JP2018112831A - Information processing apparatus, method of finding bullying, information processing system, and computer program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To further reduce a possibility of overlooking bullying.SOLUTION: There is provided an information processing apparatus comprising face image processing means for acquiring image information from photographing means, and digitizing an expression of a monitoring target person of bullying in the image information into an expression score indicating a higher numerical value as the expression is clearer; storage means for storing brightness of an ordinary expression of the monitoring target person as an ordinary expression score; and control means for performing comparison between the ordinary expression score and a present expression score of the monitoring target person, and comparison between the ordinary expression score and an another person expression score of the other person other than the monitoring target person, and for notifying an external device of alert information indicating a possibility of bullying occurrence in a case where at least one condition is met between a first condition of whether or not the present expression score is smaller than the ordinary expression score, and difference between the ordinary expression score and the present expression score is larger than a predetermined threshold; and a second condition of whether or not the present expression score is smaller than the other person expression score, and difference between the other person expression score and the present expression score is larger than a predetermined threshold.SELECTED DRAWING: Figure 1

Description

  The present invention relates to an information processing apparatus, a bullying discovery method, an information processing system, and a computer program for discovering bullying in a group life such as a school.

  Bullying in group life, especially bullying at school, is a serious problem that can seriously affect the sound and physical development of students and may cause school refusal, suicide and murder. Previously, there were measures to discover bullying, such as visual confirmation and life consultation by a homeroom teacher, but it is not only time consuming, but also depends on the skill etc. of the implementer, so the accuracy of bullying detection is not sufficient Absent. Therefore, in recent years, a method for stably finding bullying at low cost by making full use of technology such as IT (Information Technology) has been proposed.

  For example, although bullying itself is not detected, in Patent Document 1, in order to improve the accuracy of facial expression estimation, a facial expression score for final facial expression determination is not limited to a facial score for only a human face, A technique for estimating a facial expression in consideration of a human motion score is described.

  In Patent Document 2, in order to detect and deal with problems such as bullying at an early stage, a group is extracted based on the collected data. It describes techniques for sending emails and other information.

  In Patent Document 3, the technology for analyzing the situation by recording voice, noise from the microphone, noise generated by violence, screaming, screaming, etc., and the risk level exceeds a predetermined level, the civil welfare committee members and counselors have exercised It describes a technique for issuing an alarm when it is determined to be necessary.

  Japanese Patent Application Laid-Open No. 2004-151867 discloses a violent deterrent device that has an image capturing function (a so-called digital camera, and an image sensor is exposed to the outside of the case), and records an image captured by the image capturing function in a semiconductor memory element. Have been described.

JP 2014-206903 A JP 2010-176558 A JP 2003-346256 A JP 2002-245563 A

  Bullying victims may endure even if they are being bullied because bullying victims may be encouraged if their anger or sadness, etc., is expressed in emotions or actions. That is, when bullying is being performed, there is not always a great change in facial expression, behavior, speech and behavior.

  The technique of Patent Document 1 can detect relatively easy-to-understand facial expressions such as anger, sadness, and disgust, but cannot detect minute changes in facial expressions when enduring bullying. That is, in the technique of Patent Document 1, bullying may be overlooked. In the case of the technique described in Patent Literature 2-4, as in Patent Literature 1, there is a possibility that the victim is overlooked and the bullying in a state where the patient is patient with the expression.

  The present invention has been made to solve the above-described problems, and provides an information processing apparatus, a bullying discovery method, an information processing system, and a computer program that can further reduce the possibility of overlooking bullying. With the goal.

  The information processing apparatus according to the present invention acquires image information from a photographing unit, and the facial expression brightness of one or more persons including at least a bullying monitoring person in the image information is within a predetermined numerical range. A facial image processing means for digitizing and outputting a facial expression score indicating a higher numerical value as it is brighter; a storage means for storing the facial expression score obtained by quantifying the brightness of a normal facial expression of the monitoring subject as a normal facial expression score; A comparison between a normal facial expression score and a current facial expression score that is the current facial expression score of the person to be monitored; and an ordinary facial expression score and another person facial expression score that is the current facial expression score of another person other than the person to be monitored. The current facial expression score is smaller than the normal facial expression score, and the difference between the normal facial expression score and the current facial expression score exceeds a predetermined first threshold value At least one of a first condition and a second condition in which the current facial expression score is smaller than the other facial expression score and a difference between the other facial expression score and the current facial expression score exceeds a predetermined second threshold value And control means for notifying an external device of alert information indicating that bullying may have occurred.

