Human detection device, human detection system and human detection method

Abstract

Provided are a person detection device and the like which are capable of accurately detecting a person present in a restricted area. A person detection device (2) is provided with: a three-dimensional model generation means (21) which generates a three-dimensional model of a restricted area located in an area where water is discharged from a dam and in the surrounding area, on the basis of reflected light resulting from illumination of the restricted area with laser light; and a person detection means (22) which detects a person in the restricted area using the three-dimensional model.

Description

[Technical field]

[0001] The present disclosure relates to a human detection device and the like. [Background technology]

[0002] Patent Document 1 discloses a technology that uses LiDAR (Light Detection and Ranging) to monitor other objects (including people and things) approaching a specific object in an area to be monitored. That is, in the technology described in Patent Document 1, LiDAR is used to generate distance images corresponding to the area in time series. The generated multiple distance images are used to detect the movement of each object in the area. Based on the results of such detection, other approaching objects are detected as described above (see the abstract of Patent Document 1, paragraphs

[0008] to

[0010] , paragraph

[0043] , etc.).

[0003] As a related technique, the technique described in Patent Document 2 is also known. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] JP 2019-201268 A [Patent Document 2] JP 2019-168985 A Summary of the Invention [Problem to be solved by the invention]

[0005] As described above, the technology described in Patent Document 1 detects the movement of an object. In other words, the technology described in Patent Document 1 detects a moving object without distinguishing whether it is a person or an object.

[0006] For example, consider the case of checking the safety of people when water is released from a dam. In such a case, it is desirable to detect all people present within the area where the dam water is released (hereinafter sometimes referred to as the "discharge area"). It is also desirable to detect all people present within dangerous areas (hereinafter sometimes referred to as the "danger areas") that are dangerous if people are present when water is released from the surrounding areas of the discharge area. Hereinafter, the discharge area and the danger area may be collectively referred to as the "no entry areas."

[0007] That is, it is preferable to accurately detect people who are present in a no-entry area, regardless of whether the people are stationary or moving. However, by using the technology described in Patent Document 1, moving objects are detected without distinguishing whether they are people or objects. This causes a problem in that it is difficult to accurately detect people who are present in a no-entry area.

[0008] The present disclosure has been made to solve the above-mentioned problems, and aims to provide a human detection device etc. that can accurately detect a person present in a no-entry area. [Means for solving the problem]

[0009] One form of a human detection device according to the present disclosure comprises a three-dimensional model generation means for generating a three-dimensional model of a no-entry area based on reflected light corresponding to laser light irradiated onto the no-entry area in an area where water from a dam is released and the surrounding areas, and a human detection means for detecting people in the no-entry area using the three-dimensional model.

[0010] One form of the human detection system according to the present disclosure comprises a three-dimensional model generation means for generating a three-dimensional model of a no-entry area based on reflected light corresponding to laser light irradiated onto the no-entry area in an area where water is released from a dam and the surrounding areas, and a human detection means for detecting people in the no-entry area using the three-dimensional model.

[0011] One form of the human detection method disclosed herein is one in which a three-dimensional model generation means generates a three-dimensional model of a no-entry area in an area where water is released from a dam and the surrounding areas based on reflected light corresponding to laser light irradiated onto the no-entry area, and a human detection means detects people in the no-entry area using the three-dimensional model. Effect of the Invention

[0012] According to the present disclosure, it is possible to accurately detect a person present in a no-entry area. [Brief description of the drawings]

[0013] [Figure 1] FIG. 1 is a block diagram showing a main part of a human detection system according to a first embodiment. [Diagram 2] FIG. 2 is a block diagram showing a main part of the optical sensing device in the human detection system according to the first embodiment. [Diagram 3] FIG. 3 is a block diagram showing a main part of the human detection device according to the first embodiment. [Figure 4] FIG. 4 is a block diagram showing a hardware configuration of a main part of the human detection device according to the first embodiment. [Diagram 5] FIG. 5 is a block diagram showing another hardware configuration of the main parts of the human detection device according to the first embodiment. [Figure 6] FIG. 6 is a block diagram showing another hardware configuration of the main parts of the human detection device according to the first embodiment. [Figure 7] FIG. 7 is a flowchart showing the operation of the human detection device according to the first embodiment. [Figure 8] FIG. 8 is an explanatory diagram showing specific examples of entry-prohibited areas and the like. [Figure 9] FIG. 9 is a block diagram showing a main part of the other person detection system according to the first embodiment. [Figure 10] FIG. 10 is a block diagram showing a main part of the other person detection device according to the first embodiment. [Figure 11] FIG. 11 is a block diagram showing a main part of a human detection system according to the second embodiment. [Figure 12] FIG. 12 is a block diagram showing a main part of a human detection device according to the second embodiment. [Figure 13] FIG. 13 is a flowchart showing the operation of the human detection device according to the second embodiment. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0014] Hereinafter, embodiments of the present disclosure will be described in detail with reference to the accompanying drawings.

