score calculation unit 10. Examples of the calculation results of the isolation score calculation unit 10 include position information for each subject, an isolation score, etc. The storage unit 50 stores the calculation results of the individual behavior score calculation unit 20. Examples of the calculation results of the individual behavior score calculation unit 20 include position information for each subject, type and position information for each object, an individual behavior score, etc. The storage unit 50 stores the calculation results of the propensity score calculation unit 30. Examples of the calculation results of the propensity score calculation unit 30 include a typical value for each subject, correction information, a propensity score, etc. The storage unit 50 stores the calculation results of the communication score calculation unit 40. Examples of the calculation results of the communication score calculation unit 40 include a communication score, etc.

[0034] Next, a score calculation method according to the present embodiment will be described. In the score calculation method according to the present embodiment, in a facility such as a school, kindergarten, or nursery school, an image capturing device 60 captures images of various locations such as classrooms, gymnasiums, and grounds, and the captured image data is provided to the score calculation device 100.

[0035] In the score calculation device 100, the isolation score calculation unit 10 individually extracts multiple subjects such as children based on the supplied shooting data, and detects the position information of each subject for each frame. FIG. 2 is a diagram showing an example of the detection result of the subject's position information. FIG. 2 shows a schematic example of the detection result for a predetermined period. In the example shown in FIG. 2, the position information of multiple subjects P present in a classroom is shown. The following description will be given by taking as an example a case where the state shown in FIG. 2 has continued for a first hour or more. In addition, when multiple subjects P are to be distinguished, they are represented as subjects P1 to P9.

[0036] The isolation score calculation unit 10 stores the position information of each detected subject in the storage unit 50. At this time, the isolation score calculation unit 10 can prevent the supplied photographic data from being stored as an image in the storage unit 50. This can prevent excessive recording of the subject's behavior, and can prevent subjects who are the perpetrators of trouble from being inadvertently highlighted.

[0037] The isolation score calculation unit 10 calculates an average value of the distance between the subject and other subjects based on the calculated position information. In addition, the isolation score calculation unit 10 detects, for each subject, whether the subject and other subjects have been in close proximity for a first time or more based on the calculated position information, and calculates the number of events based on the detection result.

[0038] In the example shown in FIG. 2, the isolation score calculation unit 10 calculates the average value of the distance between each of the subjects P1 to P9 and the other subjects P. The isolation score calculation unit 10 also detects that the subjects P1 and P2 are in close proximity to each other for a first time or more. In this case, the number of events of the subjects P1 and P2 is counted once. The isolation score calculation unit 10 also detects that the subjects P3, P4, P5, and P6 are in close proximity to each other for a first time or more. In this case, the number of events of the subjects P3, P4, P5, and P6 is counted once. The isolation score calculation unit 10 does not count the number of events of the subjects P7 to P9 because the subjects P7 to P9 are not in close proximity to the other subjects P. The isolation score calculation unit 10 calculates the isolation score based on the calculated average value and the number of events. In the example shown in FIG. 2, the isolation scores of the subjects P7 to P9 are higher than the isolation scores of the subjects P1 to P6.

[0039] The individual behavior score calculation unit 20 extracts the objects individually based on the supplied photographic data, and detects the position information of the objects. FIG. 3 is a diagram showing an example of the detection result of the position information of the objects and the objects. FIG. 3 shows a schematic example of the detection result in a predetermined period. In the example shown in FIG. 3, the position information of a plurality of objects P and the position information of an object Q existing in a classroom are shown. The following describes an example in which the state shown in FIG. 3 continues for a second time or more. In addition, when the plurality of objects P are to be distinguished, they are written as objects P1 to P9, and when the object Q is to be distinguished, they are written as objects Q1 and Q2.

[0040] The individual behavior score calculation unit 20 detects the object Q by image processing or the like. The individual behavior score calculation unit 20 detects the type and position information of the detected object Q. Based on the position information of the object Q and the position information of the subject P detected by the isolation score calculation unit 10, the individual behavior score calculation unit 20 detects, for each subject P, whether the subject P and the object Q have been in close proximity for a second time or more. In the example shown in FIG. 3, the individual behavior scores of subjects P7 and P8 are higher than the isolation scores of subjects P1 to P6 and P9.

[0041] In the example shown in Fig. 3, the individual behavior score calculation unit 20 does not detect subjects P1 to P6 and P9 because they are not in close proximity to the object. The individual behavior score calculation unit 20 detects that subject P7 has been in close proximity to object Q1 for a second time or more. The individual behavior score calculation unit 20 detects that subject P8 has been in close proximity to object Q2 for a second time or more. The individual behavior score calculation unit 20 calculates the individual behavior score based on the detection result.

[0042] The propensity score calculation unit 30 calculates the typical value and correction information for each subject, and calculates the propensity score based on the calculation results.

[0043] The communication score calculation unit 40 calculates a communication score based on the calculated isolation score, individual behavior score, and tendency score. The communication score becomes higher as the isolation score becomes lower and the individual behavior score and tendency score become higher.

[0044] For example, in the example shown in Figures 2 and 3, subject P9 has a high isolation score and a low individual behavior score, so it is estimated that the communication score will be low. On the other hand, subject P9 may have a high tendency to actively perform individual behavior even in the absence of an object, such as a high tendency to think alone. In such a case, the propensity score of subject P9 is calculated to be high. For this reason, subject P9 can be evaluated separately from subjects who are forced to passively perform individual behavior.

[0045] An administrator who manages multiple subjects can understand the degree of communication between each subject and others in the group based on the calculated communication score. For example, by taking measures when there is a change in the communication score, such as a significant drop in the communication score, it is possible to prevent bullying and other troubles before they occur.

