JP6999399B2 - Abnormality judgment device - Google Patents

Description

The present invention relates to a technique for detecting a person who behaves abnormally.

Techniques for detecting people who behave abnormally have been proposed. For example, Patent Document 1 describes a device for determining whether or not a person is behaving abnormally based on the staying time of the person captured by the camera in a predetermined area.

Japanese Unexamined Patent Publication No. 2006-11728

According to the technique described in Patent Document 1, even if a person stays in a blind spot area that is not captured by any camera for a long time, the abnormal behavior is not detected. Therefore, in order to detect a person who is behaving abnormally without omission, it is necessary to arrange a large number of cameras so that a blind spot area does not occur.

In view of the above circumstances, the present invention provides a means for detecting a person who is behaving abnormally even if a blind spot area of the camera is generated.

In order to solve the above-mentioned problems, the present invention uses the plurality of cameras to set a region of the monitored region that is not an imaging region of any of the plurality of cameras discretely arranged in the monitored region as a blind spot region. Based on the captured image, the time during which the subject stays in the blind spot region when the subject moves on the movement path from the first imaging region to the blind spot region and exits to the second imaging region is specified, and the time spent in the blind spot region is specified. The probability of occurrence is calculated based on the probability distribution of the time that a general passerby stays in the blind spot region when moving on the movement path, and based on the probability of occurrence , is the behavior of the subject abnormal? An abnormality determination device for determining whether or not to use is provided as a first aspect.

According to the abnormality determination device of the first aspect, it is possible to detect a subject who behaves abnormally even if a blind spot region is generated.

In the abnormality determination device of the first aspect, the monitored area includes a plurality of the blind spot areas.
The configuration is such that the occurrence probability is calculated for each of the plurality of blind spot regions, and whether or not the behavior of the subject is abnormal is determined based on the occurrence probability calculated for each of the plurality of blind spot regions. May be adopted as the second aspect.

According to the abnormality determination device of the second aspect, it is possible to detect a subject who is behaving abnormally even when there are a plurality of blind spot areas.

In the abnormality determination device of the first aspect or the second aspect, when it is determined that the behavior of the subject is abnormal, the configuration of notifying the user may be adopted as the third aspect.

According to the abnormality determination device of the third aspect, the user can know the target person who is behaving abnormally.

In the abnormality determination device of the first to third aspects, the feature information of the subject extracted from the image captured by any of the plurality of cameras is stored, and the feature information is used to store the feature information of the plurality of cameras. By identifying the person and the subject in the image captured by any of the above, the period during which the subject stayed in any of the imaging regions of the plurality of cameras is specified , and the period is specified. Based on this, a configuration in which the subject stays in the blind spot region for a certain period of time may be adopted as the fourth aspect.

According to the abnormality determination device of the fourth aspect, the staying time in each of the imaging region and the blind spot region of a certain subject can be known.

In the abnormality determination device of the fourth aspect, after the time during which the characteristic information of the subject is not extracted from the image captured by any of the plurality of cameras continues to be equal to or longer than the threshold value, the characteristic information of the subject is described. A configuration in which memory is not maintained may be adopted as the fifth aspect.

According to the abnormality determination device of the fifth aspect, the inconvenience that the information about the target person who has gone out of the monitoring target area is unnecessarily stored is avoided.

