Fire detection device, fire detection method, and fire detection program

Abstract

To keep the number of infrared cameras allocated to a target area low without deteriorating the accuracy of determination of whether or not there is a fire occurring in a target area where mobile objects such as vehicles move.SOLUTION: A detection unit detects a mobile object that has stopped in a target area. An adjustment unit changes an imaging area of an infrared camera to the stop location of the mobile object that has been detected to be stopped by the detection unit. A first determination unit processes frame images taken by the infrared camera and determines whether or not a fire is occurring at the stop location.SELECTED DRAWING: Figure 3

Description

Technical Field

[0001] This invention relates to a technology for detecting the occurrence of a fire in an area where an unspecified number of moving bodies (such as vehicles and autonomous mobile robots) move.

Background Art

[0002] Conventionally, as a technology for detecting a fire occurring in a tunnel where a vehicle travels, there is, for example, the system disclosed in Patent Document 1. In this Patent Document 1, a frame image in the tunnel captured by an infrared camera is processed to detect the occurrence of a fire.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Patent Document 2

Summary of the Invention

Problems to be Solved by the Invention

[0004] However, in the system described in Patent Document 1, in order to detect a fire occurring in a tunnel, many infrared cameras must be used. Specifically, the inside of the tunnel is divided into a plurality of spaces, and at least one infrared camera must be assigned to each of the divided spaces. The infrared camera is installed so that the assigned space becomes the imaging area (imaging region). Therefore, the system described in Patent Document 1 requires more infrared cameras than the number of spaces divided in the tunnel.

[0005] Also, as described in Patent Document 1, in order to prevent a situation where a blind spot (a place not imaged by the infrared camera) occurs in the tunnel due to occlusion by a large vehicle such as a truck or a trailer traveling in the tunnel, further addition of infrared cameras is required.

[0006] Furthermore, if there are areas within the tunnel that are not captured by the infrared camera, frame images of these areas cannot be obtained, making it impossible to detect the occurrence of a fire.

[0007] In order to reduce the number of infrared cameras, increasing the size of each section within the tunnel (i.e., increasing the imaging area allocated to each infrared camera) results in coarse images (low-resolution images) captured by the infrared cameras, which reduces the accuracy of determining whether or not a fire is occurring.

[0008] The objective of this invention is to provide a technology that can reduce the number of infrared cameras assigned to a target area without reducing the accuracy of determining whether or not a fire is occurring in that area where a moving object such as a vehicle is traveling. [Means for solving the problem]

[0009] To achieve the above objective, the fire detection device of this invention is configured as follows.

[0010] The detection unit detects a moving object that has stopped within the target area. The target area is, for example, a road on which a vehicle, which is the moving object, is traveling. In this case, the target area may be inside a tunnel or not. The moving object may also be an autonomous mobile robot. In this case, the target area may be, for example, a warehouse or factory where the autonomous mobile robot, which is the moving object, is transporting goods. The moving object may be other than the vehicle or autonomous mobile robot exemplified here. The target area can be determined according to the moving object.

[0011] The adjustment unit changes the imaging area of ​​the infrared camera to the stopping position of the moving object detected by the detection unit. For example, if the infrared camera is mounted on a tripod head that can be changed in two orthogonal axes (e.g., pan and tilt directions), the adjustment unit generates a drive signal to the tripod head that changes the imaging area of ​​the infrared camera to the stopping position of the moving object detected by the detection unit. Furthermore, the adjustment unit may generate a zoom signal that changes the imaging magnification of the infrared camera according to the stopping position of the moving object detected by the detection unit.

[0012] The first determination unit processes the frame images captured by the infrared camera and determines whether a fire has occurred at the stopping position.

[0013] This configuration allows the imaging area of ​​the infrared camera to be aligned with the area where a fire is most likely to occur. This enables the determination of whether a fire is occurring by processing the frame images captured by the infrared camera without increasing the size of the area corresponding to the infrared camera's imaging area. Therefore, the accuracy of the fire detection is not reduced. Furthermore, the area assigned to the infrared camera for determining whether a fire is occurring can be expanded. In other words, the number of infrared cameras used to capture frame images for determining whether a fire is occurring within the target area where a mobile object such as a vehicle or autonomous robot is moving can be reduced.

[0014] Furthermore, for example, if the road on which the vehicle is traveling is within the target area (the target area may or may not be inside a tunnel), the system may include a second determination unit that determines whether the stop of the moving object detected by the detection unit is due to traffic congestion. In this case, the first determination unit can reduce the processing load (especially the processing load for image processing of frame images captured by the infrared camera) by configuring it so that it does not determine whether a fire has occurred at the stopping location if the second determination unit determines that the stop is due to traffic congestion.