  In the bullying discovery method of the present invention, the face image processing means acquires image information from the photographing means, and the brightness of facial expressions of one or more persons including at least a bullying monitoring person in the image information is determined by a predetermined value. A facial expression score that shows a higher numerical value as the facial expression becomes brighter within a numerical range, and is output in numerical form, and the facial expression score obtained by quantifying the brightness of the normal facial expression of the monitoring subject is stored in a storage means as a normal facial expression score, The control means compares the normal facial expression score with the current facial expression score that is the current facial expression score of the monitoring target person, and the normal facial expression score and the current facial expression score of another person other than the monitoring target person. Comparison with other facial expression score is performed, the current facial expression score is smaller than the normal facial expression score, and the difference between the normal facial expression score and the current facial expression score A second condition wherein the current facial expression score is smaller than the other facial expression score and a difference between the other facial expression score and the current facial expression score exceeds a predetermined second threshold. When at least one of the conditions is satisfied, alert information indicating that bullying may occur is notified to the external device.

  An information processing system according to the present invention includes a photographing unit, an information processing device, and an administrator terminal. The information processing device acquires image information from the photographing unit and monitors bullying in the image information. Facial image processing means for digitizing and outputting facial expression brightness of one or more persons including at least the target person into a facial expression score indicating a higher numerical value as the expression is brighter within a predetermined numerical range; Storage means for storing the expression score obtained by quantifying the brightness of the expression as a normal expression score, comparison between the normal expression score and the current expression score that is the current expression score of the person to be monitored; A comparison is made between the facial expression score and the other facial expression score, which is the current facial expression score of another person other than the monitored person, and the current facial expression score is higher than the usual facial expression score. First, a difference between the normal facial expression score and the current facial expression score exceeds a predetermined first threshold, and the current facial expression score is smaller than the other facial expression score and the other facial expression score and the current facial expression When at least one of the second conditions in which the difference from the score exceeds a predetermined second threshold is satisfied, alert information indicating that bullying may have occurred is notified to the external device. Control means.

  The computer program according to the present invention obtains image information from a photographing unit in a computer of an information processing apparatus, and determines the brightness of facial expressions of one or more persons including at least a bullying monitoring person in the image information. First processing for digitizing and outputting a facial expression score indicating a higher numerical value as the facial expression becomes brighter within the numerical range, and storing the facial expression score obtained by quantifying the brightness of the usual facial expression of the monitoring subject as a regular facial expression score A second process, a comparison between the normal facial expression score and the current facial expression score that is the current facial expression score of the monitoring subject, and the normal facial expression score and the current facial expression score of another person other than the monitoring subject. A comparison with a certain other facial expression score is performed, and the current facial expression score is smaller than the normal facial expression score, and the normal facial expression score and the current facial expression score are compared. A first condition in which a difference from a) exceeds a predetermined first threshold, and the current facial expression score is smaller than the other facial expression score, and a difference between the other facial expression score and the current facial expression score is a predetermined second threshold A third process for notifying an external device of alert information indicating that bullying may have occurred when at least one of the second conditions exceeding the second condition is satisfied It is a program.

  According to the present invention, the possibility of overlooking bullying can be further reduced.

It is a block diagram which shows the structural example of the information processing apparatus which concerns on the 1st Embodiment of this invention. It is a flowchart for demonstrating the operation example (bullying discovery method) of the information processing apparatus shown in FIG. It is a block diagram which shows the structural example of the information processing apparatus which concerns on the 2nd Embodiment of this invention. It is a block diagram which shows the structural example of the information processing system which concerns on the 3rd Embodiment of this invention. It is a graph which shows the relationship between a usual expression score and a present expression score. It is a graph which shows the relationship between a usual expression score and an other person's facial expression score. It is a flowchart for demonstrating the operation example (bullying discovery method) of the information processing system shown in FIG. It is a block diagram which shows the structural example of the information processing system which concerns on the modification of 3rd Embodiment. It is the 1st alert display example in an administrator terminal in case alert information contains bullying detailed information. It is the 2nd alert display example in an administrator terminal in case alert information contains bullying detailed information.