The figure which showed the structure of the abnormality determination system which concerns on one Embodiment. The figure which exemplifies the imaging area of a plurality of cameras which concerns on one Embodiment. The block diagram which showed the hardware configuration of the abnormality determination apparatus which concerns on one Embodiment. The figure which illustrated the data structure of the person table which concerns on one Embodiment. The figure which illustrated the data structure of the detection log DB which concerns on one Embodiment. The figure which exemplifies the movement path of the person detected in the abnormality determination system which concerns on one Embodiment. The block diagram which showed the functional structure of the abnormality determination apparatus which concerns on one Embodiment. The figure which showed the flow of the process performed by the abnormal behavior determination part of the abnormality determination apparatus which concerns on one Embodiment. The figure which illustrated the data structure of the transition probability table which concerns on one Embodiment. The figure which illustrated the data structure of the normal distribution parameter table which concerns on one Embodiment. The figure which illustrated the data structure of the undetected time table which concerns on one Embodiment. A graph illustrating the relationship between the probability and the degree of abnormality used by the abnormal behavior determination unit of the abnormality determination device according to the embodiment. The figure which exemplifies the movement path of the person detected in the abnormality determination system which concerns on one Embodiment. The figure exemplifying the screen displayed on the display of the abnormality determination system which concerns on one Embodiment. A graph illustrating the relationship between the staying time and the probability used in the abnormality determination system according to one modification. A graph illustrating the relationship between the staying time and the probability used in the abnormality determination system according to one modification. The figure which exemplifies the arrangement of a plurality of cameras which concerns on one modification. The figure which exemplifies the arrangement of a plurality of cameras which concerns on one modification.

<Embodiment>
Hereinafter, the abnormality determination system 1 according to the embodiment of the present invention will be described. The abnormality determination system 1 is a system for determining whether or not the behavior of a passerby is abnormal. In the following description, it is assumed that the monitored area of the abnormality determination system 1 is the passage S in the station yard, but the monitored area of the abnormality determination system 1 may be any area as long as it is an area where passersby pass. ..

FIG. 1 is a diagram showing the configuration of the abnormality determination system 1. The abnormality determination system 1 illustrated in FIG. 1 includes an abnormality determination device 11, a display 12, an operation device 13, a plurality of cameras 14A to 14F (hereinafter collectively referred to as “camera 14”), and a speaker 15.

The abnormality determination device 11 determines whether or not the behavior of a passerby (monitoring target person) in the passage S is abnormal based on images captured by a plurality of cameras 14 discretely arranged in the monitoring target area. It is a device.

The display 12 is connected to the abnormality determination device 11 and displays various information such as the result of the determination made by the abnormality determination device 11 to the user of the abnormality determination device 11. The operation device 13 is a keyboard, a mouse, or the like, and is connected to the abnormality determination device 11 to receive operations such as data input performed by the user of the abnormality determination device 11 to the abnormality determination device 11.

Each of the cameras 14 is connected to the abnormality determination device 11, and image data representing an image captured at a predetermined time interval in the imaging region is sequentially transmitted to the abnormality determination device 11. FIG. 2 is a diagram showing each imaging region of the plurality of cameras 14. The camera 14A captures the imaging region a, the camera 14B captures the imaging region b, the camera 14C captures the imaging region c, the camera 14D captures the imaging region d, the camera 14E captures the imaging region e, and the camera 14F captures the imaging region f. do. The passage S is a one-way street, and it is assumed that passersby pass in the direction from the left to the right in FIG.

As shown in FIG. 2, in the passage S which is the image pickup target area, there are blind spot areas g, h, and i which are not the image pickup areas of any of the cameras 14.

With reference to FIG. 1, the description of the configuration of the abnormality determination system 1 will be continued. The speaker 15 is connected to the abnormality determination device 11 and emits a message voice output by the abnormality determination device 11 to passersby in the management target area.

FIG. 3 is a block diagram showing the hardware configuration of the abnormality determination device 11. The hardware of the abnormality determination device 11 is a computer including a processor 1101, a memory 1102, and an interface 1103. The processor 1101 performs various data processing according to the program stored in the memory 1102. The memory 1102 stores various data such as a program executed by the processor 1101. The interface 1103 is connected to devices such as a display 12, an operating device 13, a camera 14, and a speaker 15, and inputs / outputs data between these devices and the processor 1101 or the memory 1102.

In the memory 1102, a person table, which is a database for storing characteristic information of a passerby (target person) passing through the monitored area and an abnormality / normal judgment result, and a detection result of a passerby passing through the monitored area are stored. The detection log DB, which is a database stored in series, is stored.

FIG. 4 is a diagram illustrating the data structure of the person table. The person table is a collection of data records for each of the people detected in the monitored area, and each data record is a person ID (Identifier) that identifies the person, feature information indicating features extracted from the image of the person, and so on. In addition, it includes a determination result indicating whether the behavior of the person is normal or abnormal.