[0015] Also, for example, the detection unit may detect a moving object stopped within the target area by processing the frame image of a visible light camera that has imaged the target area, or may be based on the detection result of a moving object by a radio wave radar (millimeter wave radar) or a lidar, or may be based on the detection result of the presence or absence of an object by a photoelectric sensor or the like, or may be performed using other types of sensors. In particular, in a target area where sensors such as a visible light camera, a radio wave radar, and a lidar are installed for traffic situation monitoring or measurement, by using this existing sensor (without newly installing a sensor), a moving object stopped within the target area can be detected.

Effects of the Invention

[0016] According to this invention, the number of infrared cameras assigned to this target area can be suppressed without reducing the determination accuracy of whether a fire has occurred within the target area where a moving object such as a vehicle moves.

Brief Description of the Drawings

[0017] [Figure 1] It is a diagram showing the fire detection system of this example. [Figure 2] It is a schematic diagram showing the target area where the occurrence of a fire is detected by the fire detection system of this example. [Figure 3] It is a block diagram showing the configuration of the main part of the fire detection device of this example. [Figure 4] It is a flowchart showing the operation of the fire detection device of this example. [Figure 5] It is a block diagram showing the configuration of the main part of the fire detection device of Modification 1. [Figure 6] It is a flowchart showing the operation of the fire detection device of Modification 1.

Modes for Carrying Out the Invention

[0018] Hereinafter, embodiments of this invention will be described.

[0019] <1. Application Example> FIG. 1 is a schematic diagram showing the fire detection system of this example. FIG. 2 is a schematic diagram showing the target area where the occurrence of a fire is detected by the fire detection system of this example. The fire detection system 100 of this example detects a fire in a tunnel in which a road on which the vehicle 110 travels is formed. In this example, the inside of the tunnel where the fire is detected corresponds to the target area referred to in this invention. Also, in this example, the vehicle 110 corresponds to the moving body referred to in this invention.

[0020] As shown in FIG. 1, the fire detection system 100 of this example includes a fire detection device 1, a plurality of infrared cameras 2 (2-1 to 2-n), a plurality of cameras 5 (5-1 to 5-m), and a guide plate 7.

[0021] The guide plate 7 is installed in front of the entrance of the tunnel. The guide plate 7 displays a guide message according to the traffic situation in the tunnel or the like to the driver of the vehicle 110 entering the tunnel. For example, when a fire is occurring in the tunnel, the guide message displayed on the guide plate 7 is, for example, "Entry prohibited, fire occurring in the tunnel". Also, when a traffic jam is occurring in the tunnel, the guide message displayed on the guide plate 7 is, for example, "Drive carefully, traffic jam in the tunnel".

[0022] The plurality of infrared cameras 2 (2-1 to 2-m) are, for example, far-infrared cameras. The infrared cameras 2 (2-1 to 2-m) are respectively attached to the corresponding pan-tilt units 3 (3-1 to 3-m). The pan-tilt unit 3 changes the angle (imaging direction) of the infrared camera 2 according to the drive signals in the pan direction and the tilt direction input from the fire detection device 1. That is, the infrared camera 2 is changed in angle in two orthogonal axes (pan direction and tilt direction) by the pan-tilt unit 3. Also, the infrared camera 2 changes the imaging magnification according to the Zoom signal input from the fire detection device 1. That is, the infrared camera 2 is operated as a so-called PTZ camera.

[0023] In this example of the fire detection system 100, the infrared cameras 2 are installed at approximately constant intervals in the direction of travel of the vehicle 110. In the example shown in Figure 2, the distance between infrared cameras 2-1 and 2-2 in the direction of travel of the vehicle 110 is approximately the same as the distance between infrared cameras 2-2 and 2-3 in the direction of travel of the vehicle 110. In this example, the installation positions of the infrared cameras 2 in the width direction of the tunnel (vehicle width direction) are alternating in the direction of travel of the vehicle 110. In the example shown in Figure 2, infrared cameras 2-1 and 2-3 are installed on the left shoulder (or roadside) in the width direction of the tunnel (vehicle width direction), and infrared camera 2-2 is installed on the right shoulder (or roadside) in the width direction of the tunnel (vehicle width direction). In this example of the fire detection system 100, multiple infrared cameras 2 are installed as shown in Figure 2 to prevent areas that cannot be imaged by any of the infrared cameras 2 (so-called blind spots) from being created due to occlusion by vehicles 110 (especially large vehicles such as trucks and trailers) traveling inside the tunnel.