[0015] [First embodiment] Fig. 1 is a block diagram showing a main part of a human detection system according to a first embodiment. Fig. 2 is a block diagram showing a main part of an optical sensing device in the human detection system according to the first embodiment. Fig. 3 is a block diagram showing a main part of a human detection device according to the first embodiment. The human detection system according to the first embodiment will be described with reference to Figs. 1 to 3.

[0016] As shown in Fig. 1, the human detection system 100 includes an optical sensing device 1, a human detection device 2, and an output device 3. As shown in Fig. 2, the optical sensing device 1 includes a light emitting unit 11 and a light receiving unit 12. As shown in Fig. 3, the human detection device 2 includes a three-dimensional model generating unit 21, a human detection unit 22, and an output control unit 23.

[0017] The optical sensing device 1 is installed inside or around the prohibited area. Here, as described above, the prohibited area may include a discharge area and may also include a prohibited area. A discharge area is an area where water is discharged from a dam. A danger area is an area around a discharge area where the presence of people is dangerous. Specific examples of prohibited areas will be described later with reference to FIG. 8.

[0018] The light emitting unit 11 uses, for example, a laser light source. The light emitting unit 11 emits pulsed laser light. Here, in the optical sensing device 1, the emission direction of the laser light by the light emitting unit 11 is variable. As a result, the light emitting unit 11 sequentially emits the laser light in a plurality of directions. As a result, the laser light is irradiated so as to scan the entry prohibited area.

[0019] The irradiated laser light is reflected as scattered light by objects (including people) in the prohibited area. Hereinafter, the reflected light may be referred to as "reflected light." The light receiving unit 12 receives the reflected light. The light receiving unit 12 uses, for example, a light receiving element.

[0020] The three-dimensional model generating unit 21 generates a three-dimensional model of the no-entry area based on the laser light emitted by the light emitting unit 11 and the reflected light received by the light receiving unit 12. More specifically, the three-dimensional model generating unit 21 generates a three-dimensional point cloud model corresponding to the shape of each object in the no-entry area. The principle of ToF (Time of Flight) LiDAR is used to generate such a three-dimensional model.

[0021] That is, the three-dimensional model generating unit 21 acquires information indicating the timing at which the laser light was emitted in each direction and information indicating the timing at which the corresponding reflected light was received. This information is acquired, for example, from the light sensing device 1. Using this information, the three-dimensional model generating unit 21 calculates a one-way propagation distance corresponding to the round-trip propagation time for the laser light emitted in each direction and the corresponding reflected light.

[0022] The three-dimensional model generating unit 21 calculates coordinate values ​​indicating the positions of the points where the laser light emitted in each direction is reflected based on the calculated one-way propagation distance. These coordinate values ​​are coordinate values ​​in a virtual three-dimensional coordinate space. By arranging points corresponding to the individual coordinate values ​​in the three-dimensional coordinate space, a three-dimensional point cloud model corresponding to the shape of each object in the entry-prohibited area is generated. In this manner, a three-dimensional model is generated.

[0023] In addition, various known techniques can be used to generate a three-dimensional model based on the principles of ToF LiDAR, and detailed descriptions of these techniques will be omitted.

[0024] The human detection unit 22 detects people present in the entry-prohibited area by using the three-dimensional model generated by the three-dimensional model generation unit 21. More specifically, the human detection unit 22 detects people present in the entry-prohibited area by determining whether or not each object present in the entry-prohibited area is a person by using the generated three-dimensional model.

[0025] That is, as described above, the three-dimensional model generated by the three-dimensional model generation unit 21 includes a three-dimensional point cloud model corresponding to the shape of each object present in the no-entry area. Here, the human detection unit 22 pre-stores information (hereinafter referred to as "reference shape information") indicating a general human shape (hereinafter referred to as "reference shape"). Alternatively, the human detection unit 22 acquires the reference shape information. The reference shape information may include information indicating a plurality of reference shapes corresponding to different postures. Furthermore, the reference shape information may include information indicating a plurality of reference shapes corresponding to different gaze directions. Furthermore, the reference shape information may include information indicating a plurality of reference shapes corresponding to different physiques.

[0026] The human detection unit 22 uses the generated three-dimensional model and the stored or acquired reference shape information to compare the shape of each object present in the intrusion-prohibited area with each reference shape. The human detection unit 22 determines whether each object is a person or not by so-called "pattern matching". Specifically, for example, the human detection unit 22 calculates the difference between the shape of each object and each reference shape. If the calculated difference is within a predetermined range for at least one reference shape, the human detection unit 22 determines that the object is a person. If not, the human detection unit 22 determines that the object is not a person. In this way, a person present in the intrusion-prohibited area is detected.

