JP7391557B2 - Notification device, notification method and notification program - Google Patents

Description

The present invention relates to a notification device, a notification method, and a notification program.

BACKGROUND ART Systems have been used that detect and notify when a person enters a road by analyzing images taken by a surveillance camera (for example, see Patent Documents 1 and 2).

JP 2015-022333 Publication International Publication No. 2018/179202

If a person is notified of an intrusion based on the presence of a person in the footage captured by a surveillance camera, the system will also notify the user if a person who has permission to work on the road is detected in advance. You will be killed. As a result, it is necessary for a supervisor or the like who has received the notification to confirm whether or not the person who has invaded the road is an authorized person through video or visually confirms the intrusion at the site.

One aspect of the disclosed technology is to provide a notification device, a notification method, and a notification program that can suppress unnecessary intrusion notifications.

One aspect of the disclosed technology is exemplified by the following notification device. This notification device is a notification device that notifies the intrusion of a person into a monitored area, and includes an acquisition unit that acquires imaged data obtained by photographing the monitored area from a camera, and a notification device that notifies the user when a person is detected in the imaged data. and a notification unit that notifies of a person's intrusion into a monitored area, and the notification unit suppresses the notification when a predetermined mark stored in a storage unit in advance is detected in the imaged data together with a person.

According to the disclosed technology, unnecessary intrusion notifications can be suppressed.

FIG. 1 is a diagram illustrating an example of an intrusion detection system according to an embodiment. FIG. 2 is a diagram illustrating an example of the hardware configuration of the detection device according to the embodiment. FIG. 3 is a diagram illustrating an example of processing blocks of the detection device according to the embodiment. FIG. 4 is a first diagram illustrating an example of a situation in which the determination unit inhibits the notification unit from transmitting an intrusion notification in the embodiment. FIG. 5 is a second diagram illustrating an example of a situation in which the determination unit inhibits the notification unit from transmitting an intrusion notification in the embodiment. FIG. 6 is a first diagram illustrating an example of a situation in which the determination unit does not suppress transmission of an intrusion notification in the embodiment. FIG. 7 is a second diagram illustrating an example of a situation in which the determination unit does not suppress transmission of an intrusion notification in the embodiment. FIG. 8 is an example of a processing flow of the intrusion detection system according to the embodiment. FIG. 9 is a diagram illustrating an example of a processing block of the detection device according to the first modification. FIG. 10 is a diagram illustrating an example of a situation in which transmission of an intrusion notification is suppressed in the first modification. FIG. 11 is an example of a processing flow of the intrusion detection system according to the first modification. FIG. 12 is a diagram illustrating an example of a situation in which an ordinary person appears in the surveillance camera video together with a maintenance worker who is to be tracked by the person detection unit in the first modification. FIG. 13 is a diagram illustrating an example in which the range of presence of a sign that suppresses transmission of an entry notification in an image is limited in the second modification. FIG. 14 is a diagram showing an example of a determination table in the third modification.

The notification device according to this embodiment has, for example, the following configuration.
A notification device that notifies of a person's intrusion into a monitored area,
an acquisition unit that acquires image data obtained by photographing the monitored area from a camera;
a notification unit that notifies the person of intrusion into the monitored area when a person is detected in the imaged data;
The notification unit suppresses the notification when a predetermined mark stored in the storage unit in advance is detected in the imaged data together with a person.

In this notification device, the monitored area is an area where the general public is restricted (or prohibited) from entering, and includes, for example, a motorway, railroad tracks, and an airstrip. The camera may be an optical camera or an infrared camera. Further, the camera may be a 360-degree semi-spherical camera. The camera is installed at a position where it can photograph the monitored area.

Examples of the predetermined sign include a guide board, a signboard, a road cone, an electronic bulletin board, etc., indicating a work place where permitted work is to be performed in the monitored area, and a tool used for permitted work. In this notification device, when a person is detected along with such a predetermined sign, it is assumed that the person is an authorized worker, and the notification of intrusion is suppressed. By performing such processing, transmission of an intrusion notification regarding a person who intrudes into the monitored area with permission is suppressed. That is, according to this notification device, unnecessary intrusion notifications can be suppressed.

This notification device may have the following features. The predetermined sign includes a vehicle, and the notification unit further suppresses the notification when detecting that the vehicle is stopped. Examples of the vehicle include maintenance vehicles used for maintenance on roads, railroads, runways, and the like. For example, if the imaging data is video data, whether or not the vehicle is stopped can be determined by determining whether the vehicle is moving if there is no change in the coordinates of the vehicle in each frame in several consecutive frames of the video data. It can be determined that there is no such thing. In addition, whether the vehicle is moving or not can be determined, for example, if the captured image data is a still image, if there is no change in the coordinates of the vehicle in each captured image data among multiple captured data captured by the camera at a predetermined interval. It can be determined that the vehicle is not moving. The maintenance vehicle transports maintenance workers to the maintenance work site, and after being transported, it also serves as a sign indicating the maintenance work site by stopping near the maintenance work site. When the notification device has such characteristics, notifications can be suppressed by using the maintenance vehicle as a predetermined sign.

