Detector

Abstract

To accurately detect an object to be detected.SOLUTION: A detector comprises: a first sensor that outputs first detection information; a second sensor that outputs second detection information; a first evaluation section that outputs a first evaluation value indicating a presence possibility ot an object to be detected on the basis of the first detection information; a second evaluation section that outputs a second evaluation value indicating the presence possibility of the object to be detected on the basis of the second detection information; and a determination section that determines whether the object to be detected is present on the basis of at least the first evaluation value. The determination section determines that the object to be detected is present on condition that the first evaluation value is equal to or greater than a first determination threshold. The determination section performs relaxation determination processing of, on a condition that the second evaluation value is equal to or greater than a first relaxation threshold (step #1), determining whether the object to be detected is present by tracking back for predetermined time (steps #3 and 6), and on a condition that the first evaluation value is equal to or greater than a second determination threshold lower than the first determination threshold, and determining that the object to be detected is present (steps #4 and 7).SELECTED DRAWING: Figure 6

Description

Technical Field

[0001] The present invention relates to a detection device for detecting a detection target object.

Background Art

[0002] As disclosed in Patent Document 1, a detection device detects a detection target object such as a person who intrudes (exists) into a monitoring area. There is a detection device that uses two different sensors in combination to integrally determine the presence of a detection target object.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] However, depending on the detection principle, the sensor may be affected by the type, size, operation mode of the detection target object, the state of the sensor, the surrounding environment, etc., and the detection accuracy may deteriorate. If the detection accuracy of any one of the sensors deteriorates, the detection device may not be able to appropriately detect the detection target object.

[0005] An object of the present invention is to accurately detect a detection target object.

Means for Solving the Problems

[0006] In order to achieve the above object, a detection device according to an embodiment of the present invention includes a first sensor that outputs first detection information, a second sensor that outputs second detection information different from the first sensor in detection principle, a first evaluation unit that outputs a first evaluation value indicating the possibility of the existence of a detection target based on the first detection information, a second evaluation unit that outputs a second evaluation value indicating the possibility of the existence of the detection target based on the second detection information, and a determination unit that determines the presence or absence of the detection target based on at least the first evaluation value. The determination unit determines that the detection target exists on the condition that the first evaluation value is equal to or greater than a first determination threshold. The determination unit determines the presence or absence of the detection target by going back in time for a predetermined time on the condition that the second evaluation value is equal to or greater than a first relaxation threshold, and performs a relaxation determination process of determining that the detection target exists on the condition that the first evaluation value is equal to or greater than a second determination threshold lower than the first determination threshold.

[0007] As described above, the detection accuracy of the sensor may deteriorate due to various conditions. For example, although the second sensor appropriately detects the detection target, the first sensor may not detect the detection target. In such a case, it becomes difficult to accurately detect the detection target.

[0008] Therefore, when the second sensor detects the detection target, the detection device determines that there is a high possibility that the detection target actually exists. Then, the detection device increases the detection sensitivity of the first sensor (relaxes the threshold for determining the presence or absence of the detection target) to determine whether the detection target exists. At this time, the detection device (determination unit) determines the presence or absence of the detection target by going back in time for a predetermined time using the relaxed threshold (second determination threshold). That is, the detection device re-determines whether the detection target existed using the relaxed threshold (second determination threshold) for the first evaluation value output after the predetermined time ago.

[0009] With such a configuration, the detection device can usually suppress the detection sensitivity of the first sensor to prevent the first sensor from making false detections. Further, the detection device re-determines whether the detection target object was present from a stage after a predetermined time has passed, using the first sensor with enhanced detection sensitivity only when the second sensor detects the detection target object and it is determined that there is a high probability that the detection target object actually exists. As a result, the detection device can accurately detect the detection target object while suppressing false detections.

[0010] Further, the determination unit determines that the detection target object is present on the condition that the second evaluation value is equal to or greater than a third determination threshold. The determination unit determines the presence or absence of the detection target object retroactively by a predetermined time on the condition that the first evaluation value is equal to or greater than a second relaxation threshold, and may perform the relaxation determination process on the condition that the second evaluation value is equal to or greater than a fourth determination threshold that is lower than the third determination threshold.

[0011] With such a configuration, the detection device can determine the presence or absence of the detection target object using the second sensor. Also in this case, the detection device can usually suppress the detection sensitivity of the second sensor to prevent the second sensor from making false detections. Further, the detection device re-determines whether the detection target object was present in the past, using the second sensor with enhanced detection sensitivity only when the first sensor detects the detection target object and it is determined that there is a high probability that the detection target object actually exists. As a result, the detection device can accurately detect the detection target object while suppressing false detections.

[0012] Further, the determination unit may perform the relaxation determination process retroactively from the time when it is detected that the second evaluation value is equal to or greater than the first relaxation threshold up to one hour before.

[0013] With such a configuration, the detection device can re-detect the detection target by the first sensor (determine whether the detection target exists) in the time period after one hour before the second sensor detects the detection target. As a result, the detection device can re-determine whether the detection target exists immediately after the possibility of the existence of the detection target occurs, and can accurately detect the detection target.

[0014] Further, the determination unit may perform the relaxation determination process until two hours after the time when it is detected that the second evaluation value is equal to or greater than the first relaxation threshold value.

[0015] If the relaxation determination process continues for a long time, the possibility of the first sensor making a false detection increases. According to the above configuration, since the detection device ends the relaxation determination process when two hours have elapsed after starting the relaxation determination process, the period during which the possibility of false detection increases can be limited, and false detection can be suppressed.

