Abnormal sound detection device, elevator device, method, and program

Abstract

To provide an abnormal sound detection device, an elevator device, a method, and a program which reduce misdetection of abnormal sound by ambient sound other than that of a moving body to be monitored.SOLUTION: An abnormal sound detection device according to an embodiment comprises a first sound collection unit, an abnormality detection unit, a position information input unit, a history storage unit, and an integration determination unit. The first sound collection unit is installed in the moving body and collects operation sound of the moving body. The abnormality detection unit detects abnormal sound of the operation sound. The position information input unit accepts input of position information of the moving body. The history storage unit stores history information in which detection results in which abnormal sound has been detected and the position information are associated with each other. The integration determination unit determines presence and absence of abnormality of the moving body by integrating the detection results per each piece of the position information on the basis of the history information.SELECTED DRAWING: Figure 2

Description

Technical Field

[0001] Embodiments of the present invention relate to an abnormal sound detection device, an elevator device, a method, and a program.

Background Art

[0002] In industrial equipment and infrastructure facilities, regular inspections are performed to maintain safety. Industrial equipment and infrastructure facilities often generate abnormal sounds when there are abnormalities. Based on this, there is an abnormal sound detection device that collects the operating sounds of industrial equipment and infrastructure facilities with a microphone and analyzes the state of the operating sounds to detect abnormalities. In this type of abnormal sound detection device, for mobile infrastructure facilities such as elevator devices, abnormalities are detected using a microphone to shorten the downtime caused by abnormalities.

[0003] However, an abnormal sound detection device may misdetect an abnormal sound due to ambient sounds other than the mobile object to be monitored as interference. Such misdetection causes unnecessary downtime of the mobile object to be monitored, so it is desirable to reduce it.

Prior Art Documents

Patent Documents

[0004]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0005] The problem to be solved by the present invention is to provide an abnormal sound detection device, an elevator device, a method, and a program that can reduce the misdetection of abnormal sounds caused by ambient sounds other than the mobile object to be monitored.

Means for Solving the Problems

[0006] The abnormal sound detection device according to the embodiment comprises a first sound collection unit, an abnormality detection unit, a location information input unit, a history storage unit, and an integrated determination unit. The first sound collection unit is installed on a moving object and collects the operating sounds of the moving object. The abnormality detection unit detects abnormal sounds among the operating sounds. The location information input unit receives the location information of the moving object. The history storage unit stores history information associating the detection results of the detected abnormal sounds with the location information. The integrated determination unit determines whether or not there is an abnormality in the moving object by integrating the detection results for each location based on the history information. [Brief explanation of the drawing]

[0007] [Figure 1] A schematic diagram showing the configuration of an elevator system equipped with an abnormal sound detection device according to the first embodiment. [Figure 2] A block diagram showing the configuration of an abnormal sound detection device according to the first embodiment. [Figure 3] A schematic diagram illustrating the history information according to the first embodiment. [Figure 4] A schematic diagram illustrating the integrated information according to the first embodiment. [Figure 5] A flowchart illustrating an example of operation in the first embodiment. [Figure 6] A diagram showing the installation of a microphone according to the second embodiment. [Figure 7] A block diagram showing the configuration of an abnormal sound detection device according to the second embodiment. [Figure 8] A schematic diagram illustrating the surrounding area determination result according to the second embodiment. [Figure 9] A schematic diagram illustrating the history information according to the second embodiment. [Figure 10] A diagram illustrating an example of integrated judgment based on anomaly detection results and surrounding normality judgment results according to the second embodiment. [Figure 11] A flowchart illustrating an example of operation in the second embodiment. [Figure 12] A diagram showing the placement of a microphone according to the third embodiment. [Figure 13] A block diagram showing the configuration of an abnormal sound detection device according to the third embodiment. [Figure 14] A block diagram showing the configuration of an abnormal sound detection device according to the fourth embodiment. [Figure 15] A diagram showing an example of a display screen according to the fourth embodiment. [Figure 16] A flowchart illustrating an example of operation in the fourth embodiment. [Figure 17] A scatter plot for estimating abnormal locations according to the fifth embodiment. [Figure 18] A diagram illustrating the hardware configuration of an abnormal sound detection device according to the sixth embodiment. [Modes for carrying out the invention]

[0008] The embodiments will be described below with reference to the drawings. In the following description, an elevator will be used as an example of a moving object, but it is not limited to this. Any moving device that passes the same location multiple times can be used as a moving object, such as a train, bus, or automated guided vehicle. The multiple travel paths may be the same or different from each other. In addition, "passing" as used here includes cases where a temporary stop is made. Furthermore, if there is a turning point in the travel path, "passing" includes "turning around" at the turning point.

[0009] <First Embodiment> Figure 1 is a schematic diagram showing the configuration of an elevator system equipped with an abnormal sound detection device according to the first embodiment. This elevator system comprises a car 1, which is a moving body; a microphone 2 with a moving body installed in the car 1; a processing board 3; and a normal elevator configuration other than the car 1 (not shown).

[0010] The cage 1 has a space for people to board and alight, and ascends and descends by the operation of a hoisting machine (not shown) or the like according to the operation of people in front of the door on each floor or the operation of people boarding from the door. In FIG. 1, each floor is each of the 1st to 4th floors, but is not limited thereto. For example, the top floor of each floor is not limited to the 4th floor. Similarly, the bottom floor of each floor is not limited to the 1st floor, and may be any underground floor, for example.

[0011] The microphone 2 with a moving body is installed in the cage 1 and collects the operating sound of the cage 1. In this example, two microphones with moving bodies are installed above the cage 1 and inside the cage 1, respectively. As the microphone 2 with a moving body inside the cage 1, an intercom microphone may be used. The microphone 2 with a moving body is an example of the first sound collection unit.

[0012] The processing board 3 notifies the abnormality of the moving body by detecting an abnormal sound from the microphone signal sent from the microphone 2 with a moving body. As the processing board 3, an edge board installed in the cage 1, an edge board fixed and installed at the top floor of the elevator or the like, an external board installed in a server capable of communicating with the elevator device, etc. can be used as appropriate.