[0027] The output control unit 23 executes control to output information based on the result of detection by the human detection unit 22 (hereinafter referred to as "detection result information"). The detection result information includes, for example, information indicating the presence or absence of a person in the intrusion prohibited area, and information indicating the position of the person in the intrusion prohibited area. In other words, the detection result information may include information regarding the person detected by the human detection unit 22.

[0028] An output device 3 is used to output the detection result information. The output device 3 includes, for example, at least one of a display device, an audio output device, and a communication device. The display device is, for example, a device using a display. The audio output device is, for example, a device using a speaker. The communication device is, for example, a device using a dedicated transmitter and receiver.

[0029] Specifically, for example, the output control unit 23 executes control to display an image corresponding to the detection result information. A display device of the output device 3 is used to display such an image. Here, such an image may include a three-dimensional model generated by the three-dimensional model generation unit 21. Alternatively, for example, the output control unit 23 executes control to output a sound corresponding to the detection result information. An audio output device of the output device 3 is used to output such a sound. Alternatively, for example, the output control unit 23 executes control to transmit a signal corresponding to the detection result information. A communication device of the output device 3 is used to transmit such a signal.

[0030] Here, such a signal may be transmitted to the dam maintenance management system 200 (see FIG. 1). In other words, the detection result information may be output to the dam maintenance management system 200. The dam maintenance management system 200 is a system that performs maintenance management of a dam. As shown in FIG. 1, the dam maintenance management system 200 is provided outside the human detection system 100. Hereinafter, the dam maintenance management system 200 may be referred to as the "external dam maintenance system." The output detection result information is used by the dam maintenance management system 200 for the maintenance management of the dam. Specifically, for example, the output detection result information is used by the dam maintenance management system 200 to issue an alarm to a person present in a no-entry area.

[0031] That is, when the detection result information indicates that a person is present in the no-entry area, the dam maintenance management system 200 issues an alarm to the person using speakers installed around the no-entry area. Alternatively, in this case, the dam maintenance management system 200 instructs another person (for example, the operator of the dam maintenance management system 200) to issue the alarm. This makes it possible to urge people present in the no-entry area to evacuate.

[0032] The information output by the output device 3 to the dam maintenance management system 200 is not limited to the detection result information (i.e., information about the detected person). For example, the output information may include, in addition to the detection result information, information about no-entry areas and information used for control of issuing the alarm.

[0033] In this manner, the main part of the human detection system 100 is configured.

[0034] Hereinafter, the light emitting unit 11 may be referred to as "light emitting means". Furthermore, the light receiving unit 12 may be referred to as "light receiving means". Furthermore, the three-dimensional model generating unit 21 may be referred to as "three-dimensional model generating means". Furthermore, the human detection unit 22 may be referred to as "human detection means". Furthermore, the output control unit 23 may be referred to as "output control means".

[0035] Next, the hardware configuration of the main parts of the human detection device 2 will be described with reference to FIGS.

[0036] As shown in each of Fig. 4 to Fig. 6, the human detection device 2 uses a computer 31. The computer 31 may be provided integrally with the optical sensing device 1. Alternatively, the computer 31 may be provided in another location (for example, in a cloud network). Alternatively, some elements of the computer 31 may be provided integrally with the optical sensing device 1, and the remaining elements of the computer 31 may be provided in another location.

[0037] 4, the computer 31 includes a processor 41 and a memory 42. The memory 42 stores programs for causing the computer 31 to function as the three-dimensional model generation unit 21, the human detection unit 22, and the output control unit 23. The processor 41 reads out and executes the programs stored in the memory 42. As a result, a function F1 of the three-dimensional model generation unit 21, a function F2 of the human detection unit 22, and a function F3 of the output control unit 23 are realized.

[0038] 5, the computer 31 includes a processing circuit 43. The processing circuit 43 executes processes for causing the computer 31 to function as the three-dimensional model generating unit 21, the human detection unit 22, and the output control unit 23. This realizes functions F1 to F3.

[0039] 6, the computer 31 includes a processor 41, a memory 42, and a processing circuit 43. In this case, some of the functions F1 to F3 are realized by the processor 41 and the memory 42, and the remaining functions of the functions F1 to F3 are realized by the processing circuit 43.

[0040] The processor 41 is composed of one or more processors. Each processor may be, for example, a central processing unit (CPU), a graphics processing unit (GPU), a microprocessor, a microcontroller, or a digital signal processor (DSP).

[0041] The memory 42 is composed of one or more memories, each of which may be, for example, a random access memory (RAM), a read only memory (ROM), a flash memory, an erasable programmable read only memory (EPROM), an electrically erasable programmable read only memory (EEPROM), a solid state drive, a hard disk drive, a flexible disk, a compact disk, a digital versatile disk (DVD), a Blu-ray disk, a magneto optical (MO) disk, or a mini disk.

