WO2023286152A1

WO2023286152A1 - Detection device, detection method, and non-transitory computer-readable medium

Info

Publication number: WO2023286152A1
Application number: PCT/JP2021/026293
Authority: WO
Inventors: 健吾大羽賀
Original assignee: 日本電気株式会社
Priority date: 2021-07-13
Filing date: 2021-07-13
Publication date: 2023-01-19

Abstract

A detection device (2000) acquires predicted passing time information (60). The predicted passing time information (60) indicates a predicted passing time at which a person who will attempt to board a vehicle (10) immediately before the departure of the vehicle (10) is predicted to pass through the imaging range of a camera (40). The camera (40) is capable of capturing images of persons heading to the departure location (20). The detection device (2000) acquires video data (50) generated by the camera (40). The detection device (2000) uses the predicted passing time information (60) to detect, from the video data (50), a person to be monitored who is predicted to attempt to board the vehicle (10) immediately before the departure of the vehicle (10).

Description

DETECTION APPARATUS, DETECTION METHOD, AND NON-TRANSITARY COMPUTER-READABLE MEDIUM

This disclosure relates to technology that supports the provision of safe transportation.

A means of transportation using vehicles such as trains and buses is provided. Such transportation is preferably provided in a safe manner. Therefore, information processing technology has been developed to support the provision of safe means of transportation.

For example, Patent Document 1 discloses a technology for opening and closing the door at a safe timing by detecting a person near the door of a train with a sensor and predicting the person's behavior based on the detection result of the sensor. ing. In addition, for example, Patent Literature 2 discloses a technique of detecting a person near the door of a train with a sensor and detecting rushing boarding based on the moving speed and moving direction of the person.

JP 2020-064570 A JP 2013-052738 A

Both the technologies of Patent Documents 1 and 2 can only monitor the behavior of people near the train door. The present invention has been made in view of the above problems, and one of its purposes is to provide a new technique capable of improving the safety of transportation.

The detection device of the present disclosure predicts that a person who is about to board the target vehicle just before the departure of the target vehicle passes through an imaging range of a camera capable of imaging a person heading to the departure place of the target vehicle. a first acquisition unit that acquires predicted passage time information indicating a passage time; a second acquisition unit that obtains video data generated by the camera; and a detection unit that detects a person to be monitored who is predicted to try to board the target vehicle just before departure of the vehicle.

The detection method of the present disclosure is executed by a computer. The detection method is a predicted passage time at which a person who is about to board the vehicle of interest is predicted to pass through an imaging range of a camera capable of imaging a person heading for the departure point of the vehicle of interest just before departure of the vehicle of interest. a first acquisition unit for acquiring predicted passage time information indicating the target vehicle from the video data using the predicted passage time information; and a detection unit that detects a person to be monitored who is predicted to try to board the vehicle of interest just before departure of the vehicle.

The non-transitory computer-readable medium of the present disclosure stores a program that causes a computer to execute the detection method of the present disclosure.

According to the present disclosure, new technology is provided that can improve the safety of transportation.

4 is a diagram illustrating an overview of the operation of the detection device of Embodiment 1; FIG. 2 is a block diagram illustrating the functional configuration of the detection device of Embodiment 1; FIG. It is a block diagram which illustrates the hardware constitutions of the computer which implement|achieves a detection apparatus. 4 is a flow chart illustrating the flow of processing executed by the detection device of Embodiment 1. FIG. It is a figure which illustrates predicted passage time information. 3 is a block diagram illustrating the functional configuration of a detection device having an output section; FIG. FIG. 4 is a diagram illustrating output information transmitted to a mobile terminal possessed by a person requiring monitoring;

Below, embodiments of the present disclosure will be described in detail with reference to the drawings. In each drawing, the same reference numerals are given to the same or corresponding elements, and redundant description will be omitted as necessary for clarity of description. Further, unless otherwise described, predetermined values such as predetermined values and reference values are stored in advance in an arbitrary storage unit in a manner accessible from a device that uses the values. Further, unless otherwise specified, the storage unit is composed of one or more arbitrary number of storage devices.

[Embodiment 1]
<Overview>
FIG. 1 is a diagram illustrating an overview of the operation of the detection device 2000 of Embodiment 1. FIG. Here, FIG. 1 is a diagram for facilitating understanding of the outline of the detection device 2000, and the operation of the detection device 2000 is not limited to that shown in FIG.

The detection device 2000 detects people who are predicted to try to board the vehicle 10 just before departure (so-called rush boarding) from among the people heading to the departure place 20 of the vehicle 10 . Hereinafter, a person predicted to try to board the vehicle 10 just before departure is referred to as a person to be monitored.

A vehicle 10 is a vehicle used to provide a means of transportation. In other words, vehicle 10 is a vehicle that carries customers. In addition, the vehicle 10 has a predetermined departure time. For example, such vehicles 10 include trains, buses, and the like. Here, "just before departure of the vehicle 10" can be defined as, for example, "after a predetermined time before the departure time of the vehicle 10". For example, the predetermined time can be set as 10 seconds.

The departure location 20 is the location from which the vehicle 10 departs. If the vehicle 10 is a train, the departure point 20 is the platform on which the train stops. If the vehicle 10 is a bus, the departure point 20 is the stop at which the bus stops.