[First Embodiment]
(Description of configuration)
FIG. 1 is a block diagram illustrating a configuration example of an information processing apparatus 10 according to the first embodiment of the present invention. The information processing apparatus 10 includes a face image processing unit 12 (an example of a face image processing unit), a storage unit 14 (an example of a storage unit), and a control unit 16 (an example of a control unit).

  The face image processing unit 12 acquires image information from a photographing unit (not shown in FIG. 1), and quantifies the brightness of facial expressions of one or more persons including at least a bullying monitoring target in the image information. Generate and output a facial expression score. The facial expression score is a numerical value within a predetermined range (for example, a numerical value range of 1 to 10), and is a score indicating a higher numerical value as the facial expression is brighter. For example, a facial expression score is lower with a strong facial expression.

  The storage unit 14 usually stores at least a facial expression score. The normal expression score is an expression score generated and stored by the face image processing unit 12 based on image information having an ordinary expression of the monitoring target person, and can be said to be an expression score of brightness when there is no expression.

  The control unit 16 compares the normal facial expression score with the current facial expression score (the current facial expression score of the monitoring target person), and the normal facial expression score and the other facial expression score (the current facial expression score of another person other than the monitoring target person) and Perform a comparison of When at least one of the first condition and the second condition is satisfied, the control unit 16 displays alert information indicating that bullying has occurred in the external device (in FIG. 1). (Not shown).

  In the above, the first condition is a condition in which the current facial expression score is smaller than the normal facial expression score and the difference between the normal facial expression score and the current facial expression score exceeds a predetermined first threshold. The second condition is a condition in which the current facial expression score is smaller than the other facial expression score and the difference between the other facial expression score and the current facial expression score exceeds a predetermined second threshold.

The first threshold value is a threshold value for determining whether or not the drop in the facial expression of the person to be monitored is a drop that can be determined to be bullying. The second threshold value is a threshold value for determining whether the difference between facial expressions of the monitoring subject and other places can be determined as bullying. The first threshold value and the second threshold value may be determined based on knowledge such as ergonomics, or may be determined based on an empirical rule of an administrator. Note that the first threshold value and the second threshold value may be different values or the same value.
(Description of operation)
FIG. 2 is a flowchart for explaining an operation example (bullying discovery method) of the information processing apparatus 10 shown in FIG.

  First, prior to the execution of the operation of the flowchart, the storage unit 14 usually stores facial expression scores.

  The face image processing unit 12 calculates facial expression scores of one or more persons including at least a bullying monitoring target person in the image information (step S1).

  The control unit 16 compares the facial expression scores (specifically, a comparison between the regular facial expression score and the current facial expression score, and a comparison between the regular facial expression score and the other facial expression score) (step S2). When at least one of the first condition and the second condition is satisfied (Yes in Step S3), the control unit 16 notifies the external device of the alert information (Step S4).

Note that the detailed configuration and operation of the first embodiment will be described in a third embodiment to be described later.
(Explanation of effect)
In the first embodiment described above, whether or not bullying (in other words, alert information notification determination) is determined is whether the brightness of the current facial expression of the monitoring target person is lower than that of a normal facial expression, that is, dark ( First condition) or lower than a facial expression of a person around, that is, dark (second condition). By making the determination as described above, for example, it is possible to find a bullying so that the facial expression of the monitoring subject is slightly clouded. Or, even if the person being monitored is patient and kills the emotions and is patient enough not to change the expression as much as usual, the surrounding person's facial expression is brighter than necessary (that is, the facial expression of the surrounding person) If the score is relatively high), it can be determined that bullying may have occurred.

Summarizing the above, according to the first embodiment, even if bullying is such that the facial expression of the monitoring subject changes little or not so much, bullying is found with higher accuracy. This makes it possible to reduce the possibility of overlooking bullying.
[Second Embodiment]
FIG. 3 is a block diagram showing a configuration example of the information processing apparatus 50 according to the second embodiment of the present invention. The information processing apparatus 50 includes a storage unit 52 and a calculation unit 54 including a processor and the like. The storage unit 52 is a computer-readable recording medium and stores a computer program 60. The computer program 60 is, for example, a program for causing the calculation unit 54 to execute the processing shown in FIG.

  According to the second embodiment described above, as with the first embodiment, even if the facial expression of the person being monitored changes little or not so much that it does not change at all, It becomes possible to find bullying with higher accuracy. Therefore, the possibility of overlooking bullying is lower.