[0042] The processing circuit 43 is composed of one or more processing circuits. Each processing circuit uses, for example, an ASIC (Application Specific Integrated Circuit), a PLD (Programmable Logic Device), an FPGA (Field Programmable Gate Array), an SoC (System on a Chip), or a system LSI (Large Scale Integration).

[0043] The processor 41 may include a dedicated processor corresponding to each of the functions F1 to F3. The memory 42 may include a dedicated memory corresponding to each of the functions F1 to F3. The processing circuit 43 may include a dedicated processing circuit corresponding to each of the functions F1 to F3.

[0044] Next, the operation of the human detection device 2 will be described with reference to the flowchart shown in FIG.

[0045] First, the three-dimensional model generating unit 21 generates a three-dimensional model of the entry-prohibited area (step ST1). ToF LiDAR technology is used to generate the three-dimensional model, as described above. Next, the human detection unit 22 detects a human being present in the entry-prohibited area (step ST2). The three-dimensional model generated in step ST1 is used for such detection, as described above. Next, the output control unit 23 executes control to output information based on the result of the detection in step ST2 (i.e., detection result information) (step ST3).

[0046] Next, a specific example of the entry-prohibited area will be described with reference to FIG.

[0047] In the figure, D indicates a dam. The dam D may be a dam for hydroelectric power generation. In the figure, R indicates a river located downstream of the dam D. In the figure, DD indicates an example of the discharge direction of the dam D into the river R. In the example shown in FIG. 8, the optical sensing device 1 is installed in a position near the outlet of the dam D (a so-called "spillway" or drain).

[0048] In the figure, A1 indicates the range downstream of the dam D. Here, as shown in FIG. 8, the shape of the range A1 is a sector. The center position of the sector corresponds to the installation position of the optical sensing device 1. The central angle of the sector is set to a predetermined value (e.g., approximately 180 degrees). The radius of the sector is set to a predetermined value (e.g., 300 meters). In other words, the range A1 corresponds to the range in which the optical sensing device 1 can irradiate laser light.

[0049] In the figure, A2 indicates a rectangular range inscribed in the sector of range A1. As shown in Fig. 8, range A2 includes a part of river R and also includes a part of the coastal area of ​​river R (more specifically, both banks).

[0050] 8, range A2 is set as a no-entry area. That is, in this case, the part of the river R included in range A2 is a discharge area. Also, the part of the coastal area of ​​the river R (more specifically, both banks) included in range A2 is a danger area.

[0051] In the figure, P1 indicates a person who is present in the no-entry area. As described above, the detection result information may be used by the dam maintenance management system 200 to notify the person P1 of an alarm. In the figure, S_1, S_2, and S_3 each indicate a speaker used to notify the person of the alarm. In the figure, P2_1 and P2_2 each indicate another person to whom the alarm is to be notified under the instruction of the dam maintenance management system 200.

[0052] Note that the intrusion prohibited area is not limited to the example shown in Fig. 8. For example, if the radius of the sector in range A1 is large (e.g., 500 meters or more) or if the length of river R is short, range A2 inscribed in range A1 may include the mouth of river R. As a result, range A2 may include the coastal area around the mouth of the river. In this case, the coastal area may be included in the danger area.

[0053] Next, the effects of using the human detection system 100 will be described.

[0054] As described above, by generating a three-dimensional model of the no-entry area, it is possible to determine whether each object present in the no-entry area is a person or not by pattern matching based on the shape of the object. Therefore, if each object is a person, the person can be detected based on the shape of the person at the time when the laser light is irradiated, regardless of whether the person is stationary or moving. In other words, the person can be detected regardless of whether the person present in the no-entry area is stationary or moving. In other words, the person can be accurately detected.

[0055] Next, a modification of the human detection system 100 will be described.

[0056] The light emitting unit 11 and the light receiving unit 12 may be provided in the human detection device 2 instead of in the optical sensing device 1. That is, the human detection device 2 may include the light emitting unit 11 and the light receiving unit 12. In this case, the optical sensing device 1 is not necessary.

[0057] Next, another modified example of the human detection system 100 will be described with reference to Fig. 9. Also, a modified example of the human detection device 2 will be described with reference to Fig. 10.

[0058] As shown in Fig. 9, the human detection system 100 may include a three-dimensional model generation unit 21 and a human detection unit 22. In other words, the main parts of the human detection system 100 may be configured by the three-dimensional model generation unit 21 and the human detection unit 22. In this case, the optical sensing device 1 may be provided outside the human detection system 100. Furthermore, the output control unit 23 may be provided outside the human detection system 100. Furthermore, the output device 3 may be provided outside the human detection system 100. Note that in the human detection system 100, each of the three-dimensional model generation unit 21 and the human detection unit 22 may be configured by an independent device.