[0016] Further, the determination unit may perform the relaxation determination process retroactively until three hours before the time when it is detected that the first evaluation value is equal to or greater than the second relaxation threshold value.

[0017] With such a configuration, the detection device can re-detect the detection target by the second sensor (determine whether the detection target exists) in the time period after three hours before the first sensor detects the detection target. As a result, the detection device can re-determine whether the detection target exists immediately after the possibility of the existence of the detection target occurs, and can accurately detect the detection target.

[0018] Further, the determination unit may perform the relaxation determination process until four hours after the time when it is detected that the first evaluation value is equal to or greater than the second relaxation threshold value.

[0019] With such a configuration, since the detection device ends the relaxation determination process when four hours have elapsed after starting the relaxation determination process, the period during which the possibility of false detection increases can be limited, and false detection can be suppressed.

[0020] Further, the first sensor may be one of at least either an infrared sensor and a radio wave sensor, and a camera sensor, and the second sensor may be the other of at least either the infrared sensor and the radio wave sensor, and the camera sensor.

[0021] With such a configuration, the detection device can accurately detect a detection target object.

[0022] Further, the detection target object may be at least either a human body and a vehicle.

[0023] With such a configuration, the detection device can accurately detect the entry of a human body or a vehicle.

Brief Description of the Drawings

[0024]

Figure 1

Figure 2

Figure 3

Figure 4

Figure 5

Figure 6

Embodiments for Carrying Out the Invention

[0025] 〔Detection Device〕 First, the overall configuration of the detection device 1 according to the present embodiment will be described with reference to FIGS. 1 to 3.

[0026] The detection device 1 detects at least one detection target object among a human body and a vehicle that exists in or has entered (hereinafter simply referred to as exists) the area to be detected by a detector. When the detection device 1 detects a detection target object, it performs a predetermined notification. Note that the detection target object is not limited to a human body or a vehicle and can be any object.

[0027] The detection device 1 includes, as a detector, a PIR3 (equivalent to a Passive Infrared Ray sensor / the first sensor), which is a type of infrared sensor, a camera sensor 4 (equivalent to the second sensor), a lighting 6, a control unit 8, and a storage unit 9. The PIR3 detects the temperature of an object existing in the area to be detected and outputs it as sensor detection information 11 (equivalent to the detection information / the first detection information), which is detection information. Note that two or more PIR3s may be provided, and they may be configured to detect an object existing in a predetermined area in combination. The camera sensor 4 captures an image of the area to be detected and outputs the captured image as video detection information 12 (equivalent to the second detection information), which is detection information.

[0028] The control unit 8 detects a detection target object existing in the area to be detected based on the video detection information 12 output by the camera sensor 4 and the sensor detection information 11 output by the PIR3. For example, the control unit 8 analyzes the temperature distribution and transition in the area to be detected from the sensor detection information 11 to determine whether a detection target object exists (detect the detection target object). Similarly, the control unit 8 determines whether a detection target object exists (detect the detection target object) by performing image analysis on the video detection information 12 acquired by the camera sensor 4. The storage unit 9 stores various types of information.

[0029] When the control unit 8 determines that a detection target object exists based on the sensor detection information 11 and also determines that a detection target object exists based on the video detection information 12, it determines that a detection target object exists in the area to be detected.

[0030] In addition, when a detection target object exists in the area to be detected, the detection device 1 gives notice that the detection target object exists. The detection device 1 is provided with a notification unit 14, and the notification may be performed by the notification unit 14, but it may also be performed by a device different from the detection device 1, such as a mobile terminal or a display device. Further, the notification can be performed by any method, such as the transmission of voice or warning sound, the display of characters or pictures, the display by a warning light, or the transmission to a communication terminal.

[0031] 〔Configuration of the control unit〕 Next, a specific configuration example of the control unit 8 of the detection device 1 will be described with reference to FIGS. 1 to 3.

[0032] The control unit 8 includes a communication unit 21, an operation control unit 22, an image evaluation unit 24 (corresponding to a detection information evaluation unit / a second evaluation unit), a sensor evaluation unit 25 (corresponding to a detection information evaluation unit / a first evaluation unit), and a determination unit 27.

[0033] The communication unit 21 performs data communication and transmits and receives various data. For example, the communication unit 21 receives information such as sensor detection information 11 and video detection information 12 from the PIR 3 and the camera sensor 4, writes the information into the storage unit 9, and further reads the information from the storage unit 9. In addition, the communication unit 21 can also perform data communication with devices external to the detection device 1.

[0034] The operation control unit 22 controls the operation of the control unit 8. That is, the communication unit 21, the image evaluation unit 24, the sensor evaluation unit 25, and the determination unit 27 operate under the control of the operation control unit 22.

[0035] The image evaluation unit 24 analyzes the video detection information 12 (captured image) output by the camera sensor 4 and stored in the storage unit 9, and evaluates the possibility of the existence of the detection target. That is, the image evaluation unit 24 analyzes the video detection information 12 and outputs a video evaluation value 15 (corresponding to the second evaluation value), which is an evaluation result indicating the possibility of the existence of the detection target. Specifically, the camera sensor 4 is an AI camera, and the image evaluation unit 24 analyzes the video detection information 12 with AI (artificial intelligence) using the learned model 29 stored in the storage unit 9 and outputs the video evaluation value 15. The learned model 29 is generated in advance so as to output a video evaluation value 15 indicating the possibility of the existence of the detection target when a captured image is input. The video evaluation value 15 is, for example, a quantification of the likelihood of the existence of the detection target, and is indicated by a percentage or a 10 - level evaluation, etc. Note that if the camera sensor 4 is not an AI camera, the image evaluation unit 24 may perform image analysis by any method and output the video evaluation value 15.