The detection device 2000 uses the video data 50 and the predicted passage time information 60 to detect the person to be monitored. The video data 50 is video data generated by the camera 40 capable of capturing images of people heading to the starting point 20 . A camera 40 is provided at a place where a person heading to the departure place 20 passes. The predicted passage time information 60 indicates the time at which the person to be monitored is predicted to pass through the imaging range of the camera 40 (hereinafter referred to as predicted passage time). The predicted passage time can also be said to be "the time at which a person who is within the imaging range of the camera 40 at the predicted passage time will get on the vehicle 10 just before departure if they hurry to the departure place 20".

The detection device 2000 uses the predicted passage time information 60 to detect persons to be monitored from among persons included in the video data 50 (persons captured by the camera 40). The person to be monitored is, among the persons detected from the video data 50, a person whose image is captured at a time close to the predicted passage time and who is assumed to be moving in a hurry. A specific method for detecting such a person will be described later.

<Example of action and effect>
According to the detection device 2000 of the present embodiment, a person who is predicted to board the vehicle 10 just before departure (a person who is predicted to board the vehicle 10 at the last minute) is detected from among the people heading to the departure point 20 of the vehicle 10. . Therefore, it is possible to grasp the possibility of rushing to board a train at an earlier timing than in the case of detecting a rushing boarding person from among the people near the train door. Therefore, transportation can be provided more safely. Although the details will be described later, for example, it is possible for the station attendant to make an announcement in advance to call attention, so that the trainees refrain from rushing to get on the train.

The detection device 2000 of this embodiment will be described in more detail below.

<Example of functional configuration>
FIG. 2 is a block diagram illustrating the functional configuration of the detection device 2000 of Embodiment 1. As shown in FIG. A first acquisition unit 2020 acquires the predicted passage time information 60 . A second acquisition unit 2040 acquires the video data 50 . The detection unit 2060 detects a person to be monitored from persons included in the video data 50 using the predicted passage time information 60 .

<Example of hardware configuration>
Each functional configuration unit of the detection device 2000 may be implemented by hardware that implements each functional configuration unit (eg, hardwired electronic circuit, etc.), or a combination of hardware and software (eg, electronic A combination of a circuit and a program that controls it, etc.). A case in which each functional component of the detection device 2000 is implemented by a combination of hardware and software will be further described below.

FIG. 3 is a block diagram illustrating the hardware configuration of the computer 500 that implements the detection device 2000. As shown in FIG. Computer 500 is any computer. For example, the computer 500 is a stationary computer such as a PC (Personal Computer) or a server machine. In addition, for example, the computer 500 is a portable computer such as a smart phone or a tablet terminal. Computer 500 may be a dedicated computer designed to implement detection apparatus 2000 or a general-purpose computer.

For example, by installing a predetermined application on the computer 500, the functions of the detection device 2000 are implemented on the computer 500. The application is composed of a program for realizing each functional component of the detection device 2000 . It should be noted that the acquisition method of the above program is arbitrary. For example, the program can be acquired from a storage medium (DVD disc, USB memory, etc.) in which the program is stored. In addition, for example, the program can be obtained by downloading the program from a server device that manages the storage unit storing the program.

Computer 500 has bus 502 , processor 504 , memory 506 , storage device 508 , input/output interface 510 and network interface 512 . The bus 502 is a data transmission path through which the processor 504, memory 506, storage device 508, input/output interface 510, and network interface 512 exchange data with each other. However, the method of connecting the processors 504 and the like to each other is not limited to bus connection.

The processor 504 is various processors such as a CPU (Central Processing Unit), GPU (Graphics Processing Unit), or FPGA (Field-Programmable Gate Array). The memory 506 is a main memory implemented using a RAM (Random Access Memory) or the like. The storage device 508 is an auxiliary storage device implemented using a hard disk, SSD (Solid State Drive), memory card, ROM (Read Only Memory), or the like.

The input/output interface 510 is an interface for connecting the computer 500 and input/output devices. For example, the input/output interface 510 is connected to an input device such as a keyboard and an output device such as a display device.

A network interface 512 is an interface for connecting the computer 500 to a network. This network may be a LAN (Local Area Network) or a WAN (Wide Area Network).

The storage device 508 stores a program that implements each functional component of the detection device 2000 (a program that implements the application described above). The processor 504 implements each functional component of the detection device 2000 by reading this program into the memory 506 and executing it.

The detection device 2000 may be realized by one computer 500 or may be realized by a plurality of computers 500. In the latter case, the configuration of each computer 500 need not be the same, and can be different.

<About camera 40>
Camera 40 is any camera that takes images and produces video data representing the results. Here, some or all of the functions of the detection device 2000 may be realized by the camera 40. FIG. As such a camera, a camera called an intelligent camera, a network camera, an IP (Internet Protocol) camera, or the like can be used.

<Process flow>
FIG. 4 is a flowchart illustrating the flow of processing executed by the detection device 2000 of the first embodiment. The first acquisition unit 2020 acquires the predicted passage time information 60 (S102). The second acquisition unit 2040 acquires the video data 50 (S104). The detection unit 2060 detects the person to be monitored using the video data 50 and the predicted passage time information 60 (S106).