[0059] 10 , the human detection device 2 may include a three-dimensional model generation unit 21 and a human detection unit 22. In other words, the main parts of the human detection device 2 may be configured by the three-dimensional model generation unit 21 and the human detection unit 22. In this case, the output control unit 23 may be provided outside the human detection device 2.

[0060] Even in these cases, the above-mentioned effects can be achieved. That is, the three-dimensional model generating unit 21 generates a three-dimensional model of the intrusion-prohibited area based on the reflected light corresponding to the laser light irradiated to the intrusion-prohibited area. The human detection unit 22 detects a person in the intrusion-prohibited area by using the three-dimensional model. Here, by using such a three-dimensional model, it is possible to determine whether each object present in the intrusion-prohibited area is a person or not by pattern matching based on the shape of the object. Therefore, when each object is a person, the person can be detected based on the shape of the person at the time when the laser light is irradiated, regardless of whether the person is stationary or moving. In other words, the person can be detected regardless of whether the person present in the intrusion-prohibited area is stationary or moving. That is, the person can be accurately detected.

[0061] [Second embodiment] Fig. 11 is a block diagram showing main parts of a human detection system according to a second embodiment. Fig. 12 is a block diagram showing main parts of a human detection device according to a second embodiment. The human detection system according to the second embodiment will be described with reference to Figs. 11 and 12. In Figs. 11 and 12, blocks similar to those shown in Figs. 1 to 3 are denoted by the same reference numerals and description thereof will be omitted.

[0062] 11, the human detection system 100a includes a light sensing device 1, a human detection device 2a, and an output device 3. As shown in Fig. 12, the human detection device 2a includes a three-dimensional model generation unit 21, a human detection unit 22a, an output control unit 23, and a movement detection unit 24.

[0063] The movement detection unit 24 detects the movement of each object present in the no-entry area based on the laser light emitted by the light emission unit 11 and the reflected light received by the light reception unit 12. The principle of Doppler LiDAR is used to detect such movement.

[0064] That is, as described in the first embodiment, the three-dimensional model generated by the three-dimensional model generating unit 21 includes a three-dimensional point cloud model corresponding to the shape of each object existing in the intrusion prohibited area. The motion detecting unit 24 acquires information indicating the frequency components contained in the corresponding laser light and the frequency components contained in the corresponding reflected light for each point in the three-dimensional point cloud model. These pieces of information are acquired, for example, from the light sensing device 1. The motion detecting unit 24 uses these pieces of information to calculate the Doppler shift amount in the corresponding reflected light for each object. The Doppler shift amount is based on the frequency of the laser light emitted by the light emitting unit 11. In other words, the Doppler shift amount is based on the difference between the frequency components contained in the corresponding laser light and the frequency components contained in the corresponding reflected light for each object.

[0065] The motion detection unit 24 detects the motion of each object present in the intrusion prohibited area based on the calculated Doppler shift amount. More specifically, the motion detection unit 24 detects the direction and speed of the motion. This also detects the presence or absence of the motion. Various known techniques can be used to detect the motion of an object based on the principle of Doppler LiDAR. Detailed explanations of these techniques will be omitted.

[0066] When an individual object that exists within the intrusion prohibited area is stationary, the human detection unit 22a judges whether or not the object is a human by a detection method similar to that used by the human detection unit 22. That is, the human detection unit 22a compares the shape of the object with a reference shape and judges whether or not the object is a human by pattern matching. On the other hand, when the object is moving, the human detection unit 22a judges whether or not the object is a human in the following manner.

[0067] That is, the human detection unit 22a stores information (hereinafter referred to as "reference range information") indicating the range (hereinafter referred to as "reference range") that a person can normally take with respect to the direction and speed of human movement in the intrusion-prohibited area. Alternatively, the human detection unit 22a acquires the reference range information. The reference range information may include information indicating a plurality of reference ranges corresponding to different positions in the intrusion-prohibited area.

[0068] Using the stored or acquired reference range information, the human detection unit 22a determines whether the movement of a moving object present in the intrusion prohibited area is within the reference range based on the result of detection by the movement detection unit 24. If the movement of the object is within the reference range, the human detection unit 22a determines that the object is a human. On the other hand, if the movement of the object is outside the reference range, the human detection unit 22a determines that the object is not a human. In this manner, a person present in the intrusion prohibited area is detected.

[0069] Alternatively, the human detection unit 22a compares the shape of a moving object present in the intrusion-prohibited area with a reference shape and determines whether the movement of the object is within a reference range. If the difference between the shape of the object and the reference shape is within a predetermined range and the movement of the object is within the reference range, the human detection unit 22a determines that the object is a person. If not, the human detection unit 22a determines that the object is not a person. In this way, a person present in the intrusion-prohibited area is detected.