This notification device may have the following features. Regarding the detected person detected together with the predetermined sign in the image data, the notification unit may detect the detected person in the second image data acquired by the acquisition unit after the image data including the predetermined sign and the detected person. The notification is suppressed even if the predetermined mark is not detected. A predetermined sign may move out of the camera's photographing range while the person to be detected is performing work or the like. This notification device can prevent the transmission of an intrusion notification to the detected person even if a predetermined sign moves.

This notification device may have the following features. The notification unit suppresses the notification when a person different from the detected person is detected together with the detected person in the second imaging data. There is a high possibility that the person who is detected together with the detected person is a person who, like the detected person, has entered the monitored area with permission. Further, even if an ordinary person who is not permitted to enter the monitored area is detected together with the detected person, it is thought that the detected person will notice the intrusion of the ordinary person. Therefore, by suppressing the transmission of intrusion notifications in such situations, unnecessary intrusion notifications can be suppressed.

The disclosed technology can also be understood as a notification method and notification program.

Hereinafter, embodiments will be further described with reference to the drawings. The configuration of the embodiment shown below is an example, and the disclosed technology is not limited to the configuration of the embodiment.

<Embodiment>
FIG. 1 is a diagram illustrating an example of an intrusion detection system according to an embodiment. The intrusion detection system 500 illustrated in FIG. 1 includes a detection device 1, a surveillance camera 2, and a surveillance server 301 installed in a surveillance center 300. The intrusion detection system 500 detects the intrusion of a person onto the road 600 based on the video of the road 600 taken by the surveillance camera 2. In this embodiment, the road 600 is, for example, a road exclusively for automobiles where pedestrians are prohibited from entering. Road 600 is an example of a "monitored area."

The surveillance camera 2 is, for example, a digital camera having a charge coupled device (CCD) sensor or a complementary metal oxide semiconductor (CMOS) sensor. The surveillance camera 2 is not limited to a camera that takes pictures using visible light, but may be an infrared camera that takes pictures using infrared rays. The surveillance camera 2 can capture at least one of a moving image and a still image. The surveillance camera 2 is provided at a position where it can photograph the road 600. For example, the surveillance camera 2 may be installed at a position on the road 600 where it can photograph a specific area (an entrance to a motorway, an entrance to a service area, a road shoulder, etc.). Further, a plurality of surveillance cameras 2 may be arranged along the road 600 so that the entire road 600 can be monitored. In this embodiment, a 360-degree hemispherical camera is used as the surveillance camera 2. Region R1 in FIG. 1 exemplifies the range that can be photographed by the surveillance camera 2, which is a 360-degree hemispherical camera. The surveillance camera 2 is an example of a "camera".

The detection device 1 is an information processing device that acquires an image from the surveillance camera 2, analyzes the acquired image, and detects the intrusion of a person onto the road 600. When detecting a human intrusion, the detection device 1 transmits an intrusion notification to a monitoring server 301 installed in a monitoring center 300. Note that in FIG. 1, the detection device 1 and the surveillance camera 2 are separate devices, but the surveillance camera 2 and the detection device 1 may be a single device. The detection device 1 is an example of a "notification device" and a "computer."

The monitoring center 300 has a guard stationed there and monitors people entering the road 600. The monitoring server 301 is an information processing device operated by a monitoring staff resident at the monitoring center 300. When the monitoring server 300 receives the intrusion notification from the detection device 1, it notifies the monitor of the intrusion of a person onto the road 600. The supervisor may be notified by outputting a warning sound from a speaker or by outputting a warning message on a display.

The network N1 is a computer network that connects information processing devices so that they can communicate with each other. The network N1 is, for example, the Internet, a leased line, a local
ea Network (LAN), and can be wired or wireless.

<Hardware configuration of detection device 1>
FIG. 2 is a diagram illustrating an example of the hardware configuration of the detection device according to the embodiment. The detection device 1 is an information processing device including a central processing unit (CPU) 101, a main storage section 102, an auxiliary storage section 103, a communication section 104, a connection terminal 105, and a connection bus B1. The CPU 101, main memory section 102, auxiliary memory section 103, communication section 104, and connection terminal 105 are interconnected by a connection bus B1.

The CPU 101 is also called a microprocessor unit (MPU) or a processor. The CPU 101 is not limited to a single processor, and may have a multiprocessor configuration. Further, a single CPU 101 connected through a single socket may have a multi-core configuration. At least part of the processing executed by the CPU 101 is performed by a processor other than the CPU 101, for example, a dedicated processor such as a Digital Signal Processor (DSP), a Graphics Processing Unit (GPU), a numerical calculation processor, a vector processor, or an image processing processor. It's okay. Further, at least a portion of the processing executed by the CPU 101 may be executed by an integrated circuit (IC) or other digital circuit. Furthermore, at least a portion of the CPU 101 may include an analog circuit. Integrated circuits include large scale integrated circuits (LSIs), application specific integrated circuits (ASICs), and programmable logic devices (PLDs). The PLD includes, for example, a field-programmable gate array (FPGA). CPU 101 may be a combination of a processor and an integrated circuit. The combination is called, for example, a microcontroller unit (MCU), a system-on-a-chip (SoC), a system LSI, a chipset, or the like. In the detection device 1, the CPU 101 loads the program stored in the auxiliary storage unit 103 into the work area of the main storage unit 102, and controls peripheral devices through execution of the program. Thereby, the detection device 1 can execute processing that meets a predetermined purpose. The main storage unit 102 and the auxiliary storage unit 103 are recording media that can be read by the detection device 1.