Here, a plurality of vehicles 10 can depart from the departure place 20 at different times. For example, in the case of trains, trains may depart from the same platform every few minutes to several tens of minutes. Therefore, the detection device 2000 detects a person to be monitored for each vehicle 10 . In addition, the number of cameras 40 is not limited to one, and cameras 40 can be installed at each of a plurality of locations. Therefore, the detection device 2000 detects a person to be monitored for each camera 40 . That is, the detection device 2000 detects a person to be monitored for each vehicle 10 and camera 40 pair.

More specifically, the detection device 2000 uses the predicted passage time information 60 to grasp the predicted passage time for each vehicle 10 and camera 40 pair. Furthermore, the detection device 2000 performs the following processing for each pair of the vehicle 10 and camera 40 . First, the detection device 2000 acquires the video data 50 at the timing based on the predicted passage time grasped for the target pair. This timing is, for example, when a predetermined time (5 seconds, 10 seconds, etc.) has passed from the predicted passage time. Then, the detection device 2000 uses the predicted passage time for the target pair and the video data 50 acquired for the pair to detect the person to be monitored for the pair.

<Obtaining Predicted Passing Time Information 60: S102>
The first acquisition unit 2020 acquires the predicted passage time information 60 (S102). The predicted passage time information 60 indicates the predicted passage time. Here, there may be multiple vehicles 10 departing from the departure location 20 . Therefore, the predicted passage time information 60 indicates the predicted passage time for each vehicle 10 . Also, a plurality of cameras 40 may exist. Therefore, the predicted passage time information 60 indicates the predicted passage time for each camera 40 . That is, the predicted passage time information 60 indicates the predicted passage time in association with the pair of the vehicle 10 and camera 40 .

FIG. 5 is a diagram illustrating predicted passage time information 60. FIG. The predicted passage time information 60 indicates a predicted passage time 66 in association with a pair of vehicle identification information 62 and camera identification information 64 . Vehicle identification information 62 indicates the identification information of the vehicle 10 . The camera identification information 64 indicates identification information of the camera 40 . Predicted passage time 66 indicates the predicted passage time for the corresponding vehicle 10 and camera 40 pair. For example, the first row record in FIG. 5 indicates that the predicted passage time is Tp1 for the vehicle v001 and camera c001 pair. This indicates that if a person who passes through the imaging range of camera c001 at time Tp1 rushes to departure point 20, there is a high probability that he will board vehicle v001 just before departure.

There are various methods for the first acquisition unit 2020 to acquire the predicted passage time information 60. For example, the predicted passage time information 60 is pre-stored in any storage unit in a form that can be obtained from the detection device 2000 . In this case, the first acquisition unit 2020 acquires the predicted passage time information 60 stored in this storage unit. Alternatively, for example, the first acquisition unit 2020 may acquire the predicted passage time information 60 by receiving the predicted passage time information 60 transmitted from another device (for example, a generation device, which will be described later).

<Generation of predicted passage time information 60>
Here, a method for generating the predicted passage time information 60 will be illustrated. The predicted passage time information 60 can be generated based on the departure time of the vehicle 10 and the required time required for movement from the imaging range of the camera 40 to the departure place 20 . Specifically, the predicted passage time indicated by the predicted passage time information 60 can be calculated by subtracting the time required for movement from the imaging range of the camera 40 to the departure place 20 from the departure time of the vehicle 10 . For example, suppose the departure time of vehicle 10 is 10:10 and the required travel time is 5 minutes. In this case, the predicted passage time can be predicted to be 10:05.

Therefore, a device having a function of generating predicted passage time information 60 (hereinafter referred to as generation device) specifies the departure time of vehicle 10 and the required time required for movement from the imaging range of camera 40 to departure place 20 . The departure times of the vehicles 10 can be specified by, for example, obtaining operation schedule information (diagram information) indicating the departure times of the vehicles 10 departing from the departure place 20 .

The time required to move from the imaging range of the camera 40 to the departure location 20 can be calculated based on the positional relationship between the imaging range of the camera 40 and the departure location 20. For example, the generation device identifies the length of the movement route by identifying the movement route from the imaging range of the camera 40 to the departure place 20 . Then, the generating device calculates the time required for moving the route of the specified length as the required time required for moving from the imaging range of the camera 40 to the departure place 20 .

Here, for example, any route search algorithm can be used as a method of specifying the travel route between two specific points and its length. For example, the generation device acquires map data of the premises of the station where the departure place 20 is provided and the surroundings of the station. It is assumed that this map data indicates the position of the departure place 20 and the position of the camera 40 . For example, the generating device specifies the movement route from the camera 40 to the starting place 20 and its length by executing the shortest route search from the camera 40 to the starting place 20 for the map data.

Note that the moving route from the camera 40 to the departure place 20 is not limited to the shortest route. For example, based on past usage records of stations, etc., a travel route frequently used by the user of the vehicle 10 (particularly, a travel route frequently used by a person in a hurry) may be specified. may be used to generate the predicted passage time information 60.

There are various methods for calculating the required time from the length of the travel route. For example, an assumed moving speed of the person to be monitored is determined in advance. In this case, the generation device calculates the required time by calculating “length of travel route/assumed travel speed”. Here, there is a high probability that the person to be monitored is moving quickly (for example, running). Therefore, it is preferable to set the speed of a person who is moving in a hurry (for example, the average speed of a person who is running) as the assumed moving speed.