Examples of the arithmetic unit 54 include a CPU (Central Processing Unit) and a memory that stores instructions executed by the CPU. The computer-readable recording medium is, for example, a non-transitory storage device. Examples of the non-transitory storage device include an optical disk, a magnetic disk, a ROM (Read Only Memory), a portable medium such as a nonvolatile semiconductor memory, and a hard disk built in the computer system. The computer-readable recording medium may be a temporary storage device. As an example of a temporary storage device, for example, a communication line in the case of transmitting a program via a network such as the Internet or a communication line such as a telephone line, or a volatile memory inside a computer system can be cited.
[Third Embodiment]
(Description of configuration)
FIG. 4 is a block diagram illustrating a configuration example of an information processing system 100 according to the third embodiment of the present invention. In this embodiment, a case where “school” is applied to this system is taken as an example, but this system can also be applied to group life (for example, a company) other than school.

  The information processing system 100 includes an information processing device 102 that detects bullying, a camera 200 (an example of a photographing unit), a microphone 210 (an example of a recording unit), an administrator terminal 220 (an example of an external device), and a receiver A terminal 230 (an example of an external device).

  The information processing apparatus 102 includes a face image processing unit 104, a storage unit 106, a sound processing unit 108, and a control unit 110.

  The face image processing unit 104 includes a face image authentication unit 120 and a facial expression score calculation unit 122.

  The face image authentication unit 120 detects a face image from images acquired from the camera 200. The detection of the face image can be realized by using a known technique (for example, template matching or the like). Then, the face image authentication unit 120 performs personal authentication by face authentication of the detected face image. The face authentication can be realized using a known technique (for example, collation between a feature pattern of a detected face image and a feature pattern registered in advance).

  The expression score calculation unit 122 calculates an expression score for each face image (face image information) for which face authentication has been performed. The calculated facial expression score is output to the control unit 110. Control unit 110 stores the facial expression score in storage unit 106. In this case, a facial expression score is stored for each authenticated individual. For example, identification information may be given for each individual, and the identification information and the facial expression score may be stored in association with each other.

  The expression score is a score obtained by digitizing the brightness of a person's facial expression for each face image for which face authentication has been performed so that the brighter the value is within a predetermined numerical range. For example, as shown in FIGS. 5 and 6, the facial expression score is expressed in 10 levels within a range from 0 to 10.

  FIG. 5 is a graph showing a relationship between a normal facial expression score (a normal facial expression score of the monitoring target person) and a current facial expression score (the current facial expression score of the monitoring target person). In FIG. 5, the vertical axis represents the expression score (10 levels), and the horizontal axis represents time (newer as it goes to the right). In FIG. 5, the usual facial expression score is indicated by a black circle (facial expression score = 8), and the current facial expression score is indicated by a white circle (facial expression score = 4).

  FIG. 6 is a graph showing the relationship between the usual facial expression score and the other facial expression score (the current facial expression score of another person other than the person to be monitored). In FIG. 6, the vertical axis indicates facial expression scores (10 levels), and the horizontal axis indicates a person. In FIG. 6, the other person's facial expression score is indicated by a black circle (facial expression score = 7), and the current facial expression score is indicated by a white circle (facial expression score = 4).

  Of course, the numerical range of the expression score and the increment of the numerical value are not limited to the above. For example, regarding the numerical range, -5 to +5 may be divided into ten. In addition, the numerical increment is not limited to an integer increment, for example, and can be divided into 1000 divisions (ie, 0.1 increments) between 0 and 100, for example.

  The calculation of the facial expression score can be realized by using a known technique. For example, a certain facial expression is quantified for each basic element of the facial expression (for example, “joy”, “anger”, “sadness”, “disgust”, etc.). Then, the facial expression score calculation unit 122 uses these numerical values to digitize the “brightness” of the facial expression. For example, the facial expression score calculation unit 122, from a plurality of pieces of facial image information within a certain period of time, a positive score indicating the facial expression score when the expression is brighter and positive, such as “joy” and “laughter”, and “anger” , "Sadness", "disgust", etc., the expression of the negative score indicating the expression score when the expression is darker and negative, and this ratio (or a value obtained by multiplying it by an integer) can be used as the expression score. . The ratio may be calculated by the control unit 110. By using the ratio of facial expressions within a certain time period as the facial expression score, it is possible to prevent the facial expression score from being determined only by a momentary facial expression, and to further improve the detection accuracy of bullying. Alternatively, the facial expression score calculation unit 122 can use a positive score (for example, a “joy” score) itself as the facial expression score. If the basic element includes a “true face” with a serious face, the expression score calculation unit 122 can also use it as an expression score.