The main storage unit 102 is exemplified as a storage unit directly accessed by the CPU 101. The main storage unit 102 includes Random Access Memory (RAM) and Read
Contains only memory (ROM).

The auxiliary storage unit 103 stores various programs and various data on a recording medium in a readable and writable manner. The auxiliary storage unit 103 is also called an external storage device. The auxiliary storage unit 103 stores an operating system (OS), various programs, various tables, and the like. The OS includes a communication interface program that exchanges data with an external device connected via the communication unit 104. External devices include, for example, other information processing devices and external storage devices connected via a computer network or the like. Note that the auxiliary storage unit 103 may be, for example, part of a cloud system that is a group of computers on a network.

The auxiliary storage unit 103 is, for example, an erasable programmable ROM (EPROM), a solid state drive (SSD), a hard disk drive (HDD), or the like. Further, the auxiliary storage unit 103 may be, for example, a Compact Disc (CD) drive device, a Digital Versatile Disc (DVD) drive device, a Blu-ray (registered trademark) Disc (BD) drive device, or the like.

The communication unit 104 is, for example, an interface with the network N1. The communication unit 104 communicates with an external device via the network N1.

The connection terminal 105 is a terminal for connecting an external device such as the surveillance camera 2. Examples of the connection terminal 105 include a Universal Serial Bus (USB) terminal, a High-Definition Multimedia Interface (HDMI (registered trademark)) terminal, and the like.

The detection device 1 may further include, for example, an input unit that receives operation instructions from a user or the like. Examples of such an input unit include input devices such as a keyboard, pointing device, touch panel, acceleration sensor, and voice input device.

The detection device 1 may include, for example, an output unit that outputs data processed by the CPU 101 and data stored in the main storage unit 102. As such an output unit, a cathode ray tube (CRT) display, a liquid crystal
Examples of output devices include a display (LCD), a plasma display panel (PDP), an electroluminescence (EL) panel, an organic EL panel, and a printer.

<Processing block of detection device 1>
FIG. 3 is a diagram illustrating an example of processing blocks of the detection device according to the embodiment. The detection device 1 includes an acquisition section 11 , a person detection section 12 , a sign detection section 13 , a determination section 14 , a notification section 15 , a human feature storage section 16 , and a sign feature storage section 17 . The detection device 1 has the acquisition unit 11, the person detection unit 12, the sign detection unit 13, the determination unit 14, The notification unit 15, the human feature storage unit 16, the marker feature storage unit 17, and other units execute processing.

The human feature amount storage unit 16 stores feature amounts indicating "people" in the video captured by the surveillance camera 2. The human feature amount storage unit 16 only needs to store enough feature amounts to be able to determine whether a person is a person or not, and does not need to have enough feature amounts to be able to identify an individual person. The mark feature amount storage unit 17 stores feature amounts indicating a predetermined mark. The predetermined sign can be, for example, a sign indicating a maintenance work location on the road 600. In this embodiment, "vehicle" is adopted as the predetermined sign. That is, in the present embodiment, the sign feature storage unit 17 stores a feature amount indicating a "vehicle" in the video captured by the surveillance camera 2. The sign feature storage unit 17 only needs to store feature amounts that can be used to determine whether or not the vehicle is a vehicle, and do not store feature amounts that can identify the model, manufacturer, etc. of the vehicle. Tomoyoshi. The human feature storage unit 16 and the marker feature storage unit 17 are constructed in the auxiliary storage unit 103, for example.

The acquisition unit 11 acquires an image of the road 600 captured by the surveillance camera 2. The acquisition unit 11 stores the acquired image of the road 600 in the auxiliary storage unit 103. The acquisition unit 11 is an example of an “acquisition unit”. Video is an example of "imaging data."

The person detection unit 12 analyzes the video acquired by the acquisition unit 11 and detects a person. The person detection unit 12 extracts feature points of the subject from the video acquired by the acquisition unit 11, for example. The person detection section 12 detects a person included in the video based on the extracted feature amount and the feature amount stored in the person feature amount storage section 16. It should be noted that the person detection unit 12 does not need to perform highly accurate detection such as identifying individual people, but only needs to be able to detect whether or not it is a person.

The sign detection unit 13 analyzes the video acquired by the acquisition unit 11 and detects a predetermined sign. As described above, since a vehicle is used as a sign in this embodiment, the sign detection unit 13 detects a vehicle. For example, the marker detection unit 13 extracts feature points of the subject from the video acquired by the acquisition unit 11. The sign detection unit 13 detects a vehicle included in the video based on the extracted feature amount and the feature amount stored in the sign feature storage unit 17. Note that the sign detection unit 13 does not need to perform highly accurate detection such as specifying what kind of vehicle the detected vehicle is, but only needs to be able to detect whether or not it is a vehicle.