FIG. 5 is a diagram illustrating the data structure of the detection log DB. The detection log DB is a collection of data tables corresponding to each of the detected persons, and in each data table, the start time of the period in which the person is detected from the image captured by any of the cameras 14 (detection start). A data record including a time), an end time (detection end time) of the period, and a camera ID for identifying the camera 14 that captured the image is stored, for example, in ascending order of the detection start time.

The data in the data table relating to the example person ID "1234" in FIG. 5 is that the person identified by the person ID "1234" stays in the imaging region a for 7 seconds and then is between the imaging region a and the imaging region b. After staying in the blind spot region g for 5 seconds, staying in the imaging region b for 4 seconds, staying in the blind spot region h between the imaging region b and the imaging region c for 6 seconds, and then staying in the imaging region c for 5 seconds. Shows. FIG. 6 is a diagram showing a movement path R1 of a person identified by the person ID “1234”, which is shown by the data in the data table relating to the person ID “1234” exemplified in FIG.

FIG. 7 is a block diagram showing a functional configuration of the abnormality determination device 11. That is, the processor 1101 operates as a device having the configuration shown in FIG. 7 by performing data processing according to the program according to the present embodiment. The functional configuration of the abnormality determination device 11 will be described below.

The image acquisition unit 111 acquires image data representing an image captured by the camera 14 from each of the plurality of cameras 14. The person detection unit 112 extracts characteristic information of a person from the image represented by the image data acquired by the image acquisition unit 111, and detects the person in the image based on the extracted characteristic information.

The person detection unit 112 identifies the similarity between the feature information extracted from the image and each of the feature information registered in the person table, and detects the person in the image in the past based on the identified similarity. The person registered in the person table is identified. In the present application, when the similarity between the two feature information is higher than a predetermined threshold value and the person in the image having the feature information can be regarded as the same person, the two feature information is the same. It shall be. If the person identified as the person in the image is not registered in the person table, the person detection unit 112 determines that a new person has been detected, adds a new data record to the person table, and newly detects the person. Stores the person ID and feature information of the person.

The detection log update unit 113 updates the detection log DB based on the detection result of a person by the person detection unit 112. The abnormal behavior determination unit 114 determines whether or not the behavior of the person detected by the person detection unit 112 is abnormal based on the data stored in the detection log DB, and writes the determination result in the person table. When the abnormal behavior determination unit 114 determines that the behavior of the passerby is abnormal, the notification unit 115 notifies the user (for example, a station staff) of the abnormality determination device 11 and the passerby to that effect. In the present embodiment, the notification unit 115 notifies the user such as the station staff by displaying an image on the display 12, and the notification unit 115 notifies the passerby by sounding the sound by the speaker 15.

The data deletion unit 116 relates to a person whose characteristic information is not extracted from any of the images captured by any of the plurality of cameras 14 by the person detection unit 112 based on the data stored in the detection log DB for a period of time exceeding the threshold value. , Delete the data related to the person from the person table and the detection log DB (the data related to the person is not maintained in the person table and the detection log DB). In addition to the person table and the detection log DB, the storage unit 117 stores various data generated when other components of the abnormality determination device 11 perform data processing.

FIG. 8 is a diagram showing a flow of processing performed by the abnormal behavior determination unit 114. The abnormal behavior determination unit 114 performs the process according to the flow of FIG. 8 for each of the persons (target persons) registered in the person table every time a predetermined time elapses.

First, the abnormal behavior determination unit 114 identifies the movement route of the target person based on the data of the target person stored in the detection log DB (step S101). For example, based on the data exemplified in FIG. 5, the abnormal behavior determination unit 114 specifies "imaging area a-> imaging area b-> imaging region c" as a movement route.

Subsequently, the abnormal behavior determination unit 114 is the probability that a general passerby moves from the image pickup area X to the image pickup area Y with respect to each of the combinations of two adjacent image pickup areas on the movement path specified in step S101. The probability p1 (X, Y) is specified (step S102).