[0070] In this manner, the main part of the human detection system 100a is configured.

[0071] Hereinafter, the human detection unit 22a may be referred to as the "human detection means", and the motion detection unit 24 may be referred to as the "motion detection means".

[0072] The hardware configuration of the main parts of the human detection device 2a is similar to that described in the first embodiment with reference to Figures 4 to 6. Therefore, a detailed description will be omitted.

[0073] That is, the function F1 of the three-dimensional model generating unit 21, the function F2a of the person detecting unit 22a, the function F3 of the output control unit 23, and the function F4 of the motion detecting unit 24 may be realized by the processor 41 and the memory 42. Alternatively, the functions F1, F2a, F3, and F4 may be realized by the processing circuit 43.

[0074] Here, the processor 41 may include dedicated processors corresponding to the functions F1, F2a, F3, and F4, the memory 42 may include dedicated memories corresponding to the functions F1, F2a, F3, and F4, and the processing circuit 43 may include dedicated processing circuits corresponding to the functions F1, F2a, F3, and F4.

[0075] Next, the operation of the human detection device 2a will be described with reference to the flowchart shown in Fig. 13. In Fig. 13, the same steps as those shown in Fig. 7 are denoted by the same reference numerals.

[0076] First, the three-dimensional model generating unit 21 generates a three-dimensional model of the intrusion-prohibited area (step ST1). ToF LiDAR technology is used to generate the three-dimensional model, as described in the first embodiment. Next, the motion detecting unit 24 detects the motion of each object present in the intrusion-prohibited area (step ST4). Doppler LiDAR technology is used to detect the motion, as described above.

[0077] Next, the human detection unit 22a detects a human being present in the entry prohibited area (step ST2a). For such detection, the three-dimensional model generated in step ST1 is used, as described in the first embodiment. Also, for such detection, the result of the movement detection in step ST4 is used, as described above. Next, the output control unit 23 executes control to output information based on the result of the detection in step ST2a (i.e., detection result information) (step ST3).

[0078] Next, the effects of using the human detection system 100a will be described.

[0079] As described above, by using the human detection system 100a, it is possible to determine whether each object present in the intrusion-prohibited area is a person, regardless of whether the object is stationary or moving. That is, it is possible to detect a person present in the intrusion-prohibited area, regardless of whether the person is stationary or moving. In other words, it is possible to accurately detect such a person.

[0080] In particular, for moving objects that exist within the entry-prohibited area, the accuracy of determining whether or not such objects are people can be improved by using a judgment based on movement in addition to a judgment based on shape, thereby improving the accuracy of detecting moving people.

[0081] Next, a modification of the human detection system 100a will be described.

[0082] The human detection system 100a can employ various modified examples similar to those described in the first embodiment.

[0083] For example, the light emitting unit 11 and the light receiving unit 12 may be provided in the human detection device 2a instead of being provided in the optical sensing device 1. In other words, the human detection device 2a may include the light emitting unit 11 and the light receiving unit 12.

[0084] Furthermore, for example, the human detection system 100a may include a three-dimensional model generation unit 21, a human detection unit 22a, and a motion detection unit 24. In other words, the three-dimensional model generation unit 21, the human detection unit 22a, and the motion detection unit 24 may constitute the main parts of the human detection system 100a.

[0085] Furthermore, for example, the human detection device 2a may include the three-dimensional model generation unit 21, the human detection unit 22a, and the movement detection unit 24. In other words, the main parts of the human detection device 2a may be configured by the three-dimensional model generation unit 21, the human detection unit 22a, and the movement detection unit 24.

[0086] Although the present disclosure has been described above with reference to the embodiments, the present disclosure is not limited to the above embodiments. Various modifications that can be understood by a person skilled in the art can be made to the configuration and details of the present disclosure within the scope of the present disclosure.

[0087] A part or all of the above-described embodiments can be described as, but is not limited to, the following supplementary notes.