The determination unit 14 determines whether the vehicle detected by the sign detection unit 13 is stopped. For example, the determination unit 14 determines that the vehicle is stopped when the position (coordinates) of the vehicle detected by the sign detection unit 13 is the same in multiple consecutive frames in the video acquired by the acquisition unit 11. . Note that, when the sign detection unit 13 detects a plurality of vehicles, the determination unit 14 may determine whether there is at least one stopped vehicle. If there is at least one stopped vehicle, the determination unit 14 prevents the notification unit 15 (described later) from transmitting an intrusion notification.

The notification unit 15 transmits an intrusion notification to the monitoring server 301 when the person detection unit 12 detects a person. Note that if the determination unit 14 determines that the vehicle detected by the vehicle detection unit 13 is stopped (not moving), the notification unit 15 will not send an intrusion notification even if the person detection unit 12 detects a person. This is suppressed by the determination unit 14. The notification section 15 is an example of a "notification section."

Here, in this embodiment, a situation in which transmission of an intrusion notification is suppressed and a situation in which an intrusion notification is transmitted will be further described with reference to the drawings.

(When suppressing the sending of intrusion notifications)
FIG. 4 is a first diagram illustrating an example of a situation in which the determination unit inhibits the notification unit from transmitting an intrusion notification in the embodiment. In FIG. 4, a maintenance worker P1 who performs maintenance work on the road 600, a maintenance vehicle V1, road cones C1, C1, and a general vehicle V2 are illustrated. The maintenance worker P1 is a maintenance worker who performs maintenance work at a work location where maintenance work is performed on the road 600. The maintenance vehicle V1 is a vehicle that transports the maintenance worker P1 to a work site. The maintenance vehicle V1 includes a guide plate V1a on which is written "maintenance work in progress". The maintenance vehicle V1 that has transported the maintenance worker P1 to the maintenance work site stops at the work site. The maintenance vehicle V1 that has stopped at the work site notifies surrounding vehicles (for example, general vehicles V2) traveling on the road 600 that maintenance work is being performed using the guide plate V1a. Furthermore, road cones C1, C1 are installed at the work place to indicate the boundaries of the work target place.

The acquisition unit 11 acquires a video of the situation illustrated in FIG. 4 from the surveillance camera 2. Based on the acquired video, the person detection section 12 detects the maintenance worker P1, and the sign detection section 13 detects the maintenance vehicle V1 and the general vehicle V2. In this situation, since the maintenance vehicle V1 among the vehicles detected by the sign detection unit 13 is stopped, the determination unit 14 prevents the notification unit 15 from transmitting the intrusion notification.

As illustrated in FIG. 4, when maintenance worker P1 performs maintenance work on road 600, maintenance vehicle V1 stops around (near) maintenance worker P1. The road 600, which is a motorway, has no or very few traffic lights, so vehicles rarely stop on the road 600. There is a high probability that a vehicle stopped on such a road 600 is a vehicle related to maintenance work, exemplified by maintenance vehicle V1. Furthermore, there is a high probability that a person present near a stopped vehicle is a maintenance worker. Therefore, in the present embodiment, if a stopped vehicle is detected in the image acquired by the acquisition unit 11, even if the person detection unit 12 detects a person, the determination unit 14 detects an intrusion by the notification unit 15. Suppress notifications from being sent. That is, in this embodiment, a person detected together with a stopped vehicle is considered to be a maintenance worker who is permitted to enter the road 600, and is excluded from the target of the intrusion notification.

FIG. 5 is a second diagram illustrating an example of a situation in which the determination unit inhibits the notification unit from transmitting an intrusion notification in the embodiment. In the example of FIG. 5, in addition to the situation illustrated in FIG. 4, an ordinary person P2 is present in the video captured by the surveillance camera 2. That is, in the situation illustrated in FIG. 5, the person detection unit 12 detects the maintenance worker P1 and the general public P2 based on the video that the acquisition unit 11 acquires from the surveillance camera 2. Similarly, the sign detection unit 13 detects the maintenance vehicle V1 and the general vehicle V2. In this embodiment, when a stopped vehicle is detected, the determination unit 14 prevents the notification unit 15 from transmitting an intrusion notification even if the people detected by the person detection unit 12 include ordinary people.

As described above, there is a high probability that a vehicle stopped on the road 600 is a vehicle related to maintenance work. It is considered that at least one maintenance worker P1 exists near such a vehicle. Therefore, even if the general public P2 is included in the plurality of people detected by the person detection unit 12, it is assumed that the maintenance worker P1 notices the presence of the general public P2. Therefore, if the vehicle detected by the sign detection unit 13 is stopped, there is no problem even if the situation where the general public P2 is included in the people detected by the person detection unit 12 is excluded from the target of the intrusion notification. is considered to be small. In the present embodiment, the intrusion notification is sent based on whether or not a vehicle that has stopped with a person is detected, rather than depending on whether the person detected by the person detection unit 12 is a maintenance worker P1 or a general person P2. It is determined whether or not to do so. Therefore, as described above, it is not necessary to cause the person detection unit 12 to perform highly accurate detection such as identifying the detected person.