[0024] Furthermore, the multiple infrared cameras 2 do not necessarily have to be installed in the configuration shown in Figure 2, as long as it is possible to prevent areas from being captured by any of the infrared cameras 2 due to occlusion caused by vehicles 110 (especially large vehicles such as trucks and trailers) traveling inside the tunnel.

[0025] Furthermore, while areas that cannot be imaged by any of the infrared cameras 2 may temporarily occur due to occlusion, a configuration may be provided that does not include either the infrared camera 2 installed on the left shoulder (or roadside) in the width direction (vehicle width direction) of the tunnel shown in Figure 2 (2-1, 2-3, etc.) or the infrared camera 2 installed on the right shoulder (or roadside) in the width direction (vehicle width direction) of the tunnel shown in Figure 2 (2-2, 2-n, etc.).

[0026] Camera 5 is a video camera that captures visible light images. The frame rate of camera 5 is several tens of frames / sec (for example, 10 to 30 frames / sec). In this example of the fire detection system 100, multiple cameras 5 are installed at approximately constant intervals in the direction of travel of the vehicle 110 so that the conditions inside the tunnel can be captured. Camera 5 is assigned to each section of the tunnel that has been divided into sections. Figure 2 also shows an example in which multiple cameras 5 are installed on one side (the right-hand shoulder) in the width direction (vehicle width direction) of the tunnel.

[0027] Furthermore, the multiple cameras 5 do not need to be installed in the configuration shown in Figure 2, as long as they can capture images of the conditions inside the tunnel. In other words, the installation positions of the multiple cameras 5 are not limited to the configuration shown in Figure 2.

[0028] Furthermore, the infrared camera 2 (with the pan / tilt head 3) and camera 5 may be mounted, for example, on the ceiling of the tunnel, or on a support installed on the roadside (or shoulder) inside the tunnel, or by any other method. In other words, the structure for installing the infrared camera 2 (with the pan / tilt head 3) and camera 5 inside the tunnel can be any structure.

[0029] The fire detection device 1 processes the frame images captured by each camera 5 and detects vehicles 110 that have stopped (parked) in the tunnel. When the fire detection device 1 detects a vehicle 110 that has stopped in the tunnel, it determines which infrared camera 2 will capture the vehicle 110's stopping position. The fire detection device 1 determines which infrared camera 2 will capture the vehicle 110's stopping position, and which infrared camera 2 will be the second closest to the vehicle 110's stopping position, in the direction of the vehicle 110's travel within the tunnel. In other words, the fire detection device 1 determines which infrared cameras will capture the vehicle 110's stopping position, and which are consecutive in the direction of the vehicle 110's travel, to be used. As is clear from the above explanation, in the example shown in Figure 2, the two infrared cameras 2 that capture the vehicle 110's stopping position are installed at different locations in the width direction of the tunnel.

[0030] For example, in the example shown in Figure 2, if the stopping position of vehicle 110 in the direction of travel within the tunnel is between infrared camera 2-1 and infrared camera 2-2, then infrared camera 2-1 and infrared camera 2-2 are determined to be infrared cameras 2 that image the stopping position of vehicle 110. Also, if the stopping position of vehicle 110 in the direction of travel within the tunnel is between infrared camera 2-2 and infrared camera 2-3, then infrared camera 2-2 and infrared camera 2-3 are determined to be infrared cameras 2 that image the stopping position of vehicle 110.

[0031] The fire detection device 1 adjusts the imaging area of ​​each infrared camera 2 that images the stopping position of the vehicle 110 to match the stopping position of the vehicle 110. For each infrared camera 2, the fire detection device 1 determines the angle (angle in the pan direction and angle in the tilt direction) to adjust the imaging area to match the stopping position of the vehicle 110, based on the installation position of the infrared camera 2 and the stopping position of the vehicle 110. The fire detection device 1 also determines the imaging magnification for each infrared camera 2 based on the distance between the installation position of the infrared camera 2 and the stopping position of the vehicle 110. For each infrared camera 2, the fire detection device 1 outputs drive signals (pan signal and tilt signal) to the pan head 3 corresponding to the pan direction angle and tilt direction angle determined for that infrared camera 2. The fire detection device 1 also outputs a zoom signal to each infrared camera 2 corresponding to the imaging magnification determined for that infrared camera 2.

[0032] The fire detection device 1 processes a frame image of the vehicle 110's stopping position, captured by an infrared camera 2 whose angle is aligned with the vehicle 110's stopping position, to obtain the temperature distribution around the vehicle 110's stopping position and determine whether a fire has occurred.

[0033] Thus, in this example of the fire detection system 100, the infrared camera 2 is used as a PTZ camera, which allows for a reduction in the number of infrared cameras assigned to the tunnel without compromising the accuracy of determining whether a fire is occurring in the target area, which is the tunnel.