[0013] Here, the abnormal sound is detected during the cage operation when the cage 1 ascends and descends, or when the door (door) of the cage 1 stops at a floor and opens and closes. Examples of the abnormal sound during the cage operation include, for example, an abnormal sound due to distortion of the guide rail in the hoistway, an abnormal sound generated when the balance weight, hoisting machine, speed governor, brake, etc. malfunction. Examples of the abnormal sound when the door opens and closes include, for example, an abnormal sound caused by a foreign object being caught between the outer door and the inner door, an abnormal sound caused by a screw loosening and contacting the door, and an abnormal sound caused by the door being rubbed due to the tightening degree of the door packing.

[0014] Note that, not limited thereto, the abnormal sound may include disturbances generated around the cage 1. Examples of the disturbance include, for example, sounds caused by voices or vibrations emitted by people inside the cage, sounds caused by voices, cleaning, vending machine replacement work, building construction, etc. of people on each floor.

[0015] Figure 2 is a block diagram showing an example of the configuration of an abnormal sound detection device according to the first embodiment. This abnormal sound detection device can be implemented using, for example, a PC, a server computer, an edge device, etc. The abnormal sound detection device comprises a microphone with a mobile body 2, an abnormality detection unit 4, a location information input unit 5, a history storage unit 6, an integration unit 7, and an abnormality notification unit 8.

[0016] Here, the portable microphone 2 is as described above.

[0017] The anomaly detection unit 4 detects abnormal sounds from the microphone signal transmitted from the mobile microphone 2. For example, the anomaly detection unit 4 detects abnormal sounds among the operating sounds of the cage 1 collected by the mobile microphone 2. Here, the anomaly detection unit 4 may also detect abnormal sounds by determining whether the collected operating sounds are normal or abnormal. The anomaly detection method of the anomaly detection unit 4 is, for example, a method that detects abnormalities by checking whether the volume of the operating sounds is above a threshold. Alternatively, for example, a rule-based method can be used to detect abnormalities by calculating the similarity between time-frequency domain features obtained by transforming the time signal of the operating sounds using a short-time Fourier transform, etc., and features that are stored in advance as a dictionary of abnormal sound types. As for the similarity, for example, the sum of absolute differences or cosine similarity can be used as appropriate. Other anomaly detection methods that can be used include supervised anomaly detection methods that learn normal and abnormal data in machine learning, and unsupervised anomaly detection methods that learn only normal data and determine what is different from normal. Examples of supervised anomaly detection methods include logistic regression, support vector machines (SVMs), decision trees, and neural network-based methods. Examples of unsupervised anomaly detection methods include neural network-based methods such as autoencoders and clustering techniques.

[0018] The location information input unit 5 inputs the location information of the elevator car 1 to the history storage unit 6. The location information input unit 5 may input the location information of the elevator car 1 at all times, or it may input the location information when an abnormal sound is detected by the abnormality detection unit 4. In the case of an elevator, the location information may include information about the floor, such as the 1st floor and the 2nd floor, and information about the intermediate space between two floors, such as between the 1st and 2nd floors. Alternatively, the location information may include information about the distance, such as 1 meter above the 1st floor. In addition to the location information, the location information input unit 5 may also input elevator control information corresponding to operations inside the elevator car 1 (e.g., destination floor, door open / close).

[0019] As shown in Figure 3, the history storage unit 6 stores history information that associates the detection result of an abnormal sound with location information. The history information may also associate the date and time with the detection result and location information. Furthermore, the history information may also include detection results when the operating sound is normal.

[0020] The integration unit 7 determines whether or not there is an abnormality in the moving object by integrating the detection results for each location based on the history information. For example, the integrated result is obtained by associating the most recent detection count (number of abnormal sound detection results) for each location. As shown in Figure 4, the integrated result 7a may further associate the most recent pass count (sum of the number of normal sound detection results and the number of abnormal sound detection results) with the location information and the most recent detection count. In any case, based on the integrated result, the integration unit 7 determines that there is an abnormality in the moving object if a certain number of detection results are obtained at approximately the same location.

[0021] If the integration unit 7 determines that there is an abnormality in the moving object, the abnormality notification unit 8 notifies a remote monitoring device (not shown) of the abnormality in the moving object.

[0022] Next, the operation of the abnormal sound detection device configured as described above will be explained using the flowchart in Figure 5.

[0023] In step ST1, the mobile microphone 2, which collects sound from the mobile body, collects the operating sound of the cage 1 and transmits a microphone signal, which is a time-series audio signal.

[0024] In step ST2, the anomaly detection unit 4 detects an anomaly from the sound collection results based on the transmitted microphone signal. Specifically, the anomaly detection unit 4 detects an abnormal sound among the collected operating sounds.

[0025] In step ST3, the position information input unit 5 inputs the position information of the moving object, the basket 1.

[0026] In step ST4, the history storage unit 6 stores history information 6a that associates the detection result of detecting an abnormal sound with location information.

[0027] In step ST5, the integration unit 7 integrates the detection results for each location based on the history information 6a. As a result of this integration, the integration unit 7 aggregates the most recent detection count for each location and associates the obtained most recent detection count with the location.

[0028] In step ST6, the integration unit 7 determines, based on the integration results, whether or not an abnormality has been repeatedly detected at the same location. Specifically, the integration unit 7 determines whether or not a certain number of detection results have been obtained at approximately the same location. If a certain number of detection results have been obtained at approximately the same location as a result of this determination (ST6: Yes), it determines that there is an abnormality in the cage 1 as a moving object and proceeds to step ST7. If the result of the determination in step ST6 is No, step ST7 is skipped and the process ends.

[0029] In step ST7, if the abnormality notification unit 8 determines, based on the results of the integration unit 7, that there is an abnormality in the cage 1, it notifies the remote monitoring device (not shown) of the abnormality in the moving body.

[0030] As described above, according to the first embodiment, the mobile microphone 2 is installed in the cage 1 and collects the operating sound of the cage 1. The abnormality detection unit 4 detects abnormal sounds among the operating sound. The location information input unit inputs the location information of the mobile unit. The history storage unit stores history information that associates the detection result of detecting an abnormal sound with the location information. The integration unit 7 determines whether or not there is an abnormality in the mobile unit by integrating the detection results for each location based on the history information. Therefore, false detection of abnormal sounds caused by ambient sounds other than the mobile unit can be reduced. To add to this, the location information of the cage 1 when an abnormal sound is detected from the operating sound of the cage 1 is acquired and stored as history information. Here, disturbances change from time to time and are unlikely to continue to occur in the same location. Considering false detections due to such disturbances, it is possible to determine from the history information, for example, that if an abnormal sound is repeatedly detected in the same location, it is a true abnormality, and if an abnormal sound is detected only once, it is a false detection due to a disturbance.