(If sending of intrusion notifications is not suppressed)
FIG. 6 is a first diagram illustrating an example of a situation in which the determination unit does not suppress transmission of an intrusion notification in the embodiment. That is, the situation illustrated in FIG. 6 is an example of a situation in which the notification unit 15 transmits an intrusion notification. In the example of FIG. 6, a general person P2 who has entered the road 600 and a general vehicle V2 traveling on the road 600 are present in the video taken by the surveillance camera 2. That is, in the case of FIG. 6, the person detection section 12 detects a general person P2, and the sign detection section 13 detects a general vehicle V2. Since the general vehicle V2 detected by the sign detection unit 13 is running (not stopped), the determination unit 14 does not prevent the notification unit 15 from transmitting the intrusion notification. As a result, the notification unit 15 transmits an intrusion notification to the monitoring server 301 to notify that the general public P2 has invaded the road 600.

FIG. 7 is a second diagram illustrating an example of a situation in which the determination unit does not suppress transmission of an intrusion notification in the embodiment. That is, the situation illustrated in FIG. 7 is also an example of a situation in which the notification unit 15 transmits an intrusion notification. In the example of FIG. 7, the maintenance vehicle V1 and the general public P2 who have entered the road 600 are present in the video taken by the surveillance camera 2. In the situation shown in FIG. 7, the maintenance vehicle V1 is traveling on the road 600 with load cones C1 and C1 loaded thereon toward the work site. That is, in the case of the situation illustrated in FIG. 7, the person detection section 12 detects the general public P2, and the sign detection section 13 detects the maintenance vehicle V1.

The determination unit 14 does not prevent the notification unit 15 from transmitting the intrusion notification because the maintenance vehicle V1 detected by the sign detection unit 13 is running (not stopped). As a result, the notification unit 15 transmits an intrusion notification to the monitoring server 301 to notify that the general public P2 has invaded the road 600. In the present embodiment, the intrusion is determined based on whether the vehicle detected with a person is stopped or not, rather than depending on whether the vehicle detected by the sign detection unit 13 is a maintenance vehicle V1 or a general vehicle V2. It is determined whether or not to send a notification. Therefore, as described above, it is not necessary to cause the sign detection unit 13 to perform highly accurate detection such as identifying what kind of vehicle the detected vehicle is.

<Processing flow>
FIG. 8 is an example of a processing flow of the intrusion detection system according to the embodiment. An example of the processing flow of the intrusion detection system 500 according to the embodiment will be described below with reference to FIG. 8.

At T1, the acquisition unit 11 acquires an image captured by the surveillance camera 2. The acquisition unit 11 stores the acquired video in the auxiliary storage unit 103.

At T2, the person detection unit 12 extracts the feature amount of the subject from the video acquired at T1. The person detection section 12 detects a person from the video acquired at T1 based on the extracted feature amount and the feature amount stored in the human feature amount storage section 16. If a person is detected from the video acquired at T1 (YES at T2), the process proceeds to T3. If no person is detected from the video acquired at T1 (NO at T2), the process ends. Note that in order to improve the accuracy of human detection, if YES is obtained at T2 for a plurality of consecutive frames, the process may proceed to T3.

At T3, the marker detection unit 13 extracts the feature amount of the subject from the video acquired at T1. The sign detection unit 13 detects a vehicle from the image acquired at T1 based on the extracted feature amount and the feature amount stored in the sign feature storage unit 17. If a vehicle is detected from the image acquired at T1 (YES at T3), the process proceeds to T4. If no vehicle is detected from the image acquired at T1 (NO at T3), it is determined that there is an intruder, and the process proceeds to T5.

At T4, the determination unit 14 determines whether the vehicle detected at T3 is stopped. Note that if a plurality of vehicles are detected at T3, the determination unit 14 may determine whether there is at least one stopped vehicle. If there is a stopped vehicle among the vehicles detected at T3 (YES at T4), the process is ended. If none of the vehicles detected at T3 have stopped (NO at T4), the process proceeds to T5.

At T5, the notification unit 15 transmits an intrusion notification to the monitoring server 301. The monitoring server 301 that has received the intrusion notification notifies the monitor of the intrusion of a person onto the road 600.

At T6, the acquisition unit 11 determines whether to perform the next detection. If the detection is performed periodically (periodically at T6), the acquisition unit 11 advances the process to T7. If the detection is to be performed continuously (continuously at T6), the acquisition unit 11 advances the process to T1. If the next detection is not to be performed (NO at T6), the process is ended. Note that the information used for the determination at T6 (information indicating whether to perform periodic detection, whether to perform continuous detection, or whether to perform the next detection) is determined by, for example, the observer, and is stored in the auxiliary memory. The information is stored in the section 103.

At T7, the acquisition unit 11 waits until the next detection cycle, and then advances the process to T1. Note that the information indicating the detection cycle may be determined by, for example, a supervisor and stored in the auxiliary storage unit 103.

<Actions and effects of embodiments>
As described above, there is a high probability that a vehicle stopping on road 600, which is a motorway, is a vehicle related to maintenance work, exemplified by maintenance vehicle V1. Furthermore, there is a high probability that a person present near such a stopped vehicle is a maintenance worker performing maintenance work on the road 600, as exemplified by maintenance worker P1. In this embodiment, whether or not the person is likely to be a maintenance worker is determined based on the presence or absence of a stopped vehicle, and when the probability that the person is a maintenance worker is high, notification of intrusion to the monitoring server 301 is suppressed. By employing such a configuration, the intrusion detection system 500 according to the embodiment is prevented from transmitting an intrusion notification even to a person who is permitted to enter the road 600.