  The storage unit 106 usually stores at least a facial expression score. Furthermore, the storage unit 106 may store various types of data used by the face image processing unit 104 (such as a template for a face image region and a feature pattern of an individual facial expression).

  It should be noted that the usual facial expression score may be a score acquired at a predetermined time (such as at the time of admission or at the start of work), or may be updated as appropriate. When the facial expression score is normally updated, the facial expression score calculation unit 122 acquires, for example, a facial image in a class that is less likely to be bullied and is more likely to have no facial expression, and calculates the facial expression score. The calculated facial expression score is output to the control unit 110 and then stored in the storage unit 106. At this time, the control unit 110 adopts the calculated facial expression score itself as a new ordinary facial expression score, or calculates an average value of the calculated facial expression score and the previous ordinary facial expression score stored in the storage unit 106. Then, the calculation result can be adopted as a new ordinary facial expression score. Of course, the method of updating the facial expression score is not limited to the above. By updating the facial expression score, it is possible to carry out bullying detection with higher accuracy in line with the actual situation.

  The sound processing unit 108 acquires recording information from the microphone 210, detects bullying sound information that is sound information related to bullying from the recording information, and information on at least one of bullying sound information and presence / absence information of bullying sound information Is output to the control unit 110. The bullying sound information is at least one of an illegal word, hoarseness, and assault sound. Bullying sound detection from recorded information can be realized by using a known voice recognition technique (for example, voice pattern matching).

  The control unit 110 compares the normal facial expression score with the current facial expression score (the current facial expression score of the monitoring target person), and compares the normal facial expression score with the other facial expression score (the current facial expression score of another person other than the monitoring target person). A comparison is executed to determine whether or not bullying sound information exists.

  The control unit 110 determines whether or not at least one of the first condition and the second condition is satisfied.

  The first condition is a condition where the current facial expression score is smaller than the normal facial expression score and the difference between the normal facial expression score and the current facial expression score exceeds a predetermined first threshold.

  The second condition is a condition in which the current facial expression score is smaller than the other facial expression score and the difference between the other facial expression score and the current facial expression score exceeds a predetermined second threshold.

  The first threshold value is a threshold value for determining whether or not the drop in the facial expression of the person to be monitored can be determined to be bullying.

  The second threshold value is a threshold value for determining whether the difference between facial expressions of the monitoring subject and other places can be determined as bullying.

  The first threshold value and the second threshold value may be determined based on knowledge such as ergonomics, or may be determined based on an empirical rule of an administrator. Note that the first threshold value and the second threshold value may be different values or the same value.

  In this embodiment, a case where the current facial expression score is “8” (see FIG. 5) and the other facial expression score is “7” (see FIG. 6) is taken as an example. Further, in the present embodiment, a case where the first threshold is “3” (see FIG. 5) and the second threshold is “2” (see FIG. 6) is taken as an example.

  For example, in the case of FIG. 5, the current facial expression score “4” is smaller than the normal facial expression score “8”, and the difference “4” between the normal facial expression score “8” and the current facial expression score “4” is the first threshold “3”. Therefore, the first condition is satisfied. In the case of FIG. 6, the current facial expression score “4” is smaller than the other facial expression score “7”, and the difference “3” between the usual facial expression score “7” and the current facial expression score “4” is the second threshold “2”. The second condition is satisfied.

  Needless to say, the normal facial expression score, the current facial expression score, the other facial expression score, the first threshold value, and the second threshold value are not limited to the above values.

  Each of the first threshold value and the second threshold value may be a fixed value or a variable value. When the first threshold value and the second threshold value are variable values, for example, the first condition and the second condition can be more easily satisfied by reducing these values. In other words, the possibility of overlooking bullying can be further reduced.

  When at least one of the first condition and the second condition is satisfied, the control unit 110 notifies the administrator terminal 220 of alert information indicating that bullying may occur. On the other hand, when neither the first condition nor the second condition is satisfied, the control unit 110 determines whether or not the bullying sound information exists, and notifies the alert information when the bullying sound exists.