[0031] Furthermore, according to the first embodiment, the integration unit 7 determines that there is an abnormality in the moving object if a certain number of detection results are obtained at substantially the same location based on the integration results. Therefore, in addition to the effects described above, it is possible to determine that there is an abnormality in the moving object according to a certain number of values ​​as a threshold. For example, the threshold for the most recent number of detections may be a certain number, or the threshold for the most recent number of detections as a percentage of the most recent number of passes may be a certain number. Alternatively, the threshold for the number of detections over a predetermined period may be a certain number. In any case, in addition to the effects described above, it is possible to determine that there is an abnormality in the moving object according to a certain number that is the threshold at the time of determination.

[0032] <Second Embodiment> In the first embodiment, the detection results of abnormal sounds are integrated for each location information of the moving object to determine whether or not there is an abnormality in the moving object.

[0033] In contrast, the second embodiment, as shown in Figures 6 and 7, further uses ambient noise around the moving object to determine whether or not there is an abnormality in the moving object. In Figures 6 and 7, parts corresponding to those in Figures 1 and 2 are denoted by the same reference numerals, and their detailed descriptions are omitted; only the different parts will be described here. Similarly, redundant descriptions will be omitted for each of the following embodiments and modifications.

[0034] As shown in Figure 6, the abnormal sound detection device of the second embodiment is equipped with a fixed microphone 11 for collecting ambient sounds, an ambient model 14a for generating an abnormality level of ambient sounds, and an abnormal sound detection model 9a for generating an abnormality level of operating sounds, compared to Figure 1. The ambient model 14a and the abnormal sound detection model 9a are mounted on the processing board 3.

[0035] The fixed microphone 11 is installed near the door on the 4th floor, which is the top floor, in the surrounding area of ​​the 1st to 4th floors along the elevator shaft, which is the movement path of the elevator car 1, and collects ambient sound at the installed location. Suitable locations near the door include, for example, the location of the floor's call button on the outside of the door, or the elevator shaft side inside the door. From the viewpoint of collecting ambient sound along the elevator car 1's movement path, it is preferable to install the fixed microphone 11 on the floor side. Note that when installing one fixed microphone 11, it is not limited to being near the door on the top floor, but may also be installed near a door on any floor among the 1st to 3rd floors. The fixed microphone 11 is an example of a second sound collection unit that is installed at least at one location in the surrounding area along the movement path of the moving object, and collects ambient sound at the installed location.

[0036] Unlike the mobile microphone 2, the fixed microphone 11 can continuously collect ambient sounds. Therefore, it is possible to collect ambient disturbances that change daily over the long term and train the ambient model 14a. Thus, by training the ambient model 14a with the daily ambient disturbances collected by the fixed microphone 11 and comparing the anomaly score generated by the trained ambient model 14a with a threshold, it is possible to determine whether the ambient sound is a normal disturbance. For example, if the anomaly detection unit 4 falsely detects an abnormal sound due to ambient disturbances, and the ambient sound is determined to be normal, it is possible to determine that a false detection due to normal disturbances has occurred. By adding processing to determine the surrounding conditions, it is possible to further reduce false detections and improve the accuracy of anomaly detection.

[0037] Figure 7 is a block diagram showing an example of the configuration of an abnormal sound detection device according to the second embodiment. In addition to the configuration shown in Figure 2, this abnormal sound detection device includes a detection model storage unit 9, a fixed microphone 11, a learning data storage unit 12, a surrounding model learning unit 13, a surrounding model storage unit 14, and a surrounding normal determination unit 15.

[0038] Here, the detection model storage unit 9 stores an abnormal sound detection model 9a, which is a trained model that generates an abnormality level of the operating sound of the cage 1 collected by the microphone 2 with a mobile body. Accordingly, the abnormality detection unit 4 uses the abnormal sound detection model 9a stored in the detection model storage unit 9 to input the operating sound to the abnormal sound detection model 9a, and determines whether or not an abnormal sound has been detected from the operating sound based on the abnormality level and threshold obtained from the abnormal sound detection model 9a. If an abnormal sound is detected (abnormal sound detection result exists), the operating sound indicates an abnormal state, and if not (no abnormal sound detection result), the operating sound indicates a normal state. The abnormality detection unit 4 also sends information including date and time information, location information, abnormality level of the mobile body, and whether or not a detection result was obtained to the history storage unit 6. The history storage unit 6 stores the sent information as history information. Note that the detection model storage unit 9, abnormal sound detection model 9a, and abnormality detection unit 4 of the second embodiment may also be applied to the first embodiment.

[0039] The fixed microphone 11 is installed near a door on the 4th floor and collects ambient sound at that location. The fixed microphone 11 also collects ambient sound and sends the microphone signal, which is a time-series audio signal, to the learning data storage unit 12 and the ambient normality determination unit 15.

[0040] The learning data storage unit 12 takes ambient sound collected by the fixed microphone 11 as input data and stores learning data in which the degree of abnormality corresponding to the degree to which the collected ambient sound differs from everyday disturbances is output data.

[0041] The ambient model learning unit 13 uses training data to perform machine learning on the ambient model 14a so that it can create a trained model that generates anomaly scores based on ambient sounds. After the machine learning is complete, the ambient model learning unit 13 sends the trained ambient model 14a to the ambient model storage unit 14.

[0042] The ambient model storage unit 14 is provided for each fixed microphone 11 and stores an ambient model 14a, which is a trained model that generates an abnormality level of the ambient sound based on the ambient sound that has been collected.