FIG. 9 is a diagram showing a data structure of a transition probability table used by the abnormal behavior determination unit 114 in the process of step S102. The transition probability table stores numerical values indicating the probabilities p1 (X, Y) for each combination of the moving source imaging region X and the moving destination imaging region Y. The transition probability table is stored in the storage unit 117 in advance. The numerical values stored in the transition probability table are calculated based on, for example, a sample of the movement routes of a large number of passersby who have been inspected in the past in the monitored area, and the total of the numerical values in each row and each column is Both are "1". In the case of the data example of FIG. 9, for example, the probability p1 (imaging region a, imaging region b), which is the probability that a general passerby moves from the imaging region a to the imaging region b, is “0.3”.

Following step S102 (FIG. 8), the abnormal behavior determination unit 114 identifies the time t during which the subject stays in each of the blind spot areas on the movement route based on the data of the subject stored in the detection log DB. (Step S103). For example, based on the data exemplified in FIG. 5, the abnormal behavior determination unit 114 takes an image from the image pickup area b for "5 seconds" as the time spent in the blind spot area g on the movement path from the image pickup area a to the image pickup area b. "6 seconds" is specified as the time spent in the blind spot area h on the movement route to the area c.

Following step S103 (FIG. 8), the abnormal behavior determination unit 114 moves the general passerby from the imaging region X on the upstream side to the blind spot region for each of the blind spot regions on the movement path of the subject. , The probability p2 (X, Y, t), which is the probability of moving to the image pickup region Y on the downstream side after staying in the blind spot region for a time t, is specified (step S104).

ただし、μは上流側の撮像領域Ｘと下流側の撮像領域Ｙの組み合わせに応じた滞在時間ｔの平均値であり、σは上流側の撮像領域Ｘと下流側の撮像領域Ｙの組み合わせに応じた滞在時間ｔの標準偏差である。 In step S104, the abnormal behavior determination unit 114 calculates the probability p2 (X, Y, t) according to the following equation 1 (normal distribution).

However, μ is the average value of the staying time t according to the combination of the upstream side imaging region X and the downstream side imaging region Y, and σ is the average value according to the combination of the upstream side imaging region X and the downstream side imaging region Y. It is the standard deviation of the staying time t.

The probability p2 (X, Y, t) shown in Equation 1 is the imaging region (upstream imaging region) in which the subject stayed before entering the blind spot region among the plurality of imaging regions adjacent to the blind spot region. This is an example of a standard specified according to the combination of the imaging region (imaging region on the downstream side) that stayed after leaving the blind spot region.

FIG. 10 is a diagram showing a data structure of a normal distribution parameter table used by the abnormal behavior determination unit 114 in the process of step S104. In the normal distribution parameter table, the average value μ of the dwell time t in the blind spot region and the standard value are shown for each of the combination of the blind spot region, the imaging region on the upstream side of the blind spot region, and the imaging region on the downstream side of the blind spot region. The value of deviation σ is stored. The normal distribution parameter table is stored in the storage unit 117 in advance.

The calculation method of the numerical values stored in the normal distribution parameter table will be described below. FIG. 11 shows the time (undetected time) in which various persons stayed in the blind spot area based on the data recorded in the detection log DB (FIG. 5) regarding the passersby who passed through the monitored area in the past. It is a table grouped by the combination of the image pickup area on the upstream side and the image pickup area on the downstream side (hereinafter referred to as “undetected time table”).

For example, according to the data exemplified in FIG. 5, the subject identified by the person ID "1234" enters the blind spot area g from the imaging area a at 12:35:24 on December 1, 2017, for 5 seconds. Later, at 12:35:29 on December 1, 2017, it can be seen that the blind spot area g entered the imaging area b. In this case, "5" seconds is entered as the sample data of the undetected time in the column corresponding to the combination of the image pickup area a on the upstream side and the image pickup area b on the downstream side of the undetected time table (FIG. 14).