[0034] Furthermore, in this example, the stopping position of the vehicle 110 is captured by two infrared cameras 2 with different angles. Therefore, even if the stopping position of the vehicle 110 cannot be captured by one infrared camera 2 due to occlusion caused by the vehicle 110 traveling inside the tunnel, the stopping position of the vehicle 110 can still be captured by the other infrared camera 2. Consequently, the delay time from the occurrence of a fire to its detection can be reduced.

[0035] <2. Example Configuration> Figure 3 shows the main components of the fire detection device in this example. The fire detection device 1 comprises a control unit 11, a PTZ signal output unit 12, an infrared image input unit 13, a visible light image input unit 14, an output unit 15, and a communication unit 16.

[0036] In this example, although not specifically shown in the diagram, the fire detection device 1 stores the installation position for each of the infrared cameras 2 (each of the pan / tilt heads 3), as well as the installation position for each of the cameras 5. Furthermore, the fire detection device 1 in this example stores camera parameters for each of the cameras 5, which convert the position on the frame image captured by that camera 5 into a position in real space.

[0037] The control unit 11 controls the operation of each part of the fire detection device 1 main body. The control unit 11 also includes a stopped vehicle detection unit 11a, a PTZ signal generation unit 11b, and a fire determination unit 11c. The stopped vehicle detection unit 11a, the PTZ signal generation unit 11b, and the fire determination unit 11c of the control unit 11 will be described later.

[0038] The PTZ signal output unit 12 outputs a pan signal and a tilt signal to the pan / tilt head 3 on which each infrared camera 2 is mounted, indicating the angle (imaging direction) of the infrared camera 2. The PTZ signal output unit 12 also outputs a zoom signal to each infrared camera 2, indicating the imaging magnification.

[0039] Each pan / tilt head 3 changes the angle of the attached infrared camera 2 in response to the pan and tilt signals from the aircraft. In addition, each infrared camera 2 changes the image magnification in response to the zoom signal from the aircraft.

[0040] The infrared image input unit 13 receives a frame image captured by each infrared camera 2. The frame image captured by the infrared camera 2 is a temperature distribution image of the imaging area.

[0041] The visible light image input unit 14 receives a frame image captured by each camera 5. The frame image captured by camera 5 is a visible light image of the imaging area.

[0042] The output unit 15 outputs guidance messages to the guidance board 7 that are to be displayed on the guidance board 7 for the driver.

[0043] The communication unit 16 is connected to the control center via a network and communicates data with the control center. For example, the communication unit 16 notifies the control center whether a fire has occurred in the tunnel.

[0044] Next, the stopped vehicle detection unit 11a, the PTZ signal generation unit 11b, and the fire detection unit 11c, which are part of the control unit 11, will be described.

[0045] The stationary vehicle detection unit 11a processes the frame images captured by each camera 5 and detects stationary vehicles 110. For example, the stationary vehicle detection unit 11a detects a vehicle as stationary if the same vehicle 110 is captured for a predetermined number of frames (3 to 10 frames) at approximately the same position on the frame images captured by each camera 5. Alternatively, the stationary vehicle detection unit 11a may process the frame images captured by each camera 5, track the vehicles 110 captured in the frame images, and detect a vehicle 110 whose moving speed is below a set speed (for example, less than a few km / h (3 to 5 km / h)) for a set time (for example, several seconds (approximately 1 to 3 seconds)) as a stationary vehicle 110. The stationary vehicle detection unit 11a may also detect stationary vehicles 110 by methods other than those exemplified above.

[0046] When the stationary vehicle detection unit 11a detects a stationary vehicle 110, it calculates the stationary vehicle 110's real-space stopping position. For example, the stationary vehicle detection unit 11a detects (or estimates) the contact position between the tire and the road surface on the frame image in which the stationary vehicle 110 is captured. The stationary vehicle detection unit 11a uses camera parameters stored for the camera 5 that captured this frame image to convert the detected contact position between the tire and the road surface on the frame image into a real-space position. The stationary vehicle detection unit 11a may also calculate the real-space stopping position of the stationary vehicle 110 using a method other than the one exemplified above.

[0047] This stopped vehicle detection unit 11a corresponds to the detection unit in this invention.

[0048] The PTZ signal generation unit 11b determines the infrared camera 2 to image the stopping position of the stopped vehicle 110 based on the real-space stopping position of the stopped vehicle 110 detected by the stopped vehicle detection unit 11a. The PTZ signal generation unit 11b determines the infrared camera 2 closest to the stopping position of the stopped vehicle 110 and the infrared camera 2 second closest to the stopping position in the direction of travel of the vehicle 110 in the tunnel to be the infrared cameras 2 that will image the stopping position of the stopped vehicle 110. In other words, the fire detection device 1 determines the two infrared cameras 2 that are continuous in the direction of travel of the stopped vehicle 110, with the stopping position of the stopped vehicle 110 in between, to be the infrared cameras 2 that will image the stopping position of the stopped vehicle 110.