[0043] The ambient normality determination unit 15 is provided for each fixed microphone 11 and determines whether the ambient sound is normal or abnormal. For example, the ambient normality determination unit 15 inputs the ambient sound into an ambient model 14a, which is a learned model, and determines whether the ambient sound is normal or not based on the degree of abnormality and threshold obtained from the ambient model 14a. In addition, as shown in Figure 8, the ambient normality determination unit 15 sends ambient history information 15a to the history storage unit 6, which associates date and time information indicating the date and time the ambient sound was collected, location information indicating the location where the fixed microphone 11 is installed, the degree of abnormality of the ambient sound, and the ambient determination result. Note that the date and time information and location information of the ambient history information 15a can be added by the history storage unit 6, so sending them to the history storage unit 6 may be omitted. Also, in the second embodiment, the degree of abnormality is not used in processing after the history storage unit 6, so sending it to the history storage unit 6 may be omitted.

[0044] Accordingly, as shown in Figure 9, the history storage unit 6 stores history information 6b, which is obtained by sequentially recording history information based on the mobile microphone 2 and ambient history information 15a based on the fixed microphone 11.

[0045] The integration unit 7 determines whether there is an abnormality in the moving body by further integrating the judgment results of the ambient normality determination unit 15. For example, based on the results of integrating the detection results for each location information, if a certain number of detection results are obtained at approximately the same location, the integration unit 7 determines that there is an abnormality in the moving body's cage 1 by further integrating the judgment results indicating that the ambient sound is abnormal. Alternatively, for example, based on the results of integrating the detection results for each location information, if a certain number of detection results are obtained at approximately the same location, the integration unit 7 suspends the determination that there is an abnormality in the cage 1 and determines that there is a possibility of false detection in the detection results by further integrating the judgment results indicating that the ambient sound is normal. Such a determination is obtained by integrating the abnormal sound detection result based on the microphone attached to the moving body 2 and the normal or abnormal ambient sound judgment result based on the fixed microphone 11, as shown in Figure 10. In Figure 10, the item "Normal" in the column for the microphone attached to the moving body corresponds to no abnormal sound detection result. The item "Abnormal" in the column for the microphone attached to the moving body corresponds to an abnormal sound detection result. For example, if the "abnormal" item in the column for the mobile microphone is merged with the "normal" item in the row for the fixed microphone, the abnormality of the mobile device suggests the possibility of a false detection due to the inclusion of normal ambient sound. Also, if the "abnormal" item in the column for the mobile microphone is merged with the "abnormal" item in the row for the fixed microphone, the abnormality of the ambient sound suggests the detection is due to the inclusion of abnormal sound from the mobile device, allowing for a more accurate determination that an abnormality has occurred in the mobile device. Furthermore, if the "normal" item in the column for the mobile microphone is merged with the "abnormal" item in the row for the fixed microphone, it can be determined that a new disturbance was introduced into the ambient sound, but was not picked up by the mobile microphone 2. In this case, the accuracy of the ambient model 14a can be improved by adding the new disturbance to the training data of the ambient model 14a.

[0046] The other configurations are the same as in the first embodiment.

[0047] Next, the operation of the abnormal sound detection device configured as described above will be explained using the flowchart in Figure 11. In the following explanation, the abnormal sound detection model 9a and the surrounding model 14a are assumed to be trained models in which machine learning has been completed.

[0048] First, the mobile microphone 2 collects the sound of the cage in operation. The anomaly detection unit 4 inputs the operation sound to the anomaly sound detection model 9a and determines whether or not an anomaly has been detected from the operation sound based on the degree of anomaly obtained from the anomaly sound detection model 9a. If an anomaly has been detected (an anomaly sound detection result is found), the operation sound indicates an abnormal state; otherwise (no anomaly sound detection result is found), the operation sound indicates a normal state. The anomaly detection unit 4 also sends information including date and time information, the degree of anomaly of the mobile object, and whether or not a detection result was found to the history storage unit 6.

[0049] Meanwhile, the fixed microphone 11 collects ambient sound near the door on the 4th floor. The ambient normality determination unit 15 inputs the collected ambient sound to the ambient model 14a and determines whether the ambient sound is normal based on the degree of abnormality obtained from the ambient model 14a. The ambient normality determination unit 15 also sends ambient history information 15a to the history storage unit 6, which associates date and time information indicating the date and time the ambient sound was collected, location information indicating the location where the fixed microphone 11 is installed, the degree of abnormality of the ambient sound, and the ambient determination result. The history storage unit 6 stores the information sent from the abnormality detection unit 4, the location information input from the location information input unit 5, and the ambient history information 15a as history information 6b.

[0050] As shown in Figures 10 and 11, the integration unit 7 integrates the detection results for each location based on the history information, and further integrates the judgment results of the surrounding normality determination unit 15 to determine whether or not there is an abnormality in the cage 1 as a moving object (steps ST11 to ST17). Steps ST11 to ST17 are an example of the procedure for making the determination shown in Figure 10, and can be modified as appropriate.

[0051] In step ST11, the integration unit 7 determines whether or not an abnormality of the moving object has been detected based on the detection results included in the history information. If an abnormality of the moving object has been detected (if there is a detection result), the process proceeds to step ST12. If the result of the determination in step ST11 is negative, the process proceeds to step ST15.

[0052] In step ST12, the integration unit 7 determines whether or not an abnormality in ambient sound has been detected based on the ambient sound determination result included in the history information. If an abnormality in ambient sound has been detected, the unit proceeds to step ST13. If the result of the determination in step ST12 is negative, the unit proceeds to step ST14.

[0053] In step ST13, the integration unit 7 determines whether there is a malfunction in the mobile body based on the abnormality in the operating sound of the cage 1 corresponding to the row of microphones 2 with mobile bodies in Figure 10, and the abnormality in the ambient sound corresponding to the row of fixed microphones 11, and sends the determination result to the malfunction notification unit 8. The malfunction notification unit 8 notifies the remote monitoring device (not shown) of the malfunction in the mobile body and terminates the process.

[0054] In step ST14, the integration unit 7 determines whether there is a possibility of a false detection of the abnormal operating sound of the cage 1 corresponding to the row of microphones 2 with moving parts in Figure 10, and whether there is a normal ambient sound corresponding to the row of fixed microphones 11, and sends the determination result to the abnormality notification unit 8. The integration unit 7 also reserves judgment on whether there is an abnormality in the cage 1 as a moving part. The abnormality notification unit 8 notifies the remote monitoring device that there is a possibility of a false detection of the abnormal operating sound and terminates the process.