[0088] [Note] [Appendix 1] a three-dimensional model generating means for generating a three-dimensional model of the prohibited area based on reflected light corresponding to the laser light irradiated onto the prohibited area in an area into which water is discharged from the dam and the surrounding area; a person detection means for detecting a person in the entry-prohibited area by using the three-dimensional model; A human detection device comprising: [Appendix 2] The human detection device described in Appendix 1, characterized in that the human detection means detects the person by using the three-dimensional model to determine whether the object is the person based on the shape of the object in the no-entry area. [Appendix 3] The human detection device according to claim 2, wherein the human detection means determines whether the object is a human by comparing a shape of the object with a reference shape. [Appendix 4] a motion detection means for detecting a motion of the object based on a difference between a frequency component included in the laser light and a frequency component included in the reflected light, The human detection means determines whether the object is a human based on the shape and movement of the object. 4. The human detection device according to claim 2 or 3, [Appendix 5] The human detection device described in Appendix 4, characterized in that the human detection means determines whether the object is the human based on the shape of the object when the object is stationary, and determines whether the object is the human based on the movement of the object when the object is moving. [Appendix 6] The human detection device described in Appendix 4, characterized in that, when the object is stationary, the human detection means determines whether the object is the person based on the shape of the object, and, when the object is moving, determines whether the object is the person based on the shape of the object and the movement of the object. [Appendix 7] an output control means for outputting information about the person to an external dam maintenance system; The information is used to notify the person of an alert. 7. A human detection device according to claim 1, wherein the human detection device comprises: [Appendix 8] a three-dimensional model generating means for generating a three-dimensional model of the prohibited area based on reflected light corresponding to the laser light irradiated onto the prohibited area in an area into which water is discharged from the dam and the surrounding area; a person detection means for detecting a person in the entry-prohibited area by using the three-dimensional model; A human detection system comprising: [Appendix 9] The human detection system of claim 8, wherein the human detection means detects the person by using the three-dimensional model to determine whether the object is the person based on the shape of the object in the no-entry area. [Appendix 10] The human detection system according to claim 9, wherein the human detection means determines whether the object is a human by comparing the shape of the object with a reference shape. [Appendix 11] a motion detection means for detecting a motion of the object based on a difference between a frequency component included in the laser light and a frequency component included in the reflected light, The human detection means determines whether the object is a human based on the shape and movement of the object. 11. The human detection system according to claim 9 or 10, [Appendix 12] The human detection system described in Appendix 11, characterized in that, when the object is stationary, the human detection means determines whether the object is the human based on the shape of the object, and, when the object is moving, determines whether the object is the human based on the movement of the object. [Appendix 13] The human detection system described in Appendix 11, characterized in that, when the object is stationary, the human detection means determines whether the object is the human based on the shape of the object, and, when the object is moving, determines whether the object is the human based on the shape of the object and the movement of the object. [Appendix 14] an output control means for outputting information about the person to an external dam maintenance system; The information is used to notify the person of an alert. 14. The human detection system according to any one of claims 8 to 13, [Appendix 15] a three-dimensional model generating means for generating a three-dimensional model of the forbidden area based on reflected light corresponding to the laser light irradiated onto the forbidden area in an area into which water is discharged from the dam and the surrounding area; A person detection means detects a person in the prohibited area using the three-dimensional model. Person detection methods. [Appendix 16] The human detection method described in Appendix 15, characterized in that the human detection means detects the person by using the three-dimensional model to determine whether the object in the no-entry area is the person based on the shape of the object. [Appendix 17] The human detection method according to claim 16, wherein the human detection means determines whether the object is a human by comparing a shape of the object with a reference shape. [Appendix 18] a motion detection means for detecting a motion of the object based on a difference between a frequency component included in the laser light and a frequency component included in the reflected light; The human detection means determines whether the object is a human based on the shape and movement of the object. 18. The human detection method according to claim 16 or 17, [Appendix 19] The human detection method described in Appendix 18, characterized in that the human detection means determines whether the object is the human based on the shape of the object when the object is stationary, and determines whether the object is the human based on the movement of the object when the object is moving. [Appendix 20] The human detection method described in Appendix 18, characterized in that, when the object is stationary, the human detection means determines whether the object is the human based on the shape of the object, and, when the object is moving, determines whether the object is the human based on the shape of the object and the movement of the object. [Appendix 21] an output control means for outputting information about the person to an external dam maintenance system; The information is used to notify the person of an alert. 21. The human detection method according to claim 15, [Appendix 22] Computer, a three-dimensional model generating means for generating a three-dimensional model of the prohibited area based on reflected light corresponding to the laser light irradiated onto the prohibited area in an area into which water is discharged from the dam and the surrounding area; a person detection means for detecting a person in the entry-prohibited area by using the three-dimensional model; A recording medium on which a program for functioning as a [Appendix 23] The recording medium described in Appendix 22, characterized in that the human detection means detects the person by using the three-dimensional model to determine whether the object in the no-entry area is the person based on the shape of the object. [Appendix 24] The recording medium according to claim 23, wherein the human detection means determines whether the object is a human by comparing a shape of the object with a reference shape. [Appendix 25] The program causes the computer to function as a motion detection unit that detects motion of the object based on a difference between a frequency component included in the laser light and a frequency component included in the reflected light, The human detection means determines whether the object is a human based on the shape and movement of the object. 25. The recording medium according to claim 23 or 24. [Appendix 26] The recording medium described in Appendix 25, characterized in that the human detection means determines whether the object is the person based on the shape of the object when the object is stationary, and determines whether the object is the person based on the movement of the object when the object is moving. [Appendix 27] The recording medium described in Appendix 25, characterized in that the human detection means, when the object is stationary, determines whether the object is the person based on the shape of the object, and, when the object is moving, determines whether the object is the person based on the shape of the object and the movement of the object. [Appendix 28] The program causes the computer to function as an output control means that outputs information about the person to an external dam maintenance system, The information is used to notify the person of an alert. 28. The recording medium according to claim 22, [Explanation of symbols]