[0036] The sensor evaluation unit 25 analyzes the sensor detection information 11 output by the PIR 3 and stored in the storage unit 9, and evaluates the possibility of the existence of the detection target. That is, the sensor evaluation unit 25 analyzes the sensor detection information 11 and outputs a sensor evaluation value 16 (corresponding to the first evaluation value), which is an evaluation result indicating the possibility of the existence of the detection target. The sensor evaluation value 16 is, for example, a quantification of the likelihood of the existence of the detection target, and is indicated by a percentage or a 10 - level evaluation, etc.

[0037] The determination unit 27 determines the presence or absence of the detection target based on the video evaluation value 15 and the sensor evaluation value 16. Specifically, the determination unit 27 compares the video evaluation value 15 and the sensor evaluation value 16 with the determination criterion 17. The determination criterion 17 is set in advance for each of the video evaluation value 15 and the sensor evaluation value 16 and stored in the storage unit 9. The determination criterion 17 corresponding to the sensor evaluation value 16 of the PIR 3 is the sensor determination criterion 17A (corresponding to the first determination threshold), and the determination criterion 17 corresponding to the video evaluation value 15 of the camera sensor 4 is the video determination criterion 17C (corresponding to the third determination threshold).

[0038] Then, when both the video evaluation value 15 and the sensor evaluation value 16 are equal to or greater than the corresponding determination criteria 17, the determination unit 27 determines that the detection target object exists. That is, when the video evaluation value 15 is equal to or greater than the video determination criteria 17C and the sensor evaluation value 16 is equal to or greater than the sensor determination criteria 17A, the determination unit 27 determines that the detection target object exists. When at least one of the video evaluation value 15 and the sensor evaluation value 16 is smaller than the corresponding determination criteria 17, the determination unit 27 determines that the detection target object does not exist. When it is determined that the detection target object exists, the determination unit 27 causes the notification unit 14 to notify that the detection target object exists (the detection target unit has been detected).

[0039] As described above, the detection accuracy of the detector deteriorates due to the environment and the like. For example, the detection accuracy of PIR3 is affected by the shape and movement state of the object, and the surrounding environment such as the ambient temperature. The detection accuracy of the camera sensor 4 is affected by the environment such as the weather and sunlight conditions, and the characteristics of the object. When the detection accuracy of at least one of PIR3 and the camera sensor 4 decreases, the detection device 1 may perform a false detection. For example, the detection device 1 may determine that the detection target object does not exist even though the detection target object exists in the area to be detected, or may determine that the detection target object exists even though the detection target object does not exist in the area to be detected. Then, the detection device 1 may perform a missed notification in which it does not notify when it should notify, or a false notification in which it notifies when it should not notify.

[0040] In order to suppress such false detections, the detection device 1 performs various processes for adjusting the detection mode by other detectors according to the detection state of one detector. For example, the detection device 1 adjusts the detection sensitivity of other detectors according to the detection state of one detector. Hereinafter, two processes that the detection device 1 can perform will be described as Embodiments 1 and 2. Note that the detection device 1 may implement Embodiments 1 and 2 alone or in combination.

[0041] [Embodiment 1] When the detection device 1 according to Embodiment 1 detects a detection target outside the range of the detection area APM where the PIR 3 detects the detection target, it executes a relaxation process for relaxing the determination criterion 17 for the PIR 3 to determine the presence of the detection target. Hereinafter, a specific configuration example of the relaxation process will be described.

[0042] 〔Detection Area〕 First, with reference to FIG. 1, the detection areas of the PIR 3 and the camera sensor 4 will be described using FIGS. 2 and 4.

[0043] The PIR 3 includes a main PIR 3A and a sub PIR 3B. The main PIR 3A and the sub PIR 3B are arranged such that the detection area APS (corresponding to the sub area) of the sub PIR 3B is located near the outside of the detection area APM (corresponding to the main area) of the main PIR 3A. For example, the detection area APM (corresponding to the main area) of the main PIR 3A is adjacent to the detection area APS (corresponding to the sub area) of the sub PIR 3B.

[0044] The detection area ACA of the camera sensor 4 is the imaging area that the camera sensor 4 captures. At least the detection area APM of the detection areas APM and APS is included inside the detection area ACA.

[0045] In the detection area APM, based on the sensor evaluation value 16 of the PIR 3, it is determined whether a detection target exists (the presence or absence of the detection target). In the area where the detection area APM of the detection area ACA overlaps, based on the video evaluation value 15 of the camera sensor 4, it is determined whether a detection target exists.

[0046] The detection area APM corresponds to the inner area, and the area outside the detection area APM in the detection area ACA corresponds to the outer area. The detection area APS belongs to the outer area. That is, the area located inside the outer area of the detection area ACA (imaging area) is the inner area, and the inner area coincides with the detection area APM.

[0047] 〔Relaxation Process〕 The relaxation process will be described using FIGS. 1 to 4.

[0048] As described above, the image evaluation unit 24 analyzes the video detection information 12 acquired over time, and outputs a video evaluation value 15 which is an evaluation result indicating the possibility of the presence of the detection target object. The determination unit 27 continuously compares the continuously output video evaluation value 15 with the video determination criterion 17C, and determines whether or not the detection target object exists (the presence or absence of the detection target object).

[0049] Similarly, the sensor evaluation unit 25 analyzes the sensor detection information 11 acquired over time, and outputs a sensor evaluation value 16 which is an evaluation result indicating the possibility of the presence of the detection target object. The determination unit 27 continuously compares the continuously output sensor evaluation value 16 with the sensor determination criterion 17A, and determines whether or not the detection target object exists (the presence or absence of the detection target object).