[0049] In this example, two infrared cameras 2 are used to image the stopping position of the stopped vehicle 110. However, the number of infrared cameras 2 used to image the stopping position of the stopped vehicle 110 may be one or three or more. In this case, the PTZ signal generation unit 11b should determine the set number of infrared cameras 2 to image the stopping position of the stopped vehicle 110, in order of proximity.

[0050] The PTZ signal generation unit 11b calculates the pan angle and tilt angle for each infrared camera 2 that images the stopping position of the stopped vehicle 110, so as to align the imaging area of ​​the infrared camera 2 with the stopping position of the vehicle 110. For example, the PTZ signal generation unit 11b calculates a straight line connecting the installation position of the infrared camera 2 and the stopping position of the stopped vehicle 110, and calculates the pan angle and tilt angle to align the angle (imaging direction) of the infrared camera with this straight line.

[0051] Furthermore, the PTZ signal generation unit 11b calculates the imaging magnification of the infrared camera 2 based on the distance between the installation position of the infrared camera 2 and the stopping position of the stopped vehicle 110. For example, the PTZ signal generation unit 11b calculates the imaging magnification of the infrared camera 2 so that the size of the imaging area of ​​the infrared camera 2 becomes the set imaging size.

[0052] The PTZ signal generation unit 11b outputs drive signals (pan signal and tilt signal) to the pan head 3 according to the calculated pan angle and tilt angle. The PTZ signal generation unit 11b also outputs a zoom signal to the infrared camera 2 according to the calculated imaging magnification. The pan signal, tilt signal, and zoom signal are output by the PTZ signal output unit 12.

[0053] This PTZ signal generation unit 11b corresponds to the adjustment unit in this invention.

[0054] The fire detection unit 11c processes the frame image captured by each infrared camera 2 that captured the stopping position of the stopped vehicle 110, obtains the temperature distribution at this stopping position, and determines whether a fire has occurred. If the fire detection unit 11c determines that a fire has occurred based on the temperature distribution obtained by processing the frame image captured by any of the infrared cameras 2, it determines that a fire has occurred near the stopping position of the stopped vehicle 110. In other words, if the fire detection unit 11c determines that no fire has occurred based on the temperature distribution obtained by processing the frame image captured by any of the infrared cameras 2, it determines that no fire has occurred at the stopping position of the vehicle 110.

[0055] This fire detection unit 11c corresponds to the first detection unit in this invention.

[0056] The control unit 11 of the fire detection device 1 is composed of a hardware CPU, memory, and other electronic circuits. When the hardware CPU executes the fire detection program according to this invention, it operates as a stopped vehicle detection unit 11a, a PTZ signal generation unit 11b, and a fire determination unit 11c. The memory also has an area for deploying the fire detection program according to this invention and an area for temporarily storing data generated when the fire detection program is executed. The control unit 11 may be an LSI integrating the hardware CPU, memory, etc. Furthermore, the hardware CPU is a computer that executes the fire detection method according to this invention.

[0057] <3. Example of operation> Figure 4 is a flowchart illustrating the operation of the fire detection device in this example. In this example, when no fire is detected in the tunnel, the fire detection system 100 receives video images of the tunnel taken by each camera 5 as input to the fire detection device 1. On the other hand, each infrared camera 2 does not take images of the tunnel when no fire is detected. The infrared cameras 2 take images of the tunnel in response to imaging instructions from the fire detection device 1.

[0058] Furthermore, the infrared camera 2, like camera 5, may be configured to output frame images of the video footage taken inside the tunnel to the fire detection device 1, even when no fire is detected inside the tunnel.

[0059] The fire detection device 1 processes the frame images of the moving images of the tunnel captured by each camera 5 and repeats the stopped vehicle detection process until it detects a stopped vehicle 110 in the tunnel (s1, s2). The stopped vehicle detection unit 11a executes the processes in s1 and s2. The stopped vehicle detection unit 11a processes the frame images of the moving images captured by each camera 5 and detects the stopped vehicle 110. The visible light image input unit 14 receives the moving images captured by each connected camera 5 as input.