[0089] 1. Optical sensing device 2,2a Human detection device 3 Output Device 11 Light emitting part 12 Light receiving part 21 3D model generation unit 22, 22a Human detection unit 23 Output control section 24 Motion detection section 31 Computer 41 Processors 42 Memory 43 Processing circuit 100,100a Human Detection System 200 Dam Maintenance Management System

Claims

1. a three-dimensional model generating means for generating a three-dimensional model of the prohibited area based on reflected light corresponding to the laser light irradiated onto the prohibited area in an area into which water is discharged from the dam and the surrounding area; a person detection means for detecting a person by using the three-dimensional model to compare a shape of an object in the entry-prohibited area with a plurality of reference shapes corresponding to different postures, thereby determining whether the object is a person; and Equipped with The human detection means calculates the difference between an object in the no-entry area and each of the reference shapes, and if the difference for at least one of the reference shapes is within a predetermined range, determines that the object in the no-entry area is the person.

2. a motion detection means for detecting a motion of the object based on a difference between a frequency component included in the laser light and a frequency component included in the reflected light, The human detection means determines whether the object is a human based on the shape and movement of the object. The human detection device according to claim 1 .

3. The human detection device according to claim 2, characterized in that, when the object is stationary, the human detection means determines whether the object is the human based on the shape of the object, and when the object is moving, the human detection means determines whether the object is the human based on the movement of the object.

4. The human detection device according to claim 2, characterized in that, when the object is stationary, the human detection means determines whether the object is the human based on the shape of the object, and, when the object is moving, determines whether the object is the human based on the shape of the object and the movement of the object.

5. an output control means for outputting information about the person to an external dam maintenance system; The information is used to notify the person of an alert. The human detection device according to any one of claims 1 to 4.

6. a three-dimensional model generating means for generating a three-dimensional model of the prohibited area based on reflected light corresponding to the laser light irradiated onto the prohibited area in an area into which water is discharged from the dam and the surrounding area; a person detection means for detecting a person by using the three-dimensional model to compare a shape of an object in the entry-prohibited area with a plurality of reference shapes corresponding to different postures, thereby determining whether the object is a person; and Equipped with The human detection means calculates the difference between an object in the no-entry area and each of the reference shapes, and if the difference for at least one of the reference shapes is within a predetermined range, determines that the object in the no-entry area is the person.

7. a motion detection means for detecting a motion of the object based on a difference between a frequency component included in the laser light and a frequency component included in the reflected light, The human detection means determines whether the object is a human based on the shape and movement of the object. The human detection system of claim 6 .

8. The human detection system according to claim 7, characterized in that, when the object is stationary, the human detection means determines whether or not the object is a human based on the shape of the object, and, when the object is moving, determines whether or not the object is a human based on the movement of the object.

9. The human detection system according to claim 7, characterized in that, when the object is stationary, the human detection means determines whether the object is the human based on the shape of the object, and, when the object is moving, determines whether the object is the human based on the shape of the object and the movement of the object.

10. an output control means for outputting information about the person to an external dam maintenance system; The information is used to notify the person of an alert. A human detection system according to any one of claims 6 to 9.

11. a three-dimensional model generating means for generating a three-dimensional model of the no-entry area based on reflected light corresponding to the laser light irradiated onto the no-entry area in an area into which water is discharged from the dam and the surrounding area; The human detection means is using the three-dimensional model, detecting a person by comparing a shape of an object in the no-entry area with a plurality of reference shapes corresponding to different poses to determine whether the object is a person; A difference between the object in the entry-prohibited area and each of the reference shapes is calculated, and if the difference for at least one of the reference shapes is within a predetermined range, the object in the entry-prohibited area is determined to be the person. Person detection methods.

12. a motion detection means for detecting a motion of the object based on a difference between a frequency component included in the laser light and a frequency component included in the reflected light; The human detection means determines whether the object is a human based on the shape and movement of the object. The method of claim 11 .

13. The human detection method according to claim 12, characterized in that, when the object is stationary, the human detection means determines whether or not the object is the human based on the shape of the object, and, when the object is moving, determines whether or not the object is the human based on the movement of the object.

14. The human detection method according to claim 12, characterized in that, when the object is stationary, the human detection means determines whether the object is the human based on the shape of the object, and, when the object is moving, determines whether the object is the human based on the shape of the object and the movement of the object.

15. an output control means for outputting information about the person to an external dam maintenance system; The information is used to notify the person of an alert. The method for detecting a person according to any one of claims 11 to 14.

Images