[0050] When the detection target object is detected in the outer region during such detection of the detection target object, the determination unit 27 performs a relaxation process of relaxing the sensor determination criterion 17A which is the determination criterion for the presence or absence of the detection target object based on the sensor evaluation value 16, and the video determination criterion 17C which is the determination criterion for the presence or absence of the detection target object based on the video evaluation value 15.

[0051] Specifically, first, the determination unit 27 determines whether or not the detection target object is detected in the external region (step #1 in FIG. 4). At this time, when the video evaluation value 15 of the object located in the external region is equal to or higher than the video determination criterion 17C, the determination unit 27 determines that the detection target object is detected in the external region. Also, when the sensor evaluation value 16 of the object located in the detection area APS which is the external region is equal to or higher than the sensor determination criterion 17A, the determination unit 27 determines that the detection target object is detected in the external region.

[0052] Next, when the detection target object is detected in the external region (Yes in step #1 in FIG. 4), the determination unit 27 performs a relaxation process of relaxing the determination criterion 17 and increasing the detection sensitivity of the PIR 3 and the camera sensor 4 (step #2 in FIG. 4).

[0053] Specifically, as a relaxation process, the determination unit 27 changes the determination criterion 17 of PIR3 from the sensor determination criterion 17A to a reduced sensor determination criterion 17B lower than the sensor determination criterion 17A, and changes the determination criterion 17 of the camera sensor 4 from the video determination criterion 17C to a reduced video determination criterion 17D lower than the video determination criterion 17C.

[0054] In this way, in the normal state, the sensitivity of the detectors (PIR3 and camera sensor 4) can be suppressed, and it is possible to suppress the detectors from falsely detecting that the detection target exists even though the detection target does not exist. And only when it is recognized that there is a high possibility that the detection target exists, the sensitivity of the detectors can be increased to suppress the detectors from falsely detecting that the detection target does not exist even though the detection target exists. As a result, the detection device 1 according to the present embodiment can accurately detect the detection target while suppressing false detection.

[0055] The relaxation process may continue thereafter, but when the detection target is no longer detected and a predetermined elapsed time (corresponding to a predetermined time) has elapsed, the determination unit 27 may end the relaxation process. If the detection target is not detected in the external area and the detection target is not detected in the internal area either, it can be said that there is a high possibility that the detection target does not exist. If the determination criterion 17 is continuously relaxed in this state, the possibility of false detection increases. By appropriately ending the relaxation process, false detection can be suppressed.

[0056] Next, the determination unit 27 compares the evaluation result with the determination criterion 17 in the internal area (detection area APM) (step #3 in FIG. 4). At this time, when it is determined that there is no detection target in the external area (No in step #1 of FIG. 4), the determination unit 27 uses the sensor determination criterion 17A and the video determination criterion 17C as the determination criterion 17. Also, while the relaxation process is being executed, the determination unit 27 uses the reduced sensor determination criterion 17B and the reduced video determination criterion 17D as the determination criterion 17.

[0057] When the evaluation result is 17 or more, the determination unit 27 determines that a detection target exists in the internal area (detection area APM) (detects the detection target) (step #4 in FIG. 4).

[0058] When it is determined that a detection target exists in the internal area (detection area APM) (step #4 in FIG. 4), the determination unit 27 performs predetermined notification as described above (step #5 in FIG. 4).

[0059] Next, when the determination unit 27 determines that a detection target exists based on the evaluation results (sensor evaluation value 16 and video evaluation value 15), the relaxation process may be terminated (step #6 in FIG. 4). That is, in the relaxation process, when the determination unit 27 determines that a detection target exists, the determination criterion 17 may be returned to the sensor determination criterion 17A and the video determination criterion 17C to terminate the relaxation process (step #6 in FIG. 4).

[0060] After a detection target is detected in the external area and then detected in the internal area, the determination criterion 17 is relaxed, indicating that the detection target to be detected has already been detected. If the relaxation process continues after that, the possibility of false detection increases. By terminating the relaxation process when a detection target is detected, false detection can be suppressed.

[0061] After the relaxation process is terminated (step #6 in FIG. 4), the determination unit 27 continues the detection process using the sensor determination criterion 17A and the video determination criterion 17C (step #7 in FIG. 4). Also, when a detection target is not detected in the internal area during the relaxation process (step #3 No in FIG. 4), the determination unit 27 may continue the relaxation process until a predetermined time has elapsed and then return the determination criterion 17 to the sensor determination criterion 17A and the video determination criterion 17C to continue the detection process (step #7 in FIG. 4). Also, when an inspection target is not detected in the external area (step #3 No in FIG. 4) and the inspection target is not detected with the determination criterion 17 remaining as the sensor determination criterion 17A and the video determination criterion 17C (step #3 No in FIG. 4), the determination unit 27 continues the detection process (step #7 in FIG. 4).