[0055] In step ST15, the integration unit 7 determines whether or not an abnormality in ambient sound has been detected based on the ambient sound determination result included in the history information. If an abnormality in ambient sound has been detected, the process proceeds to step ST16. If the result of the determination in step ST15 is negative, the process proceeds to step ST17.

[0056] In step ST16, the integration unit 7 determines that the cause of the abnormal ambient noise is a new disturbance on the floor, based on the normal operating sound of the cage 1 for the row of microphones with moving parts and the abnormal ambient noise for the row of fixed microphones in Figure 10, and sends the determination result to the abnormality notification unit 8. The abnormality notification unit 8 notifies the remote monitoring device that the cause of the abnormal ambient noise is a new disturbance on the floor and terminates the process.

[0057] In step ST17, the integration unit 7 determines the normal operation of the mobile body based on the normal operation sound of the cage 1 corresponding to the row of microphones 2 with mobile bodies in Figure 10, and the normal ambient sound corresponding to the row of fixed microphones 11, and then terminates the process.

[0058] As described above, according to the second embodiment, the fixed microphone 11 is installed in the vicinity of the door on the 4th floor within the surrounding area along the movement path of the cage 1, and collects ambient sound at the installed location. An ambient normality determination unit 15 is provided for each fixed microphone 11 and determines whether the ambient sound is normal or abnormal. The integration unit 7 further integrates the determination results of the ambient normality determination unit 15 to determine whether there is an abnormality in the cage 1 as a moving object. Therefore, in addition to the effects described above, the accuracy of determining whether there is an abnormality in the moving object can be further improved by further integrating the ambient sound determination results.

[0059] Furthermore, according to the second embodiment, the integration unit 7 determines whether there is an abnormality in the cage 1 as a moving object by further integrating the judgment results indicating an abnormality in ambient sound when a certain number of detection results exceeding a certain number are obtained at substantially the same location, based on the integration results.Therefore, in addition to the effects described above, the accuracy of determining whether there is an abnormality in the moving object can be further improved by further integrating the judgment results of ambient sound.

[0060] Furthermore, according to the second embodiment, if a certain number of detection results are obtained at approximately the same location based on the integrated results, the determination that the ambient sound is normal is further integrated, thereby suspending the determination that there is an abnormality in the mobile body, which is the cage 1, and determining that there is a possibility of false detection in the detection results. Therefore, in addition to the effects described above, by further integrating the ambient sound determination results, it is possible to determine that there is a possibility of false detection in the detection result of abnormal sound of the mobile body.

[0061] Furthermore, according to the second embodiment, the ambient model storage unit 14 is provided for each fixed microphone 11 and stores an ambient model 14a, which is a trained model that generates an abnormality score of the ambient sound based on the ambient sound that has been collected. The ambient normality determination unit 15 inputs the ambient sound to the ambient model 14a, which is the trained model, and determines whether the ambient sound is in a normal state or not based on the abnormality score obtained from the ambient model 14a. Therefore, in addition to the effects described above, an abnormality score of the ambient sound can be obtained.

[0062] <Third Embodiment> The second embodiment determines whether or not there is an abnormality in the moving object using ambient sound from a single fixed microphone 11.

[0063] In contrast, the third embodiment, as shown in Figures 12 and 13, determines whether or not there is an abnormality in the moving object using ambient sound from two fixed microphones 11 located closest to the location information of the abnormal sound detection result. For example, in an elevator system, fixed microphones 11 are installed on each floor, and when an abnormal sound is detected while the elevator car 1 is moving between the second and third floors, the presence or absence of an abnormality in the moving object is determined using ambient sound from two fixed microphones 11 on the second and third floors that are closest to the location information corresponding to the detection result. Although four fixed microphones 11 are used in the example shown in Figures 12 and 13, the number of fixed microphones 11 is not limited to four.

[0064] Figure 13 is a block diagram showing an example of the configuration of an abnormal sound detection device according to the third embodiment. This abnormal sound detection device has a configuration that includes multiple fixed microphones 11, learning data storage units 12, ambient model learning units 13, ambient model storage units 14, and ambient normality determination units 15, as shown in Figure 7. Here, the fixed microphones 11, learning data storage units 12, ambient model learning units 13, ambient model storage units 14, and ambient normality determination units 15 are provided in each of the multiple ambient sound determination units 10_1F to 10_4F. Note that each of the ambient sound determination units 10_1F to 10_4F represents the 1F to 4F floors where the fixed microphones 11 are provided, as indicated by the last two digits of the reference code.

[0065] Accordingly, in addition to the functions described above, the integration unit 7 further integrates the ambient sound determination results from the location closest to the location of the abnormal sound detection result, based on the location information. As a result, the integration unit 7 determines whether or not there is an abnormality in the moving object.

[0066] The other configurations are the same as in the second embodiment.

[0067] With the above configuration, the effects of the second embodiment can be obtained on each floor where the fixed microphone 11 is installed. In addition, when the mobile cage 1 is located between two floors, the integration unit 7 further integrates the ambient sound judgment results of the two positions closest to the location of the abnormal sound detection result based on the position information, and determines whether or not there is an abnormality in the mobile body. Therefore, in addition to the effects described above, abnormality can be determined in the ambient sound between the two floors without installing a fixed microphone between the two floors. Thus, false detections can be reduced more efficiently compared to the case where a fixed microphone is also installed between the two floors.

[0068] <Fourth Embodiment> In the third embodiment, the presence or absence of abnormality in a moving object is determined using ambient sound from the two fixed microphones 11 at the positions closest to each other in terms of location information.

[0069] In contrast, the fourth embodiment determines whether or not to update the ambient model 14a based on the ambient sound determination result. Furthermore, the fourth embodiment displays information regarding abnormal sounds and information regarding abnormal ambient sounds on the display unit.

[0070] Figure 14 is a block diagram showing an example of the configuration of an abnormal sound detection device according to the fourth embodiment. Compared to the configuration shown in Figure 13, this abnormal sound detection device further includes a display control unit 20, a display unit 21, and a model update determination unit 22.