  The camera 200 is connected to the information processing apparatus 102 directly or via a network such as a LAN (Local Area Network). The camera 200 captures an image of a predetermined imaging area, acquires image information, and outputs the acquired image information to the information processing apparatus 102. The camera 200 is, for example, a digital camera or a digital video camera. The image information may be a still image or a moving image. Further, the camera 200 may output shooting information (for example, position information of the camera 200, shooting time information, etc.) together with the image information.

  The microphone 210 is connected to the information processing apparatus 102 directly or via a network such as a LAN. The microphone 210 records sound in a predetermined recording area to acquire recording information, and outputs the acquired recording information to the information processing apparatus 102. The microphone 210 may output recording information (for example, position information of the microphone 210, recording time information, etc.) together with the recording information.

  The camera 200 and the microphone 210 are generally often provided side by side, but may be provided at different positions.

  The administrator terminal 220 is a terminal held by a person who monitors bullying in the school (for example, a homeroom teacher or a principal). The administrator terminal 220 is connected to the information processing apparatus 102 directly or via a network such as a LAN.

The recipient terminal 230 is a terminal carried by a person who receives a report on bullying (for example, a guardian or a PTA (Parent-Teacher Association)). The recipient terminal 203 is connected to the information processing apparatus 102 directly or via a network such as a LAN.
(Description of operation)
FIG. 7 is a flowchart for explaining an operation example (bullying discovery method) of the information processing system 100 shown in FIG. Prior to execution of the operation of this flowchart, it is assumed that the storage unit 106 normally stores a facial expression score.

  The face image authentication unit 120 detects a face image from images acquired from the camera 200. Then, the face image authentication unit 120 performs face authentication of the detected face image (step S10). The expression score calculation unit 122 calculates an expression score for each face image for which face authentication has been performed (step S11). The calculated facial expression score is output to the control unit 110.

  The sound processing unit 108 acquires recording information from the microphone 210, and detects bullying sound information (for example, illegal word, hoarseness, and assault sound) related to bullying from the recording information (step S12). The sound processing unit 108 outputs at least one of bullying sound information and presence / absence information of bullying sound information to the control unit 110.

  The control unit 110 performs comparison of facial expression scores (specifically, comparison between a regular facial expression score and a current facial expression score, and a comparison between a regular facial expression score and another person facial expression score) (step S13). When at least one of the first condition and the second condition is satisfied (Yes in Step S14), the control unit 110 notifies the administrator terminal 220 of the alert information (Step S15).

  On the other hand, when neither the first condition nor the second condition is satisfied (No in step S14), the control unit 110 confirms the presence or absence of bullying sound information (step S16). When the bullying sound information exists (Yes in step S16), the control unit 110 notifies the administrator terminal 220 of the alert information (step S15). When the bullying sound information does not exist (No in step S16) and when the notification of the alert information is completed, the process of this flowchart ends.

The administrator who is notified of the alert information implements appropriate measures (for example, listening to circumstances related to the on-site personnel, providing guidance to the bullying offender, and providing educational consultation to the monitoring subject).
(Explanation of effect)
In the third embodiment described above, whether or not bullying is present (in other words, alert information notification determination) is performed when the brightness of the current facial expression of the monitoring subject is lower than that of a normal facial expression, that is, dark ( First condition) or lower than a facial expression of a person around, that is, dark (second condition). By determining as described above, for example, it is possible to determine bullying even when the facial expression of the monitoring subject is slightly cloudy. Or, even if the person being monitored is patient and kills the emotions and is patient enough not to change the expression as much as usual, the surrounding person's facial expression is brighter than necessary (that is, the facial expression of the surrounding person) If the score is relatively high, it can be determined that bullying has occurred.

  Summarizing the above, according to the third embodiment, even if bullying is such that the facial expression of the monitoring subject changes little or not so much, bullying is found with higher accuracy. This makes it possible to reduce the possibility of overlooking bullying.

Furthermore, in the third embodiment, the presence or absence of bullying sound information is checked when neither the first condition nor the second condition is satisfied, and if bullying sound information exists, the alert information is sent to the administrator terminal 220. Has been notified. As a result, for example, it is possible to detect a bully that the person to be monitored pushes his emotions and corrects his / her normal facial expression, and that the surrounding people have no expression. Therefore, according to the present embodiment, the possibility of overlooking bullying can be further reduced.
(Description of modification)
In the third embodiment described above, the alert information output by the control unit 110 may include bullying detailed information. The bullying detailed information is, for example, at least one information of image information, position information of the camera 200, shooting time information, recording information, bullying sound information, position information of the microphone 210, recording time information, and information for specifying a monitoring target person. is there.