The numerical values stored in the normal distribution parameter table are the mean value μ and the standard deviation σ of the undetected time indicated by the sample data stored in the undetected time table. For example, according to the data exemplified in FIG. 11, the staying time (undetected time) in the blind spot area g of a passerby who has moved from the image pickup area a on the upstream side to the image pickup area b on the downstream side via the blind spot area g. The samples are 5 seconds, 6 seconds, 11 seconds, 3 seconds, and so on. The mean μ and standard deviation σ of these samples will be stored in the first row of the example normal distribution parameter table in FIG.

ただし、ｍｉｎ（ａ，ｂ）はａとｂのうち小さい方を示す関数であり、Ｌ_maxは異常度の上限値として設定された値である。 Following step S104 (FIG. 8), the abnormal behavior determination unit 114 calculates the degree of abnormality L (X, Y, t) for each of the blind spot regions on the movement path of the subject according to the following equation 2. Step S105).

However, min (a, b) is a function indicating the smaller of a and b, and L _max is a value set as the upper limit of the degree of abnormality.

FIG. 12 shows the probability p (X, Y, t) which is the product of the probability p1 (X, Y) and the probability p2 (X, Y, t) and the degree of abnormality L (X, Y, t) when the equation 2 is followed. ) Is a graph showing the relationship.

ただし、ｎは移動経路上の死角領域の数を示し、Ｘ_iはｉ番目の死角領域の上流側の撮像領域、Ｙ_iはｉ番目の死角領域の下流側の撮像領域、ｔ_iはｉ番目の死角領域における滞在時間を示す。 Following step S105 (FIG. 8), the abnormal behavior determination unit 114 calculates the total abnormality degree M according to the following equation 3 (step S106).

However, n indicates the number of blind spot regions on the movement path, X _i is the imaging region on the upstream side of the i-th blind spot region, Y _i is the imaging region on the downstream side of the i-th blind spot region, and t _i is the i-th. Shows the length of stay in the blind spot area.

Subsequently, the abnormal behavior determination unit 114 determines whether or not the behavior of the subject is abnormal based on the total abnormality degree M calculated in step S106 (step S107). Specifically, the abnormal behavior determination unit 114 determines that the total abnormality degree M is "normal" if it is equal to or less than a predetermined threshold value, and "abnormal" if it exceeds the threshold value.

Subsequently, the abnormal behavior determination unit 114 performs a process of writing the determination result in the person table (update of the person table) (step S108).

FIG. 13 is a diagram illustrating a movement path R2 in which the total degree of abnormality M is calculated higher than that of the movement path R1 exemplified in FIG. 6 and is likely to be determined as an abnormality. Although the detailed movement route of R2 shown in FIG. 13 (for example, the detailed movement route in the blind spot regions g and h) cannot be specified from the data stored in the detection log DB, the target person follows the movement route R2. When the movement is performed, the staying time in the blind spot areas g and h becomes longer than usual, so that the total degree of abnormality M becomes high, and the possibility that the behavior of the subject is determined to be abnormal increases.

The above is the process performed by the abnormal behavior determination unit 114. The abnormal behavior determination unit 114 determines that the behavior of the subject is abnormal, and the value of the "determination result" stored in the data record of the person table relating to the subject changes from "normal" to "abnormal". When the change is made, the notification unit 115 instructs the display 12 of the abnormality determination device 11 to indicate that a passerby having an abnormal behavior has been detected.

FIG. 14 is a diagram illustrating a screen displayed on the display 12 (hereinafter, referred to as a “monitoring screen”). In the area J of the monitoring screen, the position of a passerby currently detected in the monitored area is indicated by "○" or "●". “●” indicates the position of a passerby who is performing abnormal behavior, and “○” indicates the position of a passerby who does not have any abnormal behavior. Further, in the area K of the monitoring screen, an image showing a passerby who is performing an abnormal behavior is displayed. In the image of region K, the passerby surrounded by the ellipse is a passerby performing anomalous behavior.

The user (station staff, etc.) of the abnormality determination device 11 can check the position and behavior of a passerby who is performing an abnormal behavior by looking at the monitoring screen.