[0071] Here, the display control unit 20 displays the information contained in the history information on the display unit 21 based on the history information in the history storage unit 6. For example, as shown in Figure 15, the display control unit 20 may display information regarding abnormal sounds and information regarding abnormal ambient sounds on the display unit 21 based on the history information. In Figure 15, the information regarding abnormal sounds includes the date and time the abnormal sound was detected, the degree of abnormality, and an indication that a detection result was found (red display on the microphone), and is displayed at the detection location on the movement path of the cage 1. The information regarding abnormal ambient sounds includes date and time information, the degree of abnormality, and an indication of normal judgment (green display on the microphone) or an indication of abnormal judgment (red display on the microphone), and is displayed for each floor. To add to this, if there is a detection result for abnormal sounds and an abnormal judgment for ambient sounds, as shown in Figure 15, the red display of the fixed microphone 11 is placed near the red display of the mobile microphone 2. That is, it displays the location of one instance of abnormal detection based on the mobile microphone 2, and the history of the fixed microphone 11 for the same date and time corresponding to the history. In this case, the ambient sound detection result of the fixed microphone 11 can be seen at a glance, making it easier for users such as maintenance personnel to determine whether the detection result of abnormal sound from the moving object is a false detection or a true abnormality. In addition, the display control unit 20 may display information about abnormal sound on the right side when the cage 1 is moving, and on the left side when it is stopped, such as when the door is opened or closed. In other words, the display control unit 20 may change the display position of the information about abnormal sound depending on the operating state of the cage 1.

[0072] The display unit 21 is a display controlled by the display control unit 20 and is capable of displaying any information. The display unit 21 may be implemented as, for example, the display of a tablet terminal carried by a maintenance worker, or as a display installed inside the cage 1. The display control unit 20 and the display unit 21 may be applied to the first or second embodiment.

[0073] The model update determination unit 22 determines whether or not to update the trained surrounding model 14a based on the surrounding judgment results in the history information. For example, if the frequency of abnormal judgment results for the surrounding model 14a of a certain floor is high, the model update determination unit 22 can determine that the disturbances on that floor have changed since training, and that it is a floor where judgment by the trained model is difficult. A state that has changed since training includes, for example, a state in which the type of disturbance has changed in a disturbance that has various types, or a state in which the disturbance has a rapid change over time. In this case, it is expected that updating the surrounding model 14a of that floor will improve accuracy. In addition, the surrounding model 14a may be updated when a certain amount of time has elapsed or when a new disturbance on the floor is detected. Additional training and retraining can be used as appropriate methods for updating the surrounding model 14a. Furthermore, the model update determination unit 22 may also be applied to the second embodiment.

[0074] The other configurations are the same as in the third embodiment.

[0075] Next, the operation of the abnormal sound detection device configured as described above will be explained using the flowchart in Figure 16. In Figure 16, steps ST11 to ST17 are executed in the same manner as described above. At this time, the display control unit 20 displays the information contained in the history information on the display unit 21 based on the history information in the history storage unit 6, as shown in Figure 15. This allows users such as maintenance personnel to visually grasp information regarding abnormal sounds and abnormal ambient sounds for each location of the elevator equipment.

[0076] On the other hand, after step ST16, in step ST18, the model update determination unit 22 determines whether the frequency of abnormality in the surrounding judgment results in the history information is high or low. If the frequency of abnormality is high as a result of this determination, the process proceeds to step ST19; otherwise, the process ends.

[0077] In step 19, the model update determination unit 22 determines that the surrounding model 14a should be updated and sends this determination result to the abnormality notification unit 8. The abnormality notification unit 8 notifies the remote monitoring device of the determination result that the surrounding model 14a should be updated, and terminates the process.

[0078] As described above, according to the fourth embodiment, the model update determination unit 22 determines whether or not to update the surrounding model 14a based on the determination result in the history information. Therefore, in addition to the effects described above, it is possible to determine whether to update the surrounding model 14a when the surrounding model 14a no longer matches the current state, thereby reducing false detections caused by surrounding models 14a that do not match the current state.

[0079] Furthermore, according to the fourth embodiment, the display control unit 20 displays information regarding abnormal sounds and information regarding abnormal ambient sounds on the display unit 21 based on the history information. Therefore, in addition to the effects described above, users such as maintenance personnel can visually grasp information regarding abnormal sounds and information regarding abnormal ambient sounds from the display unit 21.

[0080] <Fifth Embodiment> The fifth embodiment is a modification of the first to fourth embodiments, and not only determines whether or not there is an abnormality in the moving object, but also integrates location information and abnormality detection results to estimate the location of the abnormality.

[0081] Specifically, in addition to the configuration of any of the first to fourth embodiments, the integration unit 7, when it determines that there is an abnormality in the moving object, the cage 1, outputs information regarding the distribution of abnormal sound detection results relative to the position of the cage 1, based on the history information.

[0082] Figure 17 is a scatter plot for estimating the location of an anomaly in the fifth embodiment. This scatter plot is an example in which the degree of anomaly detected by the mobile microphone 2 is shown on the horizontal axis, and the location information at the time the anomaly was detected is shown on the vertical axis. In Figure 17, high anomaly degrees are shown multiple times on the 4th floor. This allows maintenance personnel to determine that an anomaly has occurred near the 4th floor and to estimate, for example, a malfunction of the speed governor fixedly installed at the top of the 4th floor. Also, if high anomaly degrees occur frequently from the 2nd to the 3rd floor, it can be estimated, for example, that an anomaly is being detected when counterweights pass each other.

[0083] As described above, according to the fifth embodiment, when the integration unit 7 determines that there is an abnormality in the moving body, it outputs information regarding the distribution of detection results relative to the position of the moving body based on the history information. Therefore, in addition to the effects described above, the location of the abnormality in the moving body can be estimated based on the distribution of abnormal sound detection results.

[0084] In the fifth embodiment, a maintenance worker estimated the location of the anomaly by looking at a scatter plot, but the system is not limited to this. For example, anomaly candidate information, which associates location information with candidate anomaly locations, may be stored in memory. In this case, the integration unit 7 can read candidate anomaly locations from the anomaly candidate information in memory based on location information that has displayed a high degree of anomaly multiple times, and display them on the display unit 21. This makes it possible to present candidate anomaly locations to less experienced maintenance workers, in addition to the effects of the fifth embodiment.

[0085] <Sixth Embodiment> The sixth embodiment is a specific example of the first to fifth embodiments, and is a form in which the aforementioned abnormal sound detection device is implemented using a computer.