In addition, for example, a function that defines the relationship between the length of the travel route and the required time may be prepared in advance. In this case, the generation device can obtain the required time by inputting the length of the identified travel route to the function. As described above, it is highly probable that the person to be monitored is moving in a hurry, so it is preferable that the required time obtained from this function also represents the required time when the person moves in a hurry.

In addition, there may be facilities such as escalators, elevators, or stairs on the movement route. The generation device may calculate the required time considering the influence of these facilities. For example, the time required for movement within these facilities is fixed in advance. In this case, for example, the generation device adds the time required for movement in each piece of equipment present in the movement route to the required time calculated by the method described above.

There are various timings for generating the predicted passage time information 60. For example, first, before starting the operation of the detecting device 2000, the predicted passage time information 60 is generated using the timetable information at that time. After that, when the timetable information is updated (that is, when the timetable is revised), the updated timetable information is used before the operation of the vehicle 10 starts according to the revised timetable, and the predicted passage time information 60 is generated.

In addition, due to special events or accidents such as accidents, the vehicle 10 may be operated on a timetable that differs from the normal timetable. In such a case, the generation device may generate temporary predicted passage time information 60 using information indicating a temporary timetable. The detection device 2000 uses this predicted passage time information 60 only until the timetable returns to normal.

The generation device may be provided integrally with the detection device 2000, or may be provided separately. The former means that the detecting device 2000 has the function of operating as a generating device. In this case, the detection device 2000 has a generation unit (not shown) that functions as a generation device.

<Acquisition of video data 50: S104>
The second acquisition unit 2040 acquires the video data 50 (S104). There are various methods for the second acquisition unit 2040 to acquire the video data 50 . For example, the camera 40 stores the video data 50 in any storage unit in a manner that can be obtained from the detection device 2000 . In this case, the second acquisition unit 2040 acquires the video data 50 from the storage unit. Alternatively, for example, the video data 50 may be transmitted from the camera 40 to the detection device 2000 .

Note that the second acquisition unit 2040 may acquire part of the video data generated by the camera 40 as the video data 50. For example, the second acquisition unit 2040 acquires, as the video data 50, only the portion of the video data generated by the camera 40 that is a predetermined length of time before and after the predicted passage time.

<Detection of person requiring surveillance: S106>
The detection unit 2060 detects a person to be monitored from the video data 50 (S106). For example, the detection unit 2060 detects persons from the video data 50, and determines whether or not each detected person is a person to be monitored based on the predicted passage time. As described above, the detection unit 2060 detects a person to be monitored for each vehicle 10 and camera 40 pair. At this time, among the predicted passage times indicated in the predicted passage time information 60, the predicted passage time corresponding to the pair of the vehicle 10 and the camera 40 to be processed is used. Also, the video data 50 used is generated by the camera 40 to be processed.

For example, the detection unit 2060 calculates a score focusing on each of the following three elements for each person to be determined.
(Element 1) The image is captured at a time close to the predicted passage time.
(Element 2) Heading to departure location 20 .
(Condition 3) I am in a hurry.
Hereinafter, scores focused on element 1, element 2, and element 3 are referred to as first score, second score, and third score, respectively. A specific calculation method for each score will be described later.

For example, the detection unit 2060 uses the first to third scores to calculate a total score, and uses the total score to determine whether or not the person to be determined is a person requiring surveillance. For example, the detection unit 2060 determines that the person to be determined is a person requiring monitoring when the total score is equal to or greater than a threshold. On the other hand, when the total score is less than the threshold, the detection unit 2060 determines that the person to be determined is not a person requiring monitoring. Note that the total score is calculated as, for example, the sum, weighted sum, or product of the first to third scores. The weight given to each score is determined in advance.

In addition, for example, the detection unit 2060 determines whether each of the first score to the third score is equal to or greater than the threshold, and if all of the scores are equal to or greater than the threshold, the person to be determined is the person to be monitored. Determine that there is. In other words, in this case, the person to be determined satisfies all of the conditions that "the image is captured at a time close to the predicted passing time", the condition that "the person is heading to the starting point 20", and the condition that the person is "in a hurry". In this case, it is determined that the person is a person requiring surveillance. On the other hand, if any of the scores is less than the threshold, the detection unit 2060 determines that the person to be determined is not a person requiring monitoring. Here, the threshold values for each score may be the same or different.

It should be noted that not all of the first to third scores need to be used. For example, when the direction of movement of each person is clear, such as when the passage captured by the camera 40 is one-way, the second score may not be used.

Below, the calculation method for each of the first to third scores will be exemplified.

<<About the first score>>
The detection unit 2060 calculates, as the first score, a value representing the degree to which the period in which the person to be determined is detected from the video data 50 is close to the predicted passage time. Specifically, the detection unit 2060 calculates a higher first score as the period during which the person to be determined is detected from the video data 50 is closer to the predicted passage time. For example, if the predicted passing time is included in the period in which the person to be determined is detected from the video data 50, the detection unit 2060 calculates the maximum first score for the person to be determined. On the other hand, if the predicted passage time is not included in the period in which the person to be determined is detected from the video data 50, the detecting unit 2060 assigns a smaller first score as the difference between the period and the predicted passage time increases. calculate.