[0062] 〔Embodiment 2〕 The detection device 1 according to Embodiment 2 includes a PIR 3 that detects a detection target based on a first determination threshold that is a determination criterion 17, and a camera sensor 4 that detects a detection target based on a third determination threshold that is a determination criterion 17. When the evaluation value (second evaluation value) of the camera sensor 4 satisfies a predetermined relaxation determination criterion 35 (first relaxation determination threshold), the determination criterion 17 (first determination threshold) is decreased (decreased to the second determination threshold) to increase the detection sensitivity of the PIR 3, and it is determined whether or not the evaluation value (first evaluation value) is equal to or greater than the determination criterion 17 (second determination threshold) that has been decreased by the PIR 3 retroactively in time, and a relaxation determination process for re-determining the presence or absence of the detection target is executed. Similarly, when the evaluation value (first evaluation value) of the PIR 3 satisfies a predetermined relaxation determination criterion 35 (second relaxation determination threshold), the detection device 1 according to Embodiment 2 decreases the determination criterion 17 (third determination threshold) (decreases to the fourth determination threshold) to increase the detection sensitivity of the camera sensor 4, and determines whether or not the evaluation value (second evaluation value) is equal to or greater than the determination criterion 17 (fourth determination threshold) that has been decreased by the camera sensor 4 retroactively in time, and executes a relaxation determination process for re-determining the presence or absence of the detection target. In the following description, a case where the determination criterion 17 before the decrease and the relaxation determination criterion 35 have the same value will be described as an example.

[0063] That is, when one of the PIR 3 and the camera sensor 4 of the detection device 1 according to Embodiment 2 detects a detection target, the detection sensitivity of the other is increased, and a relaxation determination process is executed to re-detect the detection target again by going back the retroactive time 31 (predetermined time) after detecting the detection target. Hereinafter, a specific configuration example of the relaxation determination process will be described.

[0064] As described above, the image evaluation unit 24 analyzes the video detection information 12 acquired over time and outputs a video evaluation value 15 indicating the possibility of the presence of the detection target. The determination unit 27 continuously compares the output video evaluation value 15 with a video determination criterion 17C (corresponding to the third determination threshold) to determine whether or not the detection target is present (the presence or absence of the detection target).

[0065] Similarly, the sensor evaluation unit 25 analyzes the sensor detection information 11 acquired over time, and outputs a sensor evaluation value 16 indicating the possibility of the presence of the detection target. The determination unit 27 continuously compares the continuously output sensor evaluation value 16 with a sensor determination criterion 17A (corresponding to the first determination threshold), and determines whether or not the detection target exists (the presence or absence of the detection target).

[0066] When detecting such a detection target, when one of the detectors of the PIR 3 and the camera sensor 4 detects the detection target, the determination criterion 17 for determining whether or not the detection target exists (the presence or absence of the detection target) by the other detector is relaxed, and further, a relaxation determination process is performed to determine again whether or not the detection target exists by the other detector by going back a predetermined retrospective time 31.

[0067] Specifically, first, the determination unit 27 determines whether or not the video evaluation value 15 is greater than or equal to the video determination criterion 17C, and also determines whether or not the sensor evaluation value 16 is greater than or equal to the sensor determination criterion 17A (step #1 in FIG. 6). Thereby, the determination unit 27 determines whether both the PIR 3 and the camera sensor 4 have detected the detection target, whether only the PIR 3 has detected the detection target, whether only the camera sensor 4 has detected the detection target, or whether neither the PIR 3 nor the camera sensor 4 has detected the detection target.

[0068] Next, when the video evaluation value 15 is less than the video determination criterion 17C and the sensor evaluation value 16 is greater than or equal to the sensor determination criterion 17A, the determination unit 27 determines that only the PIR 3 has detected the detection target (branch of step #1 in FIG. 6).

[0069] In this case, as a relaxation determination process, the determination unit 27 relaxes the determination criterion 17 to a reduced video determination criterion 17E (corresponding to the fourth determination threshold) lower than the video determination criterion 17C in order to increase the detection sensitivity of the camera sensor 4 (step #2 in FIG. 6), traces back by a video traceback time 31C (corresponding to the third time) which is the traceback time 31 (step #3 in FIG. 6), determines whether all of the video evaluation values 15 captured and evaluated before and after by the video traceback time 31C are equal to or higher than the reduced video determination criterion 17E, and re-executes the detection of the detection target (step #4 in FIG. 6). Note that the traced-back determination in the relaxation determination process is performed by tracing back to a point in time that is the video traceback time 31C before the point in time when the determination unit 27 determines (detects) that the sensor evaluation value 16 is equal to or higher than the sensor determination criterion 17A.

[0070] Similarly, when the sensor evaluation value 16 is smaller than the sensor determination criterion 17A and the video evaluation value 15 is equal to or higher than the video determination criterion 17C, the determination unit 27 determines that only the camera sensor 4 has detected the detection target (branch at step #1 in FIG. 6).

[0071] In this case, as a relaxation determination process, the determination unit 27 relaxes the determination criterion 17 to a reduced sensor determination criterion 17F (corresponding to the second determination threshold) lower than the sensor determination criterion 17A in order to increase the detection sensitivity of the PIR 3 (step #5 in FIG. 6), traces back by a sensor traceback time 31P (corresponding to the first time) which is the traceback time 31 (step #6 in FIG. 6), determines whether all of the sensor evaluation values 16 output and evaluated before and after by the sensor traceback time 31P are equal to or higher than the reduced sensor determination criterion 17F, and re-executes the detection of the detection target (step #7 in FIG. 6). Note that the traced-back determination in the relaxation determination process is performed by tracing back to a point in time that is the sensor traceback time 31P before the point in time when the determination unit 27 determines (detects) that the video evaluation value 15 is equal to or higher than the video determination criterion 17C.

[0072] When one of the PIR 3 and the camera sensor 4 detects a detection target, even if the other does not detect the detection target, there may actually be a detection target. When there is actually a detection target, by increasing the detection sensitivity of the other detector and re-detecting during the time period when it could not be detected, it may be possible to detect the detection target. Conversely, if the detection target cannot be detected even when the detection sensitivity of the other detector is increased, it is highly likely that there is actually no detection target.