  FIG. 9 shows a first alert display example on the administrator terminal 220 when the alert information includes bullying detailed information. FIG. 10 shows a second alert display example on the administrator terminal 220 when the alert information includes bullying detailed information.

  The alert information may include bullying prediction information. For example, the control unit 110 analyzes the bullying detailed information to predict a bullying state (for example, a place where bullying is performed, the number of bullying, the content of bullying, etc.) in the future (for example, several hours after the current time). Create information. Then, the control unit 110 includes the bullying prediction information alone or together with the bullying detailed information in the alert information. In particular, when alert information includes bullying detail information and bullying prediction information, the administrator can grasp the behavior before and after the monitored person and the bullying perpetrator, so that the actual state of bullying can be grasped more accurately. Is possible.

  In the third embodiment, the information processing apparatus 102 transmits the alert information only to the administrator terminal 220, but the transmission destination of the alert information is not limited to the above. For example, the alert information may be transmitted to the receiver terminal 230 or other communication device (not shown in FIG. 4, for example, a police authority device or a communication terminal of a bullying offender's guardian). For example, the information processing apparatus 102 may transmit alert information to the recipient terminal 230 when the alert information is transmitted to the administrator terminal 220 a predetermined number of times (for example, three times).

  In addition to the configuration shown in FIG. 4, the information processing apparatus 102 can include a bullying detection history database 300 as shown in FIG. 8. The bullying detection history database 300 stores a list of bullying discovered by the information processing apparatus 102 (in other words, contents notified as alert information).

  Furthermore, the bullying detection history database 300 can provide not only a list of bullying but also additional information (treatment content, degree of improvement, etc.) for each bullying. The additional information may be written or rewritten by a direct operation of the information processing apparatus 102, or may be written or rewritten from the administrator terminal 220. By doing in this way, the administrator can grasp at a glance the number of bullying and the subsequent progress.

  In the third embodiment described above, the hardware configuration of the information processing apparatus 102 is not limited to the configuration illustrated in FIG. The hardware configuration of the information processing apparatus 102 may be, for example, the configuration illustrated in FIG. 3 (that is, a configuration in which a calculation unit including a processor realizes a function by executing a computer program).

  Further, the first to third embodiments described above are not limited to schools, and can be used in any group life such as a company.

  As mentioned above, although this invention was demonstrated using each embodiment, the technical scope of this invention is not limited to description of said each embodiment. It is obvious to those skilled in the art that various modifications or improvements can be added to the above embodiments. Therefore, it is needless to say that embodiments with such changes or improvements are also included in the technical scope of the present invention. The numerical values and names of the components used in the embodiments described above are illustrative and can be changed as appropriate.

DESCRIPTION OF SYMBOLS 10 Information processing apparatus 12 Face image processing part 14 Storage part 16 Control part 50 Information processing apparatus 52 Storage part 54 Operation part 60 Computer program 100 Information processing system 102 Information processing apparatus 104 Face image processing part 106 Storage part 108 Sound processing part 110 Control Unit 120 face image authentication unit 122 expression score calculation unit 200 camera 210 microphone 220 administrator terminal 230 receiver terminal 300 bullying detection history database

Claims (10)