Further, when the value of the "determination result" stored in the data record of the person table is changed from "normal" to "abnormal", the notification unit 115 notifies the speaker 15 arranged in the monitored area to the speaker 15. For example, you will be instructed to pronounce a message such as "If you have any problems, please raise your hand. A station clerk will come." By sounding the message voice from the speaker 15, the passerby can receive the help of the station staff or the like when he / she needs the help of someone because he / she is lost, for example. In addition, when a passerby who is behaving abnormally is trying to commit a crime such as rubbing or voyeurism, the message voice is pronounced to suppress the behavior.

<Modification example>
The above embodiments can be modified in various ways. An example of these variations is shown below. In addition, two or more modification examples shown below may be combined appropriately.

(1) Specific data examples used in the explanation of the above-described embodiment, examples of processing flows, mathematical formulas used for determining whether or not the behavior is abnormal, etc. are merely examples, and other data and processing flows. , Formulas and the like may be used. For example, the abnormal behavior determination unit 114 is based on calculating the total abnormality degree according to the mathematical formulas shown in the formulas 1 to 3, but the calculation method of the total abnormality degree is not limited to this. As long as the determination is made using the staying time in the blind spot region, various methods can be adopted as the method used by the abnormal behavior determination unit 114 for the determination.

(2) In the above-described embodiment, the abnormal behavior determination unit 114 uses the normal distribution (Equation 1) of the time that a general passerby stays in the blind spot area to determine whether or not the behavior of the subject is abnormal. Judgment is made. The probability distribution used by the abnormal behavior determination unit 114 for determination is not limited to the normal distribution. For example, a probability distribution that directly reflects the distribution of sample values indicated by the data stored in the undetected time table (FIG. 11) may be used.

Further, the probability distribution used by the abnormal behavior determination unit 114 for determination is not limited to the probability distribution in which the probability p2 (X, Y, t) continuously changes according to the change in the staying time t as in the normal distribution, and FIG. The binary function illustrated in 1 and the probability distribution according to the multivalued function illustrated in FIG. 16 may be used.

The binary function shown in FIG. 15 has a range from the lower limit value t1 to the upper limit value t2, where the lower limit value is t1 and the upper limit value is t2 among the sample values shown by the data stored in the undetected time table (FIG. 11). This is a function in which the probability of occurrence of is set to "1".

Further, the multivalued function shown in FIG. 16 sets the sample value indicated by the data stored in the undetected time table (FIG. 11) to t1 (lower limit value) or more and less than t2, t2 or more and less than t3, t3 or more and less than t4, t4. It is a function that is divided into four groups of t5 (upper limit value) or less, and the value obtained by dividing the number of sample values belonging to those groups by the total number of sample values is set as the occurrence probability of those groups.

(3) In the above-described embodiment, the abnormal behavior determination unit 114 determines whether or not the behavior of the subject is abnormal based on the degree of abnormality using the probability p1 (X, Y) regarding the movement route. When the abnormal behavior determination unit 114 makes a determination, the movement route may not be taken into consideration.

(4) In the above-described embodiment, the abnormal behavior determination unit 114 does not use the staying time in the imaging region when determining whether or not the behavior of the subject is abnormal. Instead of this, when the abnormal behavior determination unit 114 makes a determination, the staying time in the imaging region may be taken into consideration.

(5) In the above-described embodiment, the abnormal behavior determination unit 114 determines whether or not the behavior of the subject is abnormal based on the total abnormality degree indicating the degree of abnormality regarding the entire movement path of the subject. Alternatively or additionally, the abnormal behavior determination unit 114 may determine whether or not the behavior of the subject is abnormal based on the degree of abnormality regarding a part of the movement route.

For example, even if the total abnormality degree M is equal to or less than a predetermined threshold value, if the abnormality degree L (X, Y, t) relating to any blind spot region on the movement path exceeds the predetermined threshold value, the abnormal behavior determination unit 114 may determine that the subject's behavior is abnormal.

(6) In the above-described embodiment, when the abnormality determination system 1 detects a person who is behaving abnormally, the abnormality determination system 1 notifies the user to that effect by displaying the display 12. Instead of or in addition to this, the user (station staff, etc.) of the abnormality determination device 11 may be notified by the pronunciation of an alert sound or a message voice.