[0060] The stationary vehicle detection unit 11a, for example, as described above, detects a vehicle 110 as a stationary vehicle 110 if it has been continuously captured for a predetermined number of frames (3 to 10 frames) at approximately the same position on the frame image captured by each camera 5. Alternatively, the stationary vehicle detection unit 11a may process the frame image captured by each camera 5 and detect a vehicle 110 as a stationary vehicle 110 if its moving speed is below a set speed (for example, less than a few km / h (3 to 5 km / h)) for a set time (for example, several seconds (approximately 1 to 3 seconds)).

[0061] When the stationary vehicle detection unit 11a detects a stationary vehicle 110, it calculates the stationary position of the stationary vehicle 110 in real space (s3). The stationary vehicle detection unit 11a detects (or estimates) the contact position between the tire and the road surface on the frame image in which the stationary vehicle 110 is captured, for example, as described above. The stationary vehicle detection unit 11a uses camera parameters stored for the camera 5 that captured this frame image to convert the detected contact position between the tire and the road surface on the frame image into a position in real space.

[0062] The PTZ signal generation unit 11b determines an infrared camera 2 to image the stopping position of the stopped vehicle 110 based on the real-space stopping position of the stopped vehicle 110 detected by the stopped vehicle detection unit 11a (s4). For example, as described above, the PTZ signal generation unit 11b determines the infrared camera 2 closest to the stopping position of the stopped vehicle 110 and the infrared camera 2 second closest to this stopping position as the infrared cameras 2 to image the stopping position of the stopped vehicle 110 in the direction of travel of the vehicle 110 in the tunnel.

[0063] The PTZ signal generation unit 11b calculates, for each infrared camera 2 that images the stopping position of the stopped vehicle 110, the pan angle and tilt angle to align the imaging area of ​​the infrared camera 2 with the stopping position of the vehicle 110 (s5). For example, as described above, the PTZ signal generation unit 11b calculates a straight line connecting the installation position of the infrared camera 2 and the stopping position of the stopped vehicle 110, and calculates the pan angle and tilt angle to align the angle (imaging direction) of the infrared camera with this straight line.

[0064] Furthermore, the PTZ signal generation unit 11b calculates the imaging magnification of the infrared camera 2 based on the distance between the installation position of the infrared camera 2 and the stopping position of the stopped vehicle 110 (s6). The PTZ signal generation unit 11b calculates the imaging magnification of the infrared camera 2 so that the size of the imaging area of ​​the infrared camera 2 becomes the set imaging size, for example, as described above.

[0065] The PTZ signal generation unit 11b generates drive signals (pan signal and tilt signal) corresponding to the calculated pan angle and tilt angle and outputs them to the pan / tilt head 3 (s7). The PTZ signal generation unit 11b also generates a zoom signal corresponding to the calculated imaging magnification and outputs it to the infrared camera 2 (s8). The pan signal, tilt signal, and zoom signal are output by the PTZ signal output unit 12.

[0066] The PTZ signal generation unit 11b outputs an imaging instruction to the infrared camera 2, which has output pan, tilt, and zoom signals. The infrared camera 2 starts imaging in response to this instruction.

[0067] The fire determination unit 11c processes the frame images captured by the infrared camera 2 that images the stopping position of the stopped vehicle 110 determined in s4, and determines whether a fire has occurred around this stopping position (s9). In s9, the fire determination unit 11c processes the frame images of the area around the stopping position of the stopped vehicle 110 captured by each infrared camera 2 that images the stopping position of the stopped vehicle 110 determined in s4, and obtains the temperature distribution around this stopping position. If, for example, the size of the area exceeding the set temperature exceeds the set range in the obtained temperature distribution, the fire determination unit 11c determines that a fire has occurred around the stopping position of the stopped vehicle 110.

[0068] Furthermore, the fire determination unit 11c processes the frame image captured by one of the infrared cameras 2 that image the stopping position of the stopped vehicle 110 determined in s4, and even if it determines that no fire has occurred, if it processes the frame image captured by the other infrared camera 2 and determines that a fire has occurred, it determines that a fire has occurred around the stopping position of the stopped vehicle 110. In other words, the fire determination unit 11c processes the frame image captured by at least one of the infrared cameras 2 that image the stopping position of the stopped vehicle 110 determined in s4, and determines from the temperature distribution around the stopping position of the stopped vehicle 110 that it has obtained by processing the frame image captured by that infrared camera 2, it determines that a fire has occurred around this stopping position. In other words, the fire determination unit 11c processes the frame image captured by all of the infrared cameras 2 that image the stopping position of the stopped vehicle 110 determined in s4, and determines from the temperature distribution around the stopping position of the stopped vehicle 110 that it has obtained by processing the frame image captured by that infrared camera 2 that it has obtained by processing the frame image, it determines that no fire has occurred, and determines that a fire has not occurred around this stopping position.