An image score obtained by acquiring image information from the photographing means, and indicating the brightness of the facial expression of one or more persons including at least a bullying monitoring person in the image information, the higher the facial expression is within a predetermined numerical range Face image processing means for digitizing and outputting to,
Storage means for storing the facial expression score obtained by quantifying the brightness of the normal facial expression of the monitored person as a normal facial expression score;
Comparison between the normal facial expression score and the current facial expression score that is the current facial expression score of the monitoring subject, and an other facial expression score that is the current facial expression score of the other than the normal facial expression score and the monitoring subject A first condition in which the current facial expression score is smaller than the normal facial expression score and a difference between the normal facial expression score and the current facial expression score exceeds a predetermined first threshold, and the current facial expression score is Bullying occurs when at least one of the second conditions is smaller than the other facial expression score and the difference between the other facial expression score and the current facial expression score exceeds a predetermined second threshold. Control means for notifying the external device of alert information that there is a possibility that
An information processing apparatus comprising: Sound processing means for obtaining recording information from the recording means, detecting bullying sound information related to bullying from the recording information, and outputting at least one of the bullying sound information and the presence / absence information of the bullying sound information; Prepared,
When neither of the first condition and the second condition is satisfied, the control means determines whether or not the bullying sound information is present, and if the bullying sound is present, the control unit displays the alert information. The information processing apparatus according to claim 1, wherein notification is performed.   The information processing apparatus according to claim 2, wherein the bullying sound information is at least one of an illegal word, a hoarseness, and an assault sound.   The information processing apparatus according to claim 2, wherein the alert information includes bullying detailed information.   The bullying detailed information includes the image information, position information of the photographing means, photographing time information of the image information, recording information, bullying sound information, position information of the recording means, recording time information of the recording information, The information processing apparatus according to claim 4, wherein the information processing apparatus includes at least one piece of information that specifies a person to be monitored.   6. The information processing apparatus according to claim 5, wherein the alert information is obtained by analyzing the bullying detailed information, and includes bullying prediction information for predicting a future bullying situation.   The information processing apparatus according to claim 1, wherein the usual facial expression score is updated from a value registered first. The image information is acquired from the photographing means by the face image processing means, and the brightness of the facial expression of one or more persons including at least the bullying monitoring person in the image information is expressed as the expression becomes brighter within a predetermined numerical range. Digitize and output high facial expression score
The facial expression score obtained by quantifying the brightness of the normal facial expression of the monitoring subject is stored in a storage means as a normal facial expression score,
The control means compares the normal facial expression score with the current facial expression score that is the current facial expression score of the monitoring target person, and the normal facial expression score and the current facial expression score of another person other than the monitoring target person. Performing a comparison with another facial expression score, a first condition in which the current facial expression score is smaller than the normal facial expression score and a difference between the normal facial expression score and the current facial expression score exceeds a predetermined first threshold; and When a current facial expression score is smaller than the other facial expression score and a difference between the other facial expression score and the current facial expression score exceeds a predetermined second threshold, at least one of the second conditions is satisfied A bullying discovery method characterized by notifying an external device of alert information indicating that bullying may have occurred. Photographing means;
An information processing device;
An administrator terminal,
With
The information processing apparatus includes:
A facial expression that acquires image information from the photographing means, and shows the brightness of the facial expression of one or more persons including at least a bullying monitoring person in the image information as the facial expression is brighter within a predetermined numerical range. Facial image processing means for converting the score into a numerical value and outputting the score;
Storage means for storing the facial expression score obtained by quantifying the brightness of the normal facial expression of the monitored person as a normal facial expression score;
Comparison between the normal facial expression score and the current facial expression score that is the current facial expression score of the monitoring subject, and an other facial expression score that is the current facial expression score of the other than the normal facial expression score and the monitoring subject A first condition in which the current facial expression score is smaller than the normal facial expression score and a difference between the normal facial expression score and the current facial expression score exceeds a predetermined first threshold, and the current facial expression score is Bullying occurs when at least one of the second conditions is smaller than the other facial expression score and the difference between the other facial expression score and the current facial expression score exceeds a predetermined second threshold. Control means for notifying the external device of alert information that there is a possibility that
An information processing system comprising: In the computer of the information processing device,
An image score obtained by acquiring image information from the photographing means, and indicating the brightness of the facial expression of one or more persons including at least a bullying monitoring person in the image information, the higher the facial expression is within a predetermined numerical range A first process for digitizing and outputting
A second process for storing the facial expression score obtained by quantifying the brightness of the normal facial expression of the person to be monitored as a normal facial expression score;
Comparison between the normal facial expression score and the current facial expression score that is the current facial expression score of the monitoring subject, and an other facial expression score that is the current facial expression score of the other than the normal facial expression score and the monitoring subject A first condition in which the current facial expression score is smaller than the normal facial expression score and a difference between the normal facial expression score and the current facial expression score exceeds a predetermined first threshold, and the current facial expression score is Bullying occurs when at least one of the second conditions is smaller than the other facial expression score and the difference between the other facial expression score and the current facial expression score exceeds a predetermined second threshold. And a third process of notifying an external device of alert information indicating that there is a possibility of being executed.

Images