[0086] Figure 18 is a block diagram illustrating the hardware configuration of an abnormal sound detection device according to the sixth embodiment. This abnormal sound detection device 30 includes a CPU (Central Processing Unit) 31, RAM (Random Access Memory) 32, program memory 33, auxiliary storage device 34, and input / output interface 35 as hardware. The CPU 31 communicates with the RAM 32, program memory 33, auxiliary storage device 34, and input / output interface 35 via a bus.

[0087] The CPU 31 is an example of a general-purpose processor. The RAM 32 is used by the CPU 31 as working memory. The RAM 32 includes volatile memory such as SDRAM (Synchronous Dynamic Random Access Memory). The program memory 33 stores programs for implementing each part according to each embodiment. This program may be, for example, a program for the computer to implement each function of the abnormal sound detection device described above. In addition, the program memory 33 may be, for example, ROM (Read-Only Memory), a part of the auxiliary storage device 34, or a combination thereof. The auxiliary storage device 34 stores data non-temporarily. The auxiliary storage device 34 includes non-volatile memory such as an HDD (hard disk drive) or SSD (solid state drive).

[0088] The input / output interface 35 is an interface for connecting to other devices. The input / output interface 35 is used, for example, to connect to a portable microphone 2, a fixed microphone 11, a keyboard, a mouse, and a display unit 21.

[0089] The program stored in the program memory 33 includes computer executable instructions. When the program (computer executable instructions) is executed by the CPU 31, which is a processing circuit, it causes the CPU 31 to perform predetermined processing. For example, when the program is executed by the CPU 31, it causes the CPU 31 to perform a series of processes described with respect to each part in Figures 2, 7, 13, and 14. For example, when the computer executable instructions included in the program are executed by the CPU 31, they cause the CPU 31 to perform an abnormal sound detection method. The abnormal sound detection method may include each step corresponding to each function of the abnormal sound detection device described above. For example, the abnormal sound detection method may include collecting the operating sound of the mobile body using a microphone installed on the mobile body, detecting abnormal sounds among the operating sounds, inputting the location information of the mobile body, storing history information in memory that associates the detection result of detecting an abnormal sound with the location information, and determining whether or not there is an abnormality in the mobile body by integrating the detection results for each location information based on the history information. The abnormal sound detection method may also appropriately include each step shown in Figures 5, 11, and 16.

[0090] The program may be provided to the abnormal sound detection device 30 in a state where it is stored on a computer-readable storage medium. In this case, for example, the abnormal sound detection device 30 further includes a drive (not shown) for reading data from the storage medium and retrieving the program from the storage medium. Suitable storage media include, for example, magnetic disks, optical disks (CD-ROM, CD-R, DVD-ROM, DVD-R, etc.), magneto-optical disks (MO, etc.), and semiconductor memory. The storage medium may also be called a non-transitory computer-readable storage medium. Alternatively, the program may be stored on a server on a communication network, and the abnormal sound detection device 30 may download the program from the server using the input / output interface 35.

[0091] The processing circuit that executes the program is not limited to a general-purpose hardware processor such as a CPU 31, but may also use a dedicated hardware processor such as an ASIC (Application Specific Integrated Circuit). The term "processing circuit" (processing unit) includes at least one general-purpose hardware processor, at least one dedicated hardware processor, or a combination of at least one general-purpose hardware processor and at least one dedicated hardware processor. In the example shown in Figure 18, the CPU 31, RAM 32, and program memory 33 correspond to the processing circuit.

[0092] (Variations of each embodiment) Furthermore, the abnormal sound detection devices according to each embodiment and each modified example are not limited to elevators, but can be installed on any moving body such as a railway, bus, or automated guided vehicle. Also, each embodiment and each modified example may be expressed as a moving body device (e.g., an elevator device) equipped with an abnormal sound detection device. Similarly, each embodiment and each modified example may be expressed as an abnormal sound detection method or program that includes each step of the abnormal sound detection device described above.

[0093] According to the at least one embodiment described above, it is possible to reduce false detections of abnormal sounds caused by ambient noise other than that of the moving object being monitored. This is also true for the at least one modified example described above.

[0094] While several embodiments of the present invention have been described, these embodiments are presented as examples only and are not intended to limit the scope of the invention. These embodiments can be carried out in a variety of other forms, and various omissions, substitutions, and modifications can be made without departing from the spirit of the invention. These embodiments and their variations are included in the scope and spirit of the invention, as well as in the claims and their equivalents. [Explanation of Symbols]

[0095] 1...Basket, 2...Microphone with moving body, 3...Processing board, 4...Anomaly detection unit, 5...Location information input unit, 6...History storage unit, 6a,6b...History information, 7...Integration unit, 7a...Result, 8...Anomaly notification unit, 9...Detection model storage unit, 9a...Anomaly sound detection model, 10_1F~10_4F...Ambient sound determination unit, 11...Fixed microphone, 12...Learning data storage unit, 13...Ambient model learning unit, 14...Ambient model storage unit, 15...Ambient normal determination unit, 15a...Ambient history information, 20...Display control unit, 21...Display unit, 22...Model update determination unit, 30...Anomaly sound detection device, 31...CPU, 32...RAM, 33...Program memory, 34...Auxiliary storage device, 35...Input / output interface.

Claims

1. A first sound collection unit is installed on the mobile body and collects the operating sound of the mobile body, An abnormality detection unit that detects abnormal sounds among the aforementioned operating sounds, A location information input unit for inputting the location information of the aforementioned moving object, A history storage unit stores history information that associates the detection result of detecting the abnormal sound with the location information, An integration unit determines whether or not there is an abnormality in the moving object by integrating the detection results for each location based on the historical information, Equipped with, The integration unit determines that there is an abnormality in the moving body based on the integration result if the number of times the most recent abnormal sound was detected relative to the number of times the body passed through substantially the same location exceeds a first threshold, or if the number of times the abnormal sound was detected at substantially the same location over a predetermined period exceeds a second threshold.

2. A second sound collection unit is provided at at least one position in the surrounding area along the movement path of the moving body, and collects ambient sound at the position where it is provided. Each of the second sound collection units is provided with an ambient normality determination unit that determines whether the ambient sound is normal or abnormal, Furthermore, The abnormal sound detection device according to claim 1, wherein the history storage unit further stores the determination result of the surrounding normality determination unit as history information, and the integration unit determines whether or not there is an abnormality in the moving body by further integrating the determination result of the surrounding normality determination unit based on the history information.