[0073] Therefore, when one of the PIR 3 and the camera sensor 4 of the detection device 1 according to Embodiment 2 detects a detection target, after increasing the sensitivity of the other detector, a relaxation determination process is performed to re-detect the detection target for the information acquired by the PIR 3 and the camera sensor 4 in the vicinity of the time when one of the PIR 3 and the camera sensor 4 detects the detection target.

[0074] Thereby, when there is actually a detection target, even if the other detector cannot detect the existence of the detection target, if one detector can detect the detection target, the sensitivity can be increased and the detection by the other detector can be re-done, and the situation where the detection device 1 cannot detect the detection target despite the actual existence of the detection target can be alleviated.

[0075] Also, when the video evaluation value 15 is equal to or higher than the video determination criterion 17C and the sensor evaluation value 16 is equal to or higher than the sensor determination criterion 17A (branch at step #1 in FIG. 6), the determination unit 27 determines that both the PIR 3 and the camera sensor 4 satisfy the determination criterion 17 and detect the detection target (step #8 in FIG. 6). When the relaxation determination process is performed and the video evaluation value 15 is equal to or higher than the reduced video determination criterion 17E (Yes at step #4 in FIG. 6), and when the sensor evaluation value 16 is equal to or higher than the reduced sensor determination criterion 17F (Yes at step #7 in FIG. 6), the determination unit 27 similarly determines that the detection target has been detected (step #8 in FIG. 6).

[0076] In this case, the determination unit 27 notifies that the detection target has been detected (step #9 in FIG. 6). After that, it may be determined that the relaxed determination criterion 17 is appropriate as the determination criterion 17 for the detection of the detection target, and the relaxed determination criterion 17 may be continuously used, or the relaxed determination criterion 17 may be restored. For example, after detecting the detection target in the relaxed determination process, the determination unit 27 restores the reduced video determination criterion 17E, which is the relaxed determination criterion 17, to the video determination criterion 17C, or after restoring the reduced sensor determination criterion 17F, which is the relaxed determination criterion 17, to the sensor determination criterion 17A (step #10 in FIG. 6), the detection process may be continued using the video determination criterion 17C and the sensor determination criterion 17A as the determination criterion 17 (step #11 in FIG. 6).

[0077] Also, in step #1 of FIG. 6, when the video evaluation value 15 is smaller than the video determination criterion 17C and the sensor evaluation value 16 is smaller than the sensor determination criterion 17A, the determination unit 27 determines that the detection target does not exist (the detection target has not been detected).

[0078] Similarly, in a state where the relaxed determination process is being performed, when it is not determined that the video evaluation value 15 is equal to or greater than the reduced video determination criterion 17E (No in step #4 of FIG. 6), and when it is not determined that the sensor evaluation value 16 is equal to or greater than the reduced sensor determination criterion 17F (No in step #7 of FIG. 6), the determination unit 27 determines that the detection target does not exist (the detection target has not been detected) (step #12 in FIG. 6).

[0079] Note that the relaxation determination process may continue until only PIR3 or the camera sensor 4 detects the detection target object and then ends, or may continue during a predetermined retroactive period 33 and then end. Also, the retroactive period 33 may be the same period for PIR3 and the camera sensor 4, or may be different periods. Specifically, the relaxation determination process is performed only for a video retroactive period 33C (second hour) from the time when it is detected that the video evaluation value 15 is equal to or greater than the video determination criterion 17C. Also, the relaxation determination process is performed only for a sensor retroactive period 33P (fourth hour) from the time when it is detected that the sensor evaluation value 16 is equal to or greater than the sensor determination criterion 17A. Also, when the state where only PIR3 or the camera sensor 4 detects the detection target object continues, the start time of measurement of the retroactive period 33 is shifted to a later time accordingly. That is, the relaxation determination process starts from the time when only PIR3 or the camera sensor 4 detects the detection target object, but if the detection continues, the duration of the relaxation determination process is also extended, and the relaxation determination process continues until the time obtained by adding the retroactive period 33 to the time when the time is returned by the retroactive time 31 from the time when the detected PIR3 or camera sensor 4 stops detecting the detection target object. Note that depending on the retroactive period 33, the relaxation determination process may continue beyond the current time. That is, the relaxation determination process may continue the detection process with the relaxed determination criterion 17 once executed, or may end by satisfying an arbitrary end condition. The end conditions for the relaxation determination process when ending include cases where the detection period 33 has elapsed after the detector no longer satisfies the relaxation determination criterion 35, cases where the detection target object is detected and notification is performed, cases where the time has elapsed until the time when the detector satisfies the relaxation determination criterion 35 (current time) after being retroactively processed, etc., and these may be used in combination.

[0080] When the end condition is satisfied and the relaxation determination process ends, the reduced video determination criterion 17E, which is the relaxed determination criterion 17, is returned to the video determination criterion 17C (step #13 in FIG. 6), or the reduced sensor determination criterion 17F, which is the relaxed determination criterion 17, is returned to the sensor determination criterion 17A (step #14 in FIG. 6). After returning the determination criterion 17, the determination unit 27 continues the detection process using the video determination criterion 17C and the sensor determination criterion 17A as the determination criterion 17 (step #11 in FIG. 6).

[0081] 〔Alternative Embodiment〕 (1) In Embodiment 1, the reduced sensor determination criterion 17B and the reduced video determination criterion 17D relaxed by the relaxation process may always have the same reference value. However, the determination unit 27 may change the degree of relaxation 18 when relaxing the determination criterion 17 according to the detection state of the detection target in the external region. Specifically, the determination unit 27 changes the degree of relaxation 18 when relaxing the determination criterion based on at least any one of the position where the detection target is detected, the moving direction of the detection target, and the moving amount of the detection target.