For example, suppose the predicted passage time is Tp1. It is also assumed that the period during which the person to be determined is detected from the video data 50 is from time Td1 to time Td2. In this case, if Td1<=Tp1<=Td2, the detection unit 2060 calculates the maximum first score for the person to be determined. For example, when the value range of the first score is 0 or more and 1 or less, the first score is 1. On the other hand, if Tp1<Td1, the detection unit 2060 calculates a smaller first score as the value of Td1-Tp1 increases. Similarly, if Td2<Tp1, the detection unit 2060 calculates a smaller first score as Tp1-Td2 is larger.

It should be noted that when performing the process of detecting a person from the video data 50, only video frames in a predetermined period before and after the predicted passage time may be processed. By doing so, a person who is clearly not a person requiring monitoring can be excluded from score calculation targets.

<<About the second score>>
The detection unit 2060 calculates a value representing the degree of heading toward the starting point 20 as the second score. The degree of heading toward the starting point 20 can be represented by the degree of matching between the movement direction of the person to be determined and the direction toward the starting point 20 . The degree of matching is represented, for example, by the size of an angle formed by a vector representing the moving direction of the person to be determined and a vector representing the direction toward the departure place 20 . The second score increases as the size of the formed angle is closer to 0°, and decreases as the angle is further away from 0°. The correspondence relationship between the formed angle and the second score is determined in advance as a function, for example.

<<About the third score>>
The detection unit 2060 calculates, as the third score, a value representing the degree to which the person to be determined is in a hurry. For example, the third score is predetermined according to the type of movement action of the person to be determined. For example, types such as "walking slowly", "walking fast", "running slowly", and "running fast" can be defined as types of moving motions. The third score is set to a larger value for a type of movement action that allows faster movement. In the example above, the 3rd score for "walking slowly" is the lowest, the 3rd score for "walking fast" is the second lowest, and the 3rd score for "running slowly" is the third. The third score of "running fast" is the maximum. Note that the types of movement motions are not limited to the four types described above. For example, two types of "walking" and "running" may be used.

The detection unit 2060 identifies the type of moving motion by analyzing the motion of the person to be determined detected from the video data 50 . Then, the detection unit 2060 uses the third score corresponding to the specified type of moving action as the third score of the person to be determined.

Here, there are various methods for identifying the type of movement motion. For example, the detection unit 2060 detects feature points such as tips of hands and feet and joints of the person in each video frame in which the person to be determined is detected. Then, the detection unit 2060 identifies the type of movement action of the person to be determined by pattern recognition using the movement trajectory of these feature points as a feature amount.

The third score may be determined based on the speed of movement of the person to be determined. Specifically, the third score in this case is determined as a value that increases as the speed of movement of the person to be determined increases. A correspondence relationship between the speed of movement and the third score is determined in advance as a function, for example. In this case, the detection unit 2060 specifies the speed of movement of the person to be determined, and inputs the speed into the above function to calculate the third score. An existing technique can be used as a technique for calculating the movement speed of a person detected from video data.

In addition, for example, the third score may be determined based on the characteristics of the facial expression of the person to be determined. For example, the contours of the eyes and mouth, or the position, size, or movement of the pupils can be used as facial expression features. The detection unit 2060 calculates these feature amounts from one or more video frames in which the person to be determined is detected, and performs pattern recognition using these feature amounts to detect whether the person to be determined is in a hurry. Calculate the degree.

<How to use detection results>
When a person to be monitored is detected, the detecting device 2000 preferably outputs information about the person to be monitored. This information is hereinafter referred to as output information. A functional configuration unit that outputs output information is called an output unit. FIG. 6 is a block diagram illustrating the functional configuration of the detection device 2000 having the output section 2080. As shown in FIG.

Various information can be adopted as the output information. For example, the output information is information indicating that there is a person to be monitored. This output information is transmitted to terminals used by, for example, station staff and security guards. By providing such output information, station staff, security guards, etc. can make advance warning announcements in preparation for last-minute boarding and prepare a security system. The output destination terminal may be a portable terminal such as a smartphone, or a stationary terminal such as a PC.

In addition to the presence of the person under surveillance, the output information may include various information related to the person under surveillance. For example, the information related to the person to be monitored includes the position of the person to be monitored, the score calculated for the person to be monitored, information about the vehicle 10 that the person to be monitored is going to use (departure place, departure time, remaining time until departure time, etc.) , or destination, etc.), or the distance from the monitored person to the starting point 20, and the like. Here, it is preferable that the position of the person to be monitored is indicated using a map (such as a station map). For example, the output information includes a map superimposed with a mark representing the location of the person to be monitored.

Here, the output unit 2080 may set the degree of emphasis of the output information based on the information regarding the person to be monitored. For example, ``the greater the score calculated for the person to be monitored, the stronger the degree of emphasis'', ``the closer the distance from the person to be monitored to the departure place 20, the stronger the degree of emphasis'', or ``to the departure time of the vehicle 10 The shorter the remaining time is, the stronger the degree of emphasis is." As a method of increasing the degree of emphasis, for example, a method of using a more conspicuous color, a method of increasing the size, and the like are conceivable. Also, when the output information is output by voice, a method of increasing the volume may be considered as a method of increasing the degree of emphasis.