[0069] If the fire detection device 1 determines in the fire determination process in s9 that no fire has occurred around the stopping position of the stopped vehicle 110 (s10), it returns to s1. On the other hand, if the fire detection device 1 determines in the fire determination process in s9 that a fire has occurred around the stopping position of the stopped vehicle 110 (s10), it outputs a guidance message to the guidance board 7 to be displayed on the guidance board 7 (s11). The fire detection device 1 also notifies the control center of the occurrence of the fire (s12) and returns to s1.

[0070] In this example, the fire detection system 100 detects a stopped vehicle 110 in a tunnel and uses the infrared camera 2 to image the stopping position of the vehicle 110. If a fire occurs, the vehicle 110 that is on fire, and any vehicles 110 approaching the location of the fire, will stop. Therefore, in this example, the fire detection system 100 can use the infrared camera 2 to image areas where a fire is likely to occur and detect whether a fire is occurring around the stopping position of the stopped vehicle 110.

[0071] Furthermore, since infrared camera 2 is used as a PTZ camera, reducing the number of infrared cameras 2 will not decrease the accuracy of determining whether or not a fire has occurred. In other words, the number of infrared cameras 2 can be reduced without decreasing the accuracy of determining whether or not a fire has occurred.

[0072] Furthermore, in this example, since the angles of the multiple infrared cameras 2 that image the area around the stopped position of the stopped vehicle 110 are different, even if one of the infrared cameras 2 is unable to image the stopped position due to occlusion, the stopped position can still be imaged by another infrared camera 2. Therefore, the increase in the time required to detect a fire after it has occurred is also suppressed. In addition, in this example, there is no need to determine whether the frame image captured by the infrared camera 2 is a frame image that captures the stopped position of the stopped vehicle 110 or a frame image that does not capture the stopped position of the stopped vehicle 110 due to occlusion.

[0073] Furthermore, in this example, when no stationary vehicle 110 is detected, the frame images captured by the infrared camera 2 are not processed, thus reducing the processing load on the fire detection device 1.

[0074] <4. Variation> • Variation 1 Figure 5 is a block diagram showing the configuration of the main parts of the fire detection device of this modified example 1. The fire detection device 1A of this modified example 1 can also be applied to the fire detection system shown in Figure 1. The fire detection device 1A of this modified example 1 differs from the fire detection device 1 of the above example in that the control unit 11A is additionally equipped with a congestion determination unit 11d.

[0075] The congestion determination unit 11d determines whether the stopped vehicle 110 detected by the stopped vehicle detection unit 11a is stopped due to congestion. For example, the congestion determination unit 11d determines that the stopped vehicle 110 detected by the stopped vehicle detection unit 11a is stopped due to congestion if the average speed of the vehicles 110 inside the tunnel is less than several tens of percent of the speed limit inside the tunnel (for example, 10% to less than 40%) for a congestion determination period (for example, 10 to 20 seconds) or longer. Also, the congestion determination unit 11d determines that the stopped vehicle 110 detected by the stopped vehicle detection unit 11a is stopped due to congestion if the occupancy rate of the vehicles 110 inside the tunnel is 60% or more of the set occupancy rate (for example, 60% to 80%), and the speed of the group of vehicles inside the tunnel is less than several tens of percent of the speed limit inside the tunnel (for example, 10% to less than 40%) for a congestion determination period (for example, 10 to 20 seconds) or longer.

[0076] This congestion determination unit 11d corresponds to the second determination unit in this invention.

[0077] Note that, apart from the traffic congestion determination unit 11d, the configuration of the fire detection device 1A in this modified example 1 is the same as in the example above, so we will omit the explanation here.

[0078] Figure 6 is a flowchart showing the operation of the fire detection device in this modified example 1. In Figure 6, the same steps as those shown in Figure 4 are given the same step numbers.

[0079] In this modified example 1, the fire detection device 1A has a stopped vehicle detection unit 11a that performs the processes described in s1 and s2 above and detects a stopped vehicle 110. The traffic congestion determination unit 11d then determines whether the stopped vehicle 110 stopped due to traffic congestion (s21). If the stopped vehicle detection unit 11a determines that the stopped vehicle 110 detected this time stopped due to traffic congestion, it returns to s1. If the stopped vehicle detection unit 11a determines that the stopped vehicle 110 detected this time did not stop due to traffic congestion, it performs the processes described in s3 to s12 above.

[0080] Thus, in this modified example 1, the fire detection device 1A does not perform the processing described in s3 to s12 if the stopped vehicle 110 detected by the stopped vehicle detection unit 11a is due to traffic congestion and not a fire. Therefore, this modified example 1 achieves the same effect as the example described above, while further reducing the processing load.