[0046] Fig. 4 is a flowchart showing the flow of operations of the score calculation device 100 according to this embodiment. As shown in Fig. 4, in the score calculation device 100, the isolation score calculation unit 10 calculates an isolation score for each subject based on the history of position information of a plurality of subjects (step S10). In step S10, the isolation score calculation unit 10 calculates the isolation score based on the average value of the distance between the subject and other subjects during the target period and the number of events indicating the number of times the subject comes into close proximity with other subjects for a first time or more during the target period.

[0047] Next, the individual behavior score calculation unit 20 calculates an individual behavior score indicating the degree to which the individual actively performs individual behavior in relation to the group based on the behavior information of the individual (step S20). When the individual behavior score calculation unit 20 detects, as the behavior information, that the individual has come into close proximity to an object for performing an individual behavior for a second time or more during the target period, it calculates an individual behavior score based on the type of the object for each detection.

[0048] Next, the propensity score calculation unit 30 calculates a propensity score indicating a tendency of each subject regarding interactions with others (step S30). In step S30, the propensity score calculation unit 30 calculates the propensity score based on the communication score history of the subject and correction information previously set for the subject.

[0049] Next, the communication score calculation unit 40 calculates a communication score indicating the degree of communication with others in the group for each subject during the target period based on the isolation score, the individual behavior score, and the tendency score (step S40).

[0050] As described above, the score calculation device 100 of this embodiment includes an isolation score calculation unit 10 that calculates an isolation score indicating the degree of isolation from a group for each subject based on the history of location information of a plurality of subjects belonging to a specific group, an individual behavior score calculation unit 20 that calculates an individual behavior score indicating the degree to which the subject is actively engaging in individual behavior toward the group based on the subject's behavioral information, a propensity score calculation unit 30 that calculates a propensity score indicating the tendency of each subject regarding interactions with others, and a communication score calculation unit 40 that calculates a communication score indicating the degree of communication with others in the group for each subject during a target period based on the isolation score, the individual behavior score, and the propensity score.

[0051] In addition, the score calculation method according to this embodiment includes calculating an isolation score indicating the degree of isolation from a group for each subject based on the history of location information of multiple subjects belonging to a specified group, calculating an individual behavior score indicating the degree to which the subject is actively engaging in individual behavior toward the group based on the subject's behavioral information, calculating a propensity score indicating each subject's tendency regarding interactions with others, and calculating a communication score indicating the degree of communication with others in the group for each subject based on the isolation score, individual behavior score, and propensity score.

[0052] According to this configuration, a communication score can be calculated for each of a plurality of subjects belonging to a predetermined group. Therefore, by using the calculated communication score, measures can be taken before a subsequent event, a direct event, or the like is detected. This makes it possible to prevent troubles between a plurality of subjects belonging to a group in advance. In addition, by using the individual behavior score and the tendency score in calculating the communication score, the tendency of each subject, such as whether being alone is an active event or not, whether the subject likes being alone or not, is taken into consideration. Therefore, a communication score suited to the subject can be calculated.

[0053] In the score calculation device 100 according to this embodiment, the isolation score calculation unit 10 calculates the isolation score based on the average value of the distance between the subject and other subjects during the target period and the number of events indicating the number of times the subject comes into close proximity with other subjects during the target period. With this configuration, the isolation score of the subject can be calculated with high accuracy.

[0054] In the score calculation device 100 according to the present embodiment, when the personal behavior score calculation unit 20 detects, as behavior information, that the subject has come close to an object for performing a personal behavior during a target period, the personal behavior score calculation unit 20 calculates the personal behavior score based on the type of the object for each detection. With this configuration, the personal behavior score of the subject can be calculated with high accuracy.

[0055] In the score calculation device 100 according to this embodiment, the propensity score calculation unit 30 calculates the propensity score based on the history of the communication score of the subject and correction information set in advance for the subject. With this configuration, the propensity score of the subject can be calculated with high accuracy.

[0056] The technical scope of the present invention is not limited to the above embodiment, and can be modified as appropriate without departing from the spirit of the present invention. For example, in the above embodiment, the configuration for detecting the position information of the subject based on the photographing data of the photographing device 60 is described as an example, but the present invention is not limited to this. For example, the configuration for detecting the position information of the subject using other means such as an infrared sensor or GPS may be used. [Explanation of symbols]

[0057] P, P1 to P9 subjects, Q, Q1, Q2... objects, 10... isolation score calculation unit, 20... individual behavior score calculation unit, 30... tendency score calculation unit, 40... communication score calculation unit, 50... storage unit, 60... imaging device, 100... score calculation device

Claims

1. an individual behavior score calculation unit that calculates an individual behavior score indicating a degree to which the subject is actively performing an individual behavior with respect to a predetermined group based on the subject's behavior information; A propensity score calculation unit that calculates a propensity score indicating a tendency of each subject regarding interactions with others; a communication score calculation unit that calculates a communication score indicating a degree of communication with others in the group for each of the subjects during a target period based on the individual behavior score and the propensity score; A score calculation device comprising:

2. When the personal behavior score calculation unit detects, as the behavior information, that the subject has come close to an object for performing the personal behavior during the target period, the personal behavior score calculation unit calculates the personal behavior score based on the type of the object for each detection. The score calculation device according to claim 1 .

3. The propensity score calculation unit is The propensity score is calculated based on the communication score history of the subject and correction information previously set for the subject. The score calculation device according to claim 1 or 2.

4. Calculating an individual behavior score indicating a degree to which the subject is actively performing an individual behavior with respect to a predetermined group based on the subject's behavior information; Calculating a propensity score indicating a tendency for each subject to relate to others; calculating a communication score indicating a degree of communication with others in the group for each of the subjects based on the individual behavior score and the propensity score; A score calculation method including:

Images