According to the present embodiment, since the transmission of an intrusion notification to a person who is permitted to enter the road 600 is suppressed, if the intrusion notification is a false alarm, a monitor stationed at the monitoring center 300 This reduces the time and effort required to confirm whether or not there has been an intrusion.

In the embodiment, the person detection unit 12 does not need to perform highly accurate detection to identify the detected person, but only needs to be able to detect whether or not the detected person is a person. Further, the sign detection unit 13 does not need to perform highly accurate detection such as specifying what kind of vehicle the detected vehicle is, but only needs to be able to detect whether or not it is a vehicle. Therefore, the processing load on the detection device 1 can be reduced compared to the case where the person detection unit 12 and the sign detection unit 13 perform more accurate detection. Furthermore, since the person detection unit 12 and the sign detection unit 13 do not perform highly accurate detection, a camera with a lower resolution (that is, a cheaper camera) can be used as the surveillance camera 2. Therefore, according to the embodiment, the construction cost of the intrusion detection system 500 can be reduced.

<First modification example>
In the embodiment, when a person and a stopped vehicle are included in the same video, transmission of an intrusion notification is suppressed. In the first modification, a configuration will be described in which, after the transmission of the intrusion notification is suppressed, even if the stopped vehicle moves out of the photographing range of the surveillance camera 2, the transmission of the intrusion notification is suppressed. Components that are the same as those in the embodiment are denoted by the same reference numerals, and the description thereof will be omitted. The first modification will be described below with reference to the drawings.

FIG. 9 is a diagram illustrating an example of a processing block of the detection device according to the first modification. The detection device 1a illustrated in FIG. 9 differs from the detection device 1 according to the embodiment in that it includes a person detection unit 12a instead of the person detection unit 12.

In addition to detecting a person in a video, the person detection unit 12a also tracks a person detected together with a stopped vehicle. There is no particular limitation on the method by which the person detection unit 12a tracks a person, and various known techniques can be applied. The person detection unit 12a may, for example, surround the detected person in the video with a rectangle and track the movement of the rectangle in the video.

FIG. 10 is a diagram illustrating an example of a situation in which transmission of an intrusion notification is suppressed in the first modification. In FIG. 10(A), a stopped maintenance vehicle V1 is illustrated along with a maintenance worker P1. In the situation of FIG. 10(A), the determination unit 14a surrounds the maintenance worker P1 present in the video together with the stopped maintenance vehicle V1 in a rectangle R2, and sets it as a tracking target. FIG. 10(B) exemplifies a situation in which the maintenance vehicle V1 moves out of the photographing range of the surveillance camera 2 after FIG. 10(A). Since the person detection unit 12a tracks the maintenance worker P1 by surrounding it with a rectangle R2, it prevents the notification unit 15 from transmitting an intrusion notification.

FIG. 11 is an example of a processing flow of the intrusion detection system according to the first modification. Hereinafter, an example of the processing flow of the intrusion detection system according to the first modification will be described with reference to FIG. 11.

At T21, the person detection unit 12a determines whether the person detected at T2 is the person to be tracked. If the person is the person to be tracked (YES at T21), the process is ended. If the person is not the person to be tracked (NO at T21), the process proceeds to T3.

In the first modified example, if an affirmative determination is made at T4 (YES at T4), the process proceeds to T22. In the case of a negative determination at T4 (NO at T4), the process proceeds to T5.

At T22, the person detection unit 12a starts tracking the person detected at T2. For example, the person detection unit 12a surrounds the person detected at T2 with a rectangle R2, and tracks the movement of the rectangle R2 in the video.

According to the first modification, if a person is detected together with a stopped vehicle, even after the vehicle moves out of the photographing range of the surveillance camera 2, the person can be excluded from receiving an intrusion notification.

In the first modified example, FIG. 12 illustrates a situation in which an ordinary person P2 newly appears in the video of the surveillance camera 2 in a situation where the maintenance worker P1 is being tracked by the person detection unit 12a. Even in such a case, in the first modification, since the person detection unit 12a tracks the maintenance worker P1, the intrusion notification is suppressed. Since it is assumed that the maintenance worker P1 notices the presence of the general public P2, it is thought that there will be little problem even if the transmission of the intrusion notification is suppressed in such a case.

<Second modification example>
In the embodiment, an example was shown in which the transmission of an intrusion notification is suppressed when a person and a stopped vehicle are included in the same video, but when using a wide-angle camera such as a half-celestial camera that can capture a wide range Even if the person is far away from the stopped vehicle, the person may be determined to be a maintenance worker associated with the vehicle, and the transmission of an intrusion notification may be suppressed. In the second modification, when a person is detected from a camera image, an area is defined around the coordinates of the detected person in which it is determined whether or not to suppress the transmission of an intrusion notification based on the presence or absence of a vehicle, and within that area We will explain how to suppress transmission when a vehicle is present or the vehicle is stopped.