[0081] • Variation 2 In the example above, the detection of the stationary vehicle 110 was performed using frame images from a moving image captured by the camera 5. However, the stationary vehicle 110 may also be detected using sensors such as radio radar (millimeter-wave radar) or laser radar.

[0082] Furthermore, while the above example described a system for detecting a fire inside a tunnel formed on a road where a vehicle 110 travels, the system is not limited to inside a tunnel; it could also be on a road outside the tunnel. Also, although the explanation used a vehicle 110 as the mobile entity, the mobile entity could be an autonomous mobile robot. In this case, the area to be detected for fire could be, for example, an indoor area such as a warehouse or factory where the autonomous mobile robot transports goods, or it could be an outdoor area.

[0083] Furthermore, the mobile device may be other than the vehicles and autonomous robots exemplified here. The target area for fire detection can be determined according to the mobile device.

[0084] It should be noted that this invention is not limited to the embodiments described above, and the components can be modified and implemented in practice without departing from the gist of the invention. Furthermore, various inventions can be formed by appropriately combining the multiple components disclosed in the embodiments. For example, some components may be deleted from all the components shown in the embodiments. Furthermore, components from different embodiments may be appropriately combined. In addition, the order of each step in the flowcharts shown in the descriptions of all the examples above is merely an example, and may be rearranged as appropriate to the extent possible.

[0085] Furthermore, the correspondence between the configuration of this invention and the configuration of the embodiment described above can be described as follows. <Note> A detection unit (11a) that detects a moving object (110) that has stopped within the target area, An adjustment unit (11b) changes the imaging area of ​​the infrared camera (5) to the stopping position of the moving object (110) which has been detected as having stopped by the detection unit (11a), A first determination unit (11c) processes frame images captured by the infrared camera (5) and determines whether a fire has occurred at the stopping position, A fire detection device (1) equipped with the following. [Explanation of Symbols]

[0086] 1, 1A…Fire detection device 2(2-1~2-n)...Infrared camera 3(3-1~3-n)...Trip head 5(5-1~5-m)...Visible light camera 11, 11A… Control Unit 11a...Stopped vehicle detection unit 11b...PTZ signal generation section 11c…Fire judgment department 11d...Traffic congestion detection unit 12...PTZ signal output section 13…Infrared image input section 14…Visible light image input section 15…Output section 16… Communications Department 100... Fire detection system 110... Vehicle

Claims

1. A detection unit that detects a moving object that has stopped within the target area, An adjustment unit that changes the imaging area of ​​the infrared camera to the stopping position of the moving object detected by the detection unit, A first determination unit processes frame images captured by the infrared camera and determines whether a fire has occurred at the stopping position, A fire detection device equipped with the following features.

2. The infrared camera is mounted on a pan / tilt head whose angle can be changed in two orthogonal axes. The adjustment unit generates a drive signal to the pan / tilt head that changes the imaging area of ​​the infrared camera to the stopping position of the moving body detected by the detection unit. The fire detection device according to claim 1.

3. The adjustment unit generates a zoom signal that changes the imaging magnification of the infrared camera according to the stopping position of the moving object detected by the detection unit. The fire detection device according to claim 2.

4. The aforementioned moving object is a vehicle, The system includes a second determination unit that determines whether the stop of a moving object detected by the detection unit is due to traffic congestion. If the second determination unit determines that the vehicle is stopped due to traffic congestion, the first determination unit does not determine whether a fire has occurred at the stopping position. A fire detection device according to any one of claims 1 to 3.

5. The detection unit processes the frame image of the visible light camera that captured the target area and detects a moving object that has stopped within the target area. A fire detection device according to any one of claims 1 to 3.

6. The detection unit processes the frame images of the visible light camera that captured each of the multiple division areas that make up the target area, and detects a moving object that has stopped within the target area. A fire detection device according to any one of claims 1 to 3.

7. The aforementioned moving object is a vehicle, The aforementioned target area is inside a tunnel. A fire detection device according to any one of claims 1 to 3.

8. A detection step to detect a moving object that has stopped within the target area, An adjustment step to change the imaging area of ​​the infrared camera to the stopping position of the moving object that was detected to have stopped in the detection step, A determination step involves processing the frame image captured by the infrared camera and determining whether a fire has occurred at the stopping position. A fire detection method performed by a computer.

9. A detection step to detect a moving object that has stopped within the target area, An adjustment step to change the imaging area of ​​the infrared camera to the stopping position of the moving object that was detected to have stopped in the detection step, A determination step involves processing the frame image captured by the infrared camera and determining whether a fire has occurred at the stopping position. A fire detection program that causes a computer to execute.

Images