　The output information is not limited to information about the person to be monitored. For example, using a speaker installed in a station premises, etc., output information that expresses a voice message to the effect that a person trying to rush to get on the train has been detected, or a voice message that urges people to refrain from rushing to get on the train, etc. may be output. Here, this voice message is preferably heard by the person under surveillance. Therefore, for example, the output unit 2080 causes a speaker provided near the camera 40 that generated the video data 50 in which the person to be monitored is detected to output a voice message. It is also preferable to use each speaker provided on the route from the camera 40 to the starting point 20 .

In addition, for example, the output unit 2080 may transmit output information to a mobile terminal possessed by a person requiring monitoring. In this case, for example, the output information preferably includes guidance regarding alternative means of transportation. In addition, information that activates an application that can search for means of transportation, or information that guides users to a website that can search for means of transportation (information indicating a link to the website, etc.) may be sent as output information. According to these methods, it becomes easier for the person to be monitored to grasp alternative means of transportation, so that the probability that the person to be monitored will stop rushing to board the vehicle can be increased.

FIG. 7 is a diagram exemplifying output information transmitted to a mobile terminal possessed by a person requiring monitoring. This output information includes a map showing the current position of the person to be monitored and the position of the taxi stand, and a message prompting the use of a taxi.

Here, when the output information is to be transmitted to the mobile terminal possessed by the person to be monitored, it is assumed that the information on the person is registered in advance. Specifically, a pair of a person's face information (facial feature amount) and a destination address is stored in advance in the storage unit in a manner that can be obtained from the detecting device 2000 . The output unit 2080 searches for a registered person who matches the person to be monitored by matching the facial feature amount of the person to be monitored detected from the video data 50 with the facial feature amount of each registered person. do. When a registered person matching the person to be monitored is detected, the output unit 2080 transmits output information to the destination address associated with the registered person.

Note that, when a person to be monitored is detected, the detection device 2000 uses the video data 50 obtained from each camera 40 provided at the destination of the person (on the route toward the starting point 20) to detect the person to be monitored. may continue to be detected as a person to be monitored. In other words, the detection apparatus 2000 may determine whether each person to be monitored detected using each camera 40 is a person who has already been detected as a person to be monitored. When a person who has already been detected as a person requiring surveillance is detected again as a person requiring surveillance, output information to be output to the person newly detected as a person requiring surveillance is output information to be output to that person. A different one may be used. For example, the greater the number of times a person is detected as a person to be monitored, the more emphasized the content of the output information is output. Whether or not the persons to be monitored detected from different video data 50 match each other can be determined, for example, by comparing the facial features of the persons to be monitored detected from each video data 50 .

Although the present invention has been described with reference to the embodiments, the present invention is not limited to the above embodiments. Various changes that can be understood by those skilled in the art can be made to the configuration and details of the present invention within the scope of the present invention.

It should be noted that in the above examples, the program includes instructions (or software code) that, when read into a computer, cause the computer to perform one or more functions described in the embodiments. The program may be stored in a non-transitory computer-readable medium or tangible storage medium. By way of example, and not limitation, computer readable media or tangible storage media may include random-access memory (RAM), read-only memory (ROM), flash memory, solid-state drives (SSD) or other memory technology, CDs - ROM, digital versatile disc (DVD), Blu-ray disc or other optical disc storage, magnetic cassette, magnetic tape, magnetic disc storage or other magnetic storage device. The program may be transmitted on a transitory computer-readable medium or communication medium. By way of example, and not limitation, transitory computer readable media or communication media include electrical, optical, acoustic, or other forms of propagated signals.