FIG. 13 is a diagram illustrating an example in which the range of presence of a sign that suppresses transmission of an entry notification in an image is limited in the second modification. FIG. 13 shows this situation. When a maintenance worker P1 is detected from the image, a deterrent area 101 is set around the maintenance worker P1, and when an ordinary person P2 is detected, an inhibition area 102 is set around the ordinary person P2. . In the example of FIG. 13, since the vehicle V3 is included in the suppression area 101, transmission of the intrusion notification is suppressed for the maintenance worker P1. On the other hand, since no vehicle is included in the deterrent area 102, an intrusion notification will be sent to the general public P2. In this way, by setting a suppression area for each detected person, it is possible to prevent intrusion notifications from being omitted. This effect is particularly effective in cameras that monitor a wide area with a single camera, such as a hemispherical camera.

<Third modification example>
In the embodiment, since the explanation has been made using a flowchart, only vehicles as a sign are judged. However, the predetermined sign may be not only a vehicle but also a road cone or a guide board indicating construction work, or a combination of these may be used for determination. If these combinations are expressed as decision branches, it will become complicated and difficult to implement. In such a case, various situations can be handled by storing the detection state of a predetermined marker, which serves as a determination material, in a determination table and making a determination with reference to the determination table. Furthermore, by storing information with different detection times in the determination table, it is possible to perform determinations that correspond to temporal changes. In the third modification, determination using such a determination table will be described.

FIG. 14 is a diagram illustrating an example of a determination table according to the third modification. The determination table 14a illustrated in FIG. 14 is stored in the auxiliary storage unit 103, for example. In the third modification, the determination unit 14 refers to the determination table 14a and determines whether there is an intrusion notification.

The determination table 14a illustrated in FIG. 14 is a table used to determine whether there is an intrusion notification based on the human detection result, vehicle detection result, road cone detection result, and the previous and current detection results for each detection result. This is an example. The determination table 14a is stored in the auxiliary storage unit 103, for example. The determination table 14a includes the following items: "No.", "Previous detection result", "Current detection result", and "Intrusion notification".

Each of the "previous detection result" and the "current detection result" includes the following items: "person", "vehicle", and "road cone". In each item of "person," "vehicle," and "road cone," "- (hyphen)" indicates a non-detected state, and "detected" indicates a detected state. "*" indicates that it does not need to be detected. In addition, for each item of "person", "vehicle", and "road cone", "moving" and "stopping" are defined as "stopping" if the change in the detection coordinates of the current detection result is within a predetermined range compared to the previous detection result. If a coordinate change exceeding a predetermined value is detected, it is determined to be a "movement".

A1 to A4 illustrate determinations made regardless of the presence or absence of a predetermined sign such as a vehicle or a road cone. The determination unit 14 does not suppress the intrusion notification (that is, the notification unit 15 transmits the intrusion notification) only when a person is detected twice in a row (the previous detection result and the current detection result), thereby reducing noise in the image. It is possible to prevent intrusion notifications from being sent due to temporary false detections caused by the above.

B1 and B2 are related to determination of a combination of vehicle detection and human detection; in B2, a person is detected and the vehicle is moving, so the determination unit 14 does not suppress transmission of an intrusion notification; Since the maintenance vehicle and the maintenance worker are stopped, the determination unit 14 determines that the maintenance vehicle and maintenance personnel are detected, and suppresses the transmission of the intrusion notification. This determination can have the effect of suppressing unnecessary intrusion notifications.

C1 and C2 relate to determining the combination of road cone detection and human detection. In C1, since the installed road cone and the person were detected at the same time, the determination unit 14 determines that the person is a worker and suppresses the transmission of the intrusion notification. At C2, the road cone is moving and it can be determined that a person is holding the road cone and setting up the construction area, so in this case as well, the determining unit 14 suppresses the transmission of the intrusion notification. By using the determination table 14a, it is possible to easily change the determination depending on the difference in the markers.

D1 to D5 are related to determination of combinations of vehicle, road cone, and human detection. Since either the vehicle or the road cone is stopped from D1 to D3, the determination unit 14 determines that the detected person is a person related to the construction work and suppresses the transmission of the intrusion notification. At D4, since the vehicle and the road cone are moving, it is considered that the construction work has been completed and the worker is cleaning up the road cone, so the determination unit 14 suppresses the transmission of the intrusion notification. If a person remains for a long time even after the vehicle and road cone have moved in D5, it may be determined that something unusual has occurred and an alarm may be issued. When changing the determination based on the elapsed time, by incorporating previous detection results corresponding to the elapsed time and configuring the determination table 14a, it becomes possible to determine various situations and appropriately set the presence or absence of an intrusion notification. .

<Other variations>
Although the intrusion detection system 500 according to the embodiment detects the intrusion of a person onto the road 600, the target for which the intrusion detection system detects the intrusion of a person is not limited to the road 600. An intrusion detection system may detect, for example, intrusion of a person onto railway tracks or an intrusion of a person onto an airport runway. When detecting human intrusion onto a railway track, a maintenance vehicle such as a rail grinding vehicle, electric inspection vehicle, snow removal train, etc. used for track maintenance work is used as a predetermined sign, and the characteristic values of the maintenance vehicle are used. may be stored in the marker feature storage unit 17. Then, the sign detection unit 13 detects the maintenance vehicle in the video, and the determination unit 14 determines whether the maintenance vehicle is stopped or moved. Similarly, when detecting human intrusion onto an airport runway, maintenance vehicles such as towing cars, cargo tractors, and high-lift loaders used for runway maintenance work should be used as designated signs. Bye.