[0082] The closer the position where the detection target is detected is to the detection area APM (internal area), the higher the possibility that the detection target will enter the internal area later. If the detection target is moving in the direction toward the detection area APM (internal area), the possibility that the detection target will enter the internal area later is high. The larger the moving amount of the detection target, the higher the possibility that the detection target will enter the internal area later. And the higher the possibility that the detection target will enter the internal area later, the greater the relaxation of the determination criterion 17 to increase the sensitivity of the detectors (PIR3 and camera sensor 4). Although it can be expected that the possibility of false detection is low, the detection target can be detected efficiently and accurately.

[0083] (2) In Embodiment 1 or the above-described alternative embodiments, when the video evaluation value 15 of an object located in the external region is equal to or greater than the video determination criterion 17C, or when the sensor evaluation value 16 of an object located in the external region is equal to or greater than the sensor determination criterion 17A, the determination unit 27 may determine that a detection target object exists in the external region and execute relaxation processing. However, in the external region, relaxation processing may be executed when both the PIR 3 and the camera sensor 4 detect the detection target object. That is, when the video evaluation value 15 of an object located in the external region is equal to or greater than the video determination criterion 17C and the sensor evaluation value 16 of an object located in the external region is equal to or greater than the sensor determination criterion 17A, the determination unit 27 may determine that a detection target object exists in the external region and execute relaxation processing. Thereby, relaxation processing can be performed only when there is a higher possibility that a detection target object exists in the internal region, false detection can be suppressed with higher accuracy, and the detection target object can be detected with higher accuracy.

[0084] (3) In Embodiment 1 or the above-described alternative embodiments, the reference threshold for executing relaxation processing may be the sensor determination criterion 17A and the video determination criterion 17C, which are the determination criteria 17 for detecting a detection target object. However, the reference threshold for executing relaxation processing may be determined separately.

[0085] (4) In Embodiment 1 or the above-described alternative embodiments, the determination unit 27 is not limited to a configuration that relaxes both the sensor determination criterion 17A and the video determination criterion 17C as relaxation processing, and may relax only one of them. Thereby, the detection accuracy of the detection target object can be improved while performing the minimum necessary relaxation processing according to the situation. For example, when the PIR 3 detects a detection target object in the external region, the determination unit 27 relaxes the video determination criterion 17C without changing the sensor determination criterion 17A, and when the camera sensor 4 detects a detection target object in the external region, the determination unit 27 relaxes the sensor determination criterion 17A without relaxing the video determination criterion 17C. Thereby, without increasing the sensitivity of the detectors (PIR 3 and camera sensor 4) more than necessary, the detection device 1 can accurately detect the detection target object.

[0086] (5) In Embodiment 1 or each of the above-described respective embodiments, the detection area APS may be an internal area instead of an external area. That is, even in the detection area APS, the presence or absence of the detection target object may be determined. As a result, the area for detecting the detection target object is expanded, and the determination unit 27 can accurately detect the detection target object. Further, the PIR 3 may be configured not to include the sub-PIR 3B.

[0087] (6) In Embodiment 2, the relaxation determination process may be executed when either one of the PIR 3 and the camera sensor 4 detects the detection target object. However, the detector that determines whether to execute the relaxation determination process may be fixed to either one of the PIR 3 and the camera sensor 4. That is, the relaxation determination process may be configured to be executed when the camera sensor 4 detects the detection target object but not executed when the PIR 3 detects the detection target object. Conversely, it may be configured to be executed when the PIR 3 detects the detection target object but not executed when the camera sensor 4 detects the detection target object.

[0088] (7) In Embodiment 2 or the above-described other embodiments, the retrospective time 31 may be different times (video retrospective time 31C and sensor retrospective time 31P) for the PIR 3 and the camera sensor 4, or may be the same time. When the retrospective times 31 are different, the detection device 1 can perform the relaxation determination process more accurately according to the detection characteristics of the PIR 3 and the camera sensor 4 and the situation of the detection target object. When the retrospective times 31 are the same time, the detection device 1 can execute the relaxation determination process with a simple configuration.

[0089] (8) In Embodiment 2 or each of the above-described respective embodiments, the relaxation determination process may be executed on the condition that the video evaluation value 15 is equal to or greater than the video determination criterion 17C, or may be executed based on a video relaxation determination criterion 35C (corresponding to the first relaxation threshold value), which is a relaxation determination criterion 35 different from the determination criterion 17. That is, the relaxation determination process is executed on the condition that the video evaluation value 15 is equal to or greater than the video relaxation determination criterion 35C. Further, the relaxation determination process may be executed on the condition that the sensor evaluation value 16 is equal to or greater than the sensor determination criterion 17A, or may be executed based on a sensor relaxation determination criterion 35P (corresponding to the second relaxation threshold value), which is a relaxation determination criterion 35 different from the determination criterion 17. That is, the relaxation determination process is executed on the condition that the sensor evaluation value 16 is equal to or greater than the sensor relaxation determination criterion 35P.

[0090] For example, increasing the determination criterion 17 means lowering the detection sensitivity. Detecting a detection target with a low detection sensitivity can be presumed to highly likely indicate the presence of the detection target. Making the relaxation determination criterion 35 larger than the determination criterion 17 means that when either the PIR 3 or the camera sensor 4 with a low detection sensitivity detects a detection target, the relaxation determination process is executed. By adopting a configuration in which the relaxation determination process is executed when the evaluation result is equal to or greater than the relaxation determination criterion 35, which is larger than the determination criterion 17, even if one of the PIR 3 and the camera sensor 4 does not detect the detection target, when the other detects the detection target with a low detection sensitivity, it is possible to increase the detection sensitivity of one detector to make it easier to detect the detection target. As a result, when it is presumed that there is a high possibility that the detection target exists, even when detecting a detection target that is difficult to detect, it is possible to easily detect the detection target by increasing the detection sensitivity of the detector that has not detected it.