Some or all of the above-described embodiments can also be described in the following supplementary remarks, but are not limited to the following.
(Appendix 1)
Predicted passage time indicating a predicted passage time at which a person who is about to board the target vehicle is predicted to pass through an imaging range of a camera capable of capturing an image of a person heading to the departure place of the target vehicle just before the departure of the target vehicle. a first acquisition unit that acquires information;
a second acquisition unit for acquiring video data generated by the camera;
a detection unit that detects a person to be monitored who is predicted to board the target vehicle just before departure of the target vehicle from the video data using the predicted passage time information.
(Appendix 2)
For each person detected from the video data, the detection unit obtains a first score representing the degree to which the period during which the person was captured by the camera is close to the predicted passage time, and Any one or more of a second score representing the degree of urgency and a third score representing the degree of urgency of the person is calculated, and based on the calculated score, it is determined whether or not the person is the person to be monitored. 2. The detection device of claim 1, wherein the detection device determines.
(Appendix 3)
The detection unit calculates the third score for each person detected from the video data based on the type of movement of the person, the speed of movement of the person, or the facial expression of the person. , Supplementary Note 2.
(Appendix 4)
Based on the positional relationship between the camera and the departure place, a predicted required time, which is the time required for movement from the imaging range of the camera to the departure place, is calculated, and the departure time and the predicted required time of the target vehicle. 4. The detection device according to any one of appendices 1 to 3, comprising a generation unit that generates the predicted passage time information based on and.
(Appendix 5)
Output for outputting output information indicating any one or more of information related to the person to be monitored detected by the detection unit, information indicating alternative means of transportation, and information facilitating search for alternative means of transportation 5. The detection device according to any one of appendices 1 to 4, comprising a portion.
(Appendix 6)
Information that associates a person's facial features with a destination address of the person is stored in a storage unit,
The output unit identifies, from the facial feature amounts stored in the storage unit, those that match the facial feature amount of the person to be monitored obtained from the video data, and determines the identified feature amount. 6. The detection device according to appendix 5, wherein the output information is output to the associated destination address.
(Appendix 7)
A computer-implemented detection method comprising:
Predicted passage time indicating a predicted passage time at which a person who is about to board the target vehicle is predicted to pass through an imaging range of a camera capable of capturing an image of a person heading to the departure place of the target vehicle just before the departure of the target vehicle. a first obtaining step of obtaining information;
a second acquisition step of acquiring video data generated by the camera;
and a detecting step of detecting from the video data a person to be monitored who is predicted to board the target vehicle just before departure of the target vehicle, using the predicted passage time information.
(Appendix 8)
In the detecting step, for each person detected from the video data, a first score representing the degree to which the period during which the person was captured by the camera is close to the predicted passage time, the person is heading to the departure location; Any one or more of a second score representing the degree of urgency and a third score representing the degree of urgency of the person is calculated, and based on the calculated score, it is determined whether or not the person is the person to be monitored. The detection method according to appendix 7, wherein:
(Appendix 9)
In the detection step, for each person detected from the video data, the third score is calculated based on the type of movement of the person, the speed of movement of the person, or the characteristics of facial expression of the person. , appendix 8.
(Appendix 10)
Based on the positional relationship between the camera and the departure place, a predicted required time, which is the time required for movement from the imaging range of the camera to the departure place, is calculated, and the departure time and the predicted required time of the target vehicle. 10. The detection method according to any one of appendices 7 to 9, comprising a generating step of generating the predicted transit time information based on:
(Appendix 11)
An output that outputs output information indicating any one or more of information related to the person to be monitored detected by the detecting step, information indicating alternative means of transportation, and information facilitating search for alternative means of transportation. 11. A detection method according to any one of clauses 7 to 10, comprising steps.
(Appendix 12)
Information that associates a person's facial features with a destination address of the person is stored in a storage unit,
In the output step, among the facial feature amounts stored in the storage unit, those that match the facial feature amount of the person to be monitored obtained from the video data are specified, and the specified feature amount is specified. 12. The detection method according to appendix 11, wherein the output information is output to the associated destination address.
(Appendix 13)
to the computer,
Predicted passage time indicating a predicted passage time at which a person who is about to board the target vehicle is predicted to pass through an imaging range of a camera capable of capturing an image of a person heading to the departure place of the target vehicle just before the departure of the target vehicle. a first obtaining step of obtaining information;
a second acquisition step of acquiring video data generated by the camera;
a detection step of detecting a person to be monitored who is predicted to board the target vehicle just before departure of the target vehicle from the video data using the predicted passage time information; and non-transitory computer-readable medium.
(Appendix 14)
In the detecting step, for each person detected from the video data, a first score representing the degree to which the period during which the person was captured by the camera is close to the predicted passage time, the person is heading to the departure location; Any one or more of a second score representing the degree of urgency and a third score representing the degree of urgency of the person is calculated, and based on the calculated score, it is determined whether or not the person is the person to be monitored. 14. The computer-readable medium of Clause 13, wherein determining.
(Appendix 15)
In the detection step, for each person detected from the video data, the third score is calculated based on the type of movement of the person, the speed of movement of the person, or the characteristics of facial expression of the person. 15. The computer-readable medium of claim 14.
(Appendix 16)
Based on the positional relationship between the camera and the departure place, a predicted required time, which is the time required for movement from the imaging range of the camera to the departure place, is calculated, and the departure time and the predicted required time of the target vehicle. 16. The computer-readable medium of any one of Clauses 13-15, comprising a generating step of generating the predicted transit time information based on:
(Appendix 17)
The program provides to the computer any one or more of information related to the person to be monitored detected by the detecting step, information indicating alternative means of transportation, and information facilitating search for alternative means of transportation. 17. The computer-readable medium according to any one of clauses 13 to 16, causing an output step of outputting output information indicative of:
(Appendix 18)
Information that associates a person's facial features with a destination address of the person is stored in a storage unit,
In the output step, among the facial feature amounts stored in the storage unit, those that match the facial feature amount of the person to be monitored obtained from the video data are specified, and the specified feature amount is specified. 18. The computer-readable medium of clause 17, outputting the output information to an associated destination address.