In the embodiment, "vehicle" is used as the predetermined sign, but the predetermined sign is not limited to vehicles. The predetermined sign may be, for example, a road cone or a guide board indicating construction work. For example, in the situation illustrated in FIG. 4, when the sign detection unit 13 detects the road cone C1 and the road cone C1 is detected together with a person, transmission of the intrusion notification may be suppressed. When the sign detection unit 13 detects a road cone, the sign feature storage unit 17 may store a feature amount indicating the road cone.

Note that since the road cone C1 itself does not run on the road 600, the determination by the determination unit 14 as to whether or not the road cone C1 is stopped may be omitted. However, as illustrated in FIG. 7, a situation may also be considered in which the maintenance vehicle V1 loads the road cone C1 and heads to the work site. Considering this situation, even when the road cone C1 is employed as the predetermined sign, it is more preferable for the determination unit 14 to determine whether the road cone C1 is stopped.

In the embodiment, a 360-degree hemispherical camera is used as the surveillance camera 2, but the surveillance camera 2 is not limited to a 360-degree hemispherical camera. The surveillance camera 2 may be a normal camera other than a 360-degree hemispherical camera as long as it can photograph the road 600. In addition, by using the surveillance camera 2 as a 360-degree hemispherical camera, the photographing range of the surveillance camera 2 can be expanded as much as possible, so the number of surveillance cameras 2 installed on the road 600 can be reduced. , the installation cost of the intrusion detection system 500 can be reduced.

In the embodiment, the detection device 1 transmits an intrusion notification to the monitoring server 301, and the monitoring server 301 that receives the intrusion notification notifies the monitoring officer of the intrusion of a person onto the road 600. A message urging the general public P2 who has entered the road 600 to vacate the road 600 may be outputted via an electronic bulletin board or speaker installed on the road 600. Note that the detection device 1 may output the message urging eviction via an electronic bulletin board installed on the road 600, a speaker, or the like.

The embodiments and modifications disclosed above can be combined.

<<Computer-readable recording medium>>
An information processing program that causes a computer or other machine or device (hereinafter referred to as a computer or the like) to realize any of the above functions can be recorded on a computer-readable recording medium. Then, by causing a computer or the like to read and execute the program on this recording medium, the function can be provided.

Here, a computer-readable recording medium is a recording medium that stores information such as data and programs through electrical, magnetic, optical, mechanical, or chemical action and can be read by a computer, etc. means. Examples of such recording media that are removable from computers and the like include flexible disks, magneto-optical disks, Compact Disc Read Only Memory (CD-ROM), Compact Disc-Recordable (CD-R), and Compact Disc-ReWriterable. (CD-RW), Digital Versatile Disc (DVD), Blu-ray Disc (BD), Digital Audio Tape (DAT), 8mm tape, and memory cards such as flash memory. In addition, there are hard disks, ROMs, etc. as recording media fixed to computers and the like.

1, 1a...Detection device 101...CPU
102... Main memory section 103... Auxiliary memory section 104... Communication section 105... Connection terminal 11... Acquisition section 12... Person detection section 13... Sign detection section 14... Judgment unit 15... Notification unit 16... Person feature storage unit 17... Sign feature storage unit 101, 102... Suppression area 300... Monitoring center 301... Monitoring server 500... Intrusion detection system 600...road B1...connecting bus V1...maintenance vehicle V1a...information board V2...general vehicle P1...maintenance worker P2...general public C1...road cone R1 ...Region R2...Rectangle

Claims (5)

A notification device that notifies of a person's intrusion into a monitored area,
an acquisition unit that acquires image data obtained by photographing the monitored area from a camera;
a notification unit that notifies the person of intrusion into the monitored area when a person is detected in the imaged data;
The notification unit suppresses the notification when detecting a vehicle together with a person in the imaged data and detecting that the vehicle is stopped .
Notification device. The notification unit may detect a person detected together with the vehicle in the image data, if the vehicle is detected in second image data acquired by the acquisition unit after image data including the vehicle and the person. suppressing said notification even if it is not done;
The notification device according to claim 1 . The notification unit suppresses the notification when a person different from the detected person is detected together with the detected person in the second imaging data.
The notification device according to claim 2 . A notification method for notifying human intrusion into a monitored area, comprising:
The computer is
A process of acquiring image data obtained by photographing the monitored area from a camera;
executing a process of notifying an intrusion of a person into the monitored area when a person is detected in the imaged data;
In the notification processing, when a vehicle is detected in the imaged data together with the person , and it is detected that the vehicle is stopped, the notification is suppressed;
Notification method. A notification program that notifies of human intrusion into a monitored area,
to the computer,
A process of acquiring image data obtained by photographing the monitored area from a camera;
executing a process of notifying the intrusion of a person into the monitored area when a person is detected in the imaged data;
In the notification processing, when a vehicle is detected in the imaged data together with the person , and it is detected that the vehicle is stopped, the notification is suppressed;
Notification program.

Images