[0091] Conversely, if the relaxation determination criterion 35 is made smaller than the determination criterion 17 to detect the detection target, it can be said that there is a certain degree of possibility even if the possibility of the detection target existing is low. In such a case, by increasing the detection sensitivity of the detector that has not detected the detection target, even if the detection target is difficult to detect, it can be easily detected. As a result, it is possible to suppress false detection in which the detection target cannot be detected even though the detection target exists.

[0092] (9) In each of the above embodiments or each of the above alternative embodiments, the detector is not limited to the PIR3 or the camera sensor 4, and various detectors capable of recognizing an object, such as infrared sensors and radio wave sensors with other configurations, can be arbitrarily combined and used as long as the object detection principles are different. Further, the detector may include at least the camera sensor 4.

[0093] (10) In each of the above embodiments or each of the above alternative embodiments, at least a part of the functions of the control unit 8 and the storage unit 9 may be provided in another device such as a server that is provided separately from the detection device 1 and can communicate with the detection device 1. For example, when the camera sensor 4 is an AI camera, the image evaluation unit 24 may be built into the camera sensor 4.

[0094] (11) In each of the above embodiments or each of the above alternative embodiments, the control unit 8 is not limited to being composed of the above functional blocks, and may be composed of arbitrary functional blocks. For example, each functional block of the control unit 8 may be further subdivided, or conversely, a part or all of the functional blocks may be combined. Further, at least a part of the functions of the control unit 8 may be configured outside the detection device 1. Further, the functions of the control unit 8 are not limited to the above functional blocks, and may be realized by a method executed by arbitrary functional blocks. Further, a part or all of the functions of the control unit 8 may be configured by software. The program related to the software is stored in an arbitrary storage device such as the storage unit 9, and is executed by a processor such as a CPU provided in the control unit 8 or a separately provided processor such as a server.

Industrial Applicability

[0095] The present invention can be applied to a detection device that detects a detection target such as a human body or a vehicle.

Explanation of Signs

[0096] 1 Detection device 3 PIR (Sensor First Sensor) 4 Camera sensor (Second sensor) 8 Control unit 9 Storage unit 11 Sensor detection information (Detection information First detection information) 12 Video detection information (Second detection information) 17 Judgment criteria 17A Sensor judgment criteria (First judgment threshold) 17C Video judgment criteria (Third judgment threshold) 17E Reduced video judgment criteria (Fourth judgment threshold) 17F Reduced sensor judgment criteria (Second judgment threshold) 24 Image evaluation unit (Detection information evaluation unit Second evaluation unit) 25 Sensor evaluation unit (Detection information evaluation unit First evaluation unit) 27 Judgment unit 31 Retroactive time 31C Video retroactive time (Third time) 31P Sensor retroactive time (First time) 35 Relaxation judgment criteria 35C Video relaxation judgment criteria (First relaxation threshold) 35P Sensor relaxation judgment criteria (Second relaxation threshold)

Claims

1. a first sensor that outputs first detection information; a second sensor that outputs second detection information different in detection principle from the first sensor; a first evaluation unit that outputs a first evaluation value indicating the possibility of the presence of a detection target based on the first detection information; a second evaluation unit that outputs a second evaluation value indicating the possibility of the presence of the detection target based on the second detection information; a determination unit that determines the presence or absence of the detection target based on at least the first evaluation value, wherein the determination unit determines that the detection target is present on the condition that the first evaluation value is equal to or greater than a first determination threshold; wherein the determination unit determines the presence or absence of the detection target retroactively for a predetermined time on the condition that the second evaluation value is equal to or greater than a first relaxation threshold, and performs a relaxation determination process of determining that the detection target is present on the condition that the first evaluation value is equal to or greater than a second determination threshold lower than the first determination threshold. A detection device.

2. wherein the determination unit determines that the detection target is present on the condition that the second evaluation value is equal to or greater than a third determination threshold; wherein the determination unit determines the presence or absence of the detection target retroactively for a predetermined time on the condition that the first evaluation value is equal to or greater than a second relaxation threshold, and performs the relaxation determination process on the condition that the second evaluation value is equal to or greater than a fourth determination threshold lower than the third determination threshold. The detection device according to claim 1.

3. The detection device according to claim 1, wherein the determination unit performs the relaxation determination process retroactively from the time when it is detected that the second evaluation value is equal to or greater than the first relaxation threshold to one hour before.

4. The detection device according to claim 3, wherein the determination unit performs the relaxation determination process from the time when it is detected that the second evaluation value is equal to or greater than the first relaxation threshold to two hours later.

5. The detection device according to claim 2, wherein the determination unit performs the relaxation determination process retroactively from the time when it is detected that the first evaluation value is equal to or greater than the second relaxation threshold to three hours before.

6. The detection device according to claim 5, wherein the determination unit performs the relaxation determination process from the time when it is detected that the first evaluation value is equal to or greater than the second relaxation threshold to four hours later.

7. wherein the first sensor is at least one of an infrared sensor and a radio wave sensor, and one of a camera sensor; The detection device according to any one of claims 1 to 6, wherein the second sensor is the other of at least one of the infrared sensor and the radio wave sensor, and the camera sensor.

8. The detection device according to any one of claims 1 to 6, wherein the object to be detected is at least one of a human body and a vehicle.

Images