10 vehicle 20 starting point 40 camera 50 video data 60 predicted passage time information 62 vehicle identification information 64 camera identification information 66 predicted passage time 500 computer 502 bus 504 processor 506 memory 508 storage device 510 input/output interface 512 network interface 2000 detector 2020 1 acquisition unit 2040 2nd acquisition unit 2060 detection unit 2080 output unit

Claims

Predicted passage time indicating a predicted passage time at which a person who is about to board the target vehicle is predicted to pass through an imaging range of a camera capable of capturing an image of a person heading to the departure place of the target vehicle just before the departure of the target vehicle. a first acquisition unit that acquires information;
a second acquisition unit for acquiring video data generated by the camera;
a detection unit that detects a person to be monitored who is predicted to board the target vehicle just before departure of the target vehicle from the video data using the predicted passage time information.
For each person detected from the video data, the detection unit obtains a first score representing the degree to which the period during which the person was captured by the camera is close to the predicted passage time, and Any one or more of a second score representing the degree of urgency and a third score representing the degree of urgency of the person is calculated, and based on the calculated score, it is determined whether or not the person is the person to be monitored. 2. The detection device of claim 1, for determining.
The detection unit calculates the third score for each person detected from the video data based on the type of movement of the person, the speed of movement of the person, or the facial expression of the person. 3. A detection device according to claim 2.
Based on the positional relationship between the camera and the departure place, a predicted required time, which is the time required for movement from the imaging range of the camera to the departure place, is calculated, and the departure time and the predicted required time of the target vehicle. 4. The detection device according to any one of claims 1 to 3, further comprising a generation unit that generates the predicted passage time information based on and.
Output for outputting output information indicating any one or more of information related to the person to be monitored detected by the detection unit, information indicating alternative means of transportation, and information facilitating search for alternative means of transportation 5. A detection device according to any one of claims 1 to 4, comprising a portion.
Information that associates a person's facial features with a destination address of the person is stored in a storage unit,
The output unit identifies, from the facial feature amounts stored in the storage unit, those that match the facial feature amount of the person to be monitored obtained from the video data, and determines the identified feature amount. 6. The detecting device according to claim 5, wherein said output information is output to the associated destination address.
A computer-implemented detection method comprising:
Predicted passage time indicating a predicted passage time at which a person who is about to board the target vehicle is predicted to pass through an imaging range of a camera capable of capturing an image of a person heading to the departure place of the target vehicle just before the departure of the target vehicle. a first obtaining step of obtaining information;
a second acquisition step of acquiring video data generated by the camera;
and a detecting step of detecting from the video data a person to be monitored who is predicted to board the target vehicle just before departure of the target vehicle, using the predicted passage time information.
In the detecting step, for each person detected from the video data, a first score representing the degree to which the period during which the person was captured by the camera is close to the predicted passage time, the person is heading to the departure location; Any one or more of a second score representing the degree of urgency and a third score representing the degree of urgency of the person is calculated, and based on the calculated score, it is determined whether or not the person is the person to be monitored. 8. A detection method according to claim 7, comprising determining.
In the detection step, for each person detected from the video data, the third score is calculated based on the type of movement of the person, the speed of movement of the person, or the characteristics of facial expression of the person. 9. The detection method according to claim 8.
Based on the positional relationship between the camera and the departure place, a predicted required time, which is the time required for movement from the imaging range of the camera to the departure place, is calculated, and the departure time and the predicted required time of the target vehicle. 10. A detection method according to any one of claims 7 to 9, comprising a generating step of generating said predicted transit time information based on:
An output that outputs output information indicating any one or more of information related to the person to be monitored detected by the detecting step, information indicating alternative means of transportation, and information facilitating search for alternative means of transportation. 11. A detection method according to any one of claims 7 to 10, comprising steps.
Information that associates a person's facial features with a destination address of the person is stored in a storage unit,
In the output step, among the facial feature amounts stored in the storage unit, those that match the facial feature amount of the person to be monitored obtained from the video data are specified, and the specified feature amount is specified. 12. The detection method according to claim 11, wherein the output information is output to the associated destination address.
to the computer,
Predicted passage time indicating a predicted passage time at which a person who is about to board the target vehicle is predicted to pass through an imaging range of a camera capable of capturing an image of a person heading to the departure place of the target vehicle just before the departure of the target vehicle. a first obtaining step of obtaining information;
a second acquisition step of acquiring video data generated by the camera;
a detection step of detecting a person to be monitored who is predicted to board the target vehicle just before departure of the target vehicle from the video data using the predicted passage time information; and non-transitory computer-readable medium.
In the detecting step, for each person detected from the video data, a first score representing the degree to which the period during which the person was captured by the camera is close to the predicted passage time, the person is heading to the departure location; Any one or more of a second score representing the degree of urgency and a third score representing the degree of urgency of the person is calculated, and based on the calculated score, it is determined whether or not the person is the person to be monitored. 14. The computer readable medium of claim 13, wherein determining.
In the detection step, for each person detected from the video data, the third score is calculated based on the type of movement of the person, the speed of movement of the person, or the characteristics of facial expression of the person. 15. The computer readable medium of claim 14.
Based on the positional relationship between the camera and the departure place, a predicted required time, which is the time required for movement from the imaging range of the camera to the departure place, is calculated, and the departure time and the predicted required time of the target vehicle. 16. The computer-readable medium of any one of claims 13-15, comprising a generating step of generating the predicted transit time information based on:
The program provides to the computer any one or more of information related to the person to be monitored detected by the detecting step, information indicating alternative means of transportation, and information facilitating search for alternative means of transportation. 17. A computer readable medium according to any one of claims 13 to 16, causing an output step to be performed to output output information indicative of a.
Information that associates a person's facial features with a destination address of the person is stored in a storage unit,
In the output step, among the facial feature amounts stored in the storage unit, those that match the facial feature amount of the person to be monitored obtained from the video data are specified, and the specified feature amount is specified. 18. The computer-readable medium of claim 17, outputting the output information to an associated destination address.