(7) In the above-described embodiment, it is assumed that the passage S, which is the monitored area, is a one-way street. The monitored area of the abnormality determination system according to the present invention is not limited to a one-way passage, and may be a two-way passage, a floor capable of passing in any direction, or the like.

For example, if the monitored area is a bidirectional passage T, as illustrated in FIG. 17, some of the plurality of cameras 14 are photographed from the front of a passerby moving in the first direction through the passage T. Some of the plurality of cameras 14 may be arranged so as to capture a passerby moving the passage T in the second direction opposite to the first direction from the front.

Further, for example, if the monitored area is a floor U that can pass in any direction, as shown in FIG. 18, a plurality of cameras 14 having various imaging directions are arranged so as to be scattered on the floor U. Just do it.

(8) In the above-described embodiment, the abnormal behavior determination unit 114 determines that the total abnormality degree M is "normal" if it is equal to or less than a predetermined threshold value, and "abnormal" if it exceeds the threshold value. The threshold value of the total abnormality degree M used by the abnormal behavior determination unit 114 to determine whether the behavior of the subject is normal or abnormal may be changed according to the movement route of the subject. For example, if the blind spot area included in the movement route of the target person includes facilities such as toilets and shops, the behavior of the target person is not as long as the target person stays in those blind spot areas for a very long time. Should not be considered anomalous. Therefore, in determining the behavior of the subject including those blind spot areas in the movement path, a threshold value larger than the threshold value used for determining the behavior of the subject who does not include those blind spot areas in the movement path may be used.

Further, when the degree of abnormality L (X, Y, t) relating to any blind spot region on the movement path described in the above-mentioned modification (5) exceeds a predetermined threshold value, the abnormal behavior determination unit 114 is targeted. In the mode of determining the behavior of a person as abnormal, the determination using different threshold values may be performed depending on the facility included in the blind spot area.

1 ... Abnormality determination system, 11 ... Abnormality determination device, 12 ... Display, 13 ... Operation device, 14 ... Camera, 15 ... Speaker, 111 ... Image acquisition unit, 112 ... Person detection unit, 113 ... Detection log update unit, 114 ... Abnormal behavior determination unit, 115 ... Notification unit, 116 ... Data deletion unit, 117 ... Storage unit, 1101 ... Processor, 1102 ... Memory, 1103 ... Interface.

Claims (5)

When the area of the monitoring target area that is not the imaging area of any of the plurality of cameras discretely arranged in the monitoring target area is set as the blind spot area , the subject is the first based on the images captured by the plurality of cameras. The time spent in the blind spot area when moving on the movement path from the image pickup area to the blind spot area and exiting to the second image pickup area is specified, and the probability of occurrence of the time is determined by a general passerby. An abnormality determination device that calculates based on the probability distribution of the time spent in the blind spot region when moving on a route, and determines whether or not the behavior of the subject is abnormal based on the probability of occurrence . The monitored area includes a plurality of the blind spot areas.
Claim 1 for calculating the occurrence probability for each of the plurality of blind spot regions and determining whether or not the behavior of the subject is abnormal based on the occurrence probability calculated for each of the plurality of blind spot regions. The abnormality determination device described in 1. The abnormality determination device according to claim 1 or 2 , wherein when it is determined that the behavior of the target person is abnormal, the user is notified. The person and the person appearing in the image captured by any of the plurality of cameras by storing the characteristic information of the subject extracted from the image captured by any of the plurality of cameras and using the characteristic information. By identifying the subject, the period during which the subject stays in the imaging region of any of the plurality of cameras is specified , and the time during which the subject stays in the blind spot region is specified based on the period . The abnormality determination device according to any one of claims 1 to 3 . The fourth aspect of the present invention, wherein the memory of the characteristic information of the subject is not maintained after the time during which the characteristic information of the subject is not extracted from the image captured by any of the plurality of cameras continues to be equal to or longer than the threshold value. Abnormality determination device.

Images