3. A first sound collection unit installed on a mobile body for collecting the operating sound of the mobile body, An abnormality detection unit that detects abnormal sounds among the aforementioned operating sounds, A location information input unit for inputting the location information of the aforementioned moving object, A history storage unit stores history information that associates the detection result of detecting the abnormal sound with the location information, An integration unit determines whether or not there is an abnormality in the moving object by integrating the detection results for each location based on the historical information, A second sound collection unit is provided at at least one position in the surrounding area along the movement path of the moving body, and collects ambient sound at the position where it is provided. Each of the second sound collection units is provided with an ambient normality determination unit that determines whether the ambient sound is normal or abnormal, Equipped with, The history storage unit further stores the determination result of the surrounding normality determination unit as history information, and the integration unit determines whether or not there is an abnormality in the moving body by further integrating the determination result of the surrounding normality determination unit based on the history information. The integration unit determines whether there is an abnormality in the moving body by further integrating the determination results indicating an abnormality in the ambient sound when a certain number of detection results exceeding a certain number are obtained at substantially the same location, based on the results of the integration.

4. A first sound collection unit installed on a mobile body for collecting the operating sound of the mobile body, An abnormality detection unit that detects abnormal sounds among the aforementioned operating sounds, A location information input unit for inputting the location information of the aforementioned moving object, A history storage unit stores history information that associates the detection result of detecting the abnormal sound with the location information, An integration unit determines whether or not there is an abnormality in the moving object by integrating the detection results for each location based on the historical information, A second sound collection unit is provided at at least one position in the surrounding area along the movement path of the moving body, and collects ambient sound at the position where it is provided. Each of the second sound collection units is provided with an ambient normality determination unit that determines whether the ambient sound is normal or abnormal, Equipped with, The history storage unit further stores the determination result of the surrounding normality determination unit as history information, and the integration unit determines whether or not there is an abnormality in the moving body by further integrating the determination result of the surrounding normality determination unit based on the history information. The integration unit, based on the results of the integration, determines that there is a possibility of false detection in the detection results when a certain number of the detection results are obtained at substantially the same location, further integrates the determination results indicating that the ambient sound is normal, thereby suspending the determination that there is an abnormality in the moving body, and determining that there is a possibility of false detection in the detection results.

5. A model storage unit is provided for each of the second sound collection units, which stores a trained model that generates an abnormality level of the ambient sound based on the ambient sound collected. Furthermore, The abnormal sound detection device according to claim 2, wherein the ambient normality determination unit inputs the ambient sound to the learned model and determines whether the ambient sound is in a normal state based on the degree of abnormality obtained from the learned model.

6. A first sound collection unit installed on a mobile body for collecting the operating sound of the mobile body, An abnormality detection unit that detects abnormal sounds among the aforementioned operating sounds, A location information input unit for inputting the location information of the aforementioned moving object, A history storage unit stores history information that associates the detection result of detecting the abnormal sound with the location information, An integration unit determines whether or not there is an abnormality in the moving object by integrating the detection results for each location based on the historical information, Multiple second sound-collecting units are provided at multiple locations within the surrounding area along the movement path of the aforementioned moving body, and each unit collects ambient sound at the location where it is provided. Each of the second sound collection units is provided with a plurality of ambient normality determination units that determine whether the ambient sound is normal or abnormal, Equipped with, The history storage unit further stores the determination result of the surrounding normality determination unit as history information, and the integration unit determines whether or not there is an abnormality in the moving body by further integrating the determination result of the surrounding normality determination unit based on the history information. The integration unit further integrates the determination results of the two positions closest to the detection result position based on the position information, thereby providing an abnormal sound detection device.

7. A first sound collection unit installed on a mobile body for collecting the operating sound of the mobile body, An abnormality detection unit that detects abnormal sounds among the aforementioned operating sounds, A location information input unit for inputting the location information of the aforementioned moving object, A history storage unit stores history information that associates the detection result of detecting the abnormal sound with the location information, An integration unit determines whether or not there is an abnormality in the moving object by integrating the detection results for each location based on the historical information, A second sound collection unit is provided at at least one position in the surrounding area along the movement path of the moving body, and collects ambient sound at the position where it is provided. Each of the second sound collection units is provided with an ambient normality determination unit that determines whether the ambient sound is normal or abnormal, A model storage unit is provided for each of the two sound collection units, which stores a trained model that generates an abnormality level of the ambient sound based on the ambient sound collected, Model update determination unit, Equipped with, The history storage unit further stores the determination result of the surrounding normality determination unit as history information, and the integration unit determines whether or not there is an abnormality in the moving body by further integrating the determination result of the surrounding normality determination unit based on the history information. The ambient normality determination unit inputs the ambient sound to the trained model and determines whether the ambient sound is normal based on the abnormality score obtained from the trained model. The model update determination unit determines whether or not to update the trained model based on the determination result in the history information; this is an abnormal sound detection device.

8. The abnormal sound detection device according to claim 2, further comprising a display control unit that displays information regarding the abnormal sound and information regarding the abnormal ambient sound on a display unit based on the history information.

9. A first sound collection unit installed on a mobile body for collecting the operating sound of the mobile body, An abnormality detection unit that detects abnormal sounds among the aforementioned operating sounds, A location information input unit for inputting the location information of the aforementioned moving object, A history storage unit stores history information that associates the detection result of detecting the abnormal sound with the location information, An integration unit determines whether or not there is an abnormality in the moving object by integrating the detection results for each location based on the historical information, Equipped with, The integration unit, when it determines that there is an abnormality in the moving body, outputs information regarding the distribution of the detection results with respect to the position of the moving body, based on the history information, as an abnormal sound detection device.

10. An elevator system equipped with an abnormal sound detection device according to any one of claims 1 to 8.

11. A microphone installed on the mobile unit collects the operating sound of the mobile unit, To detect abnormal sounds among the aforementioned operating sounds, Inputting the location information of the aforementioned moving object, The detection result of detecting the aforementioned abnormal sound and the location information are associated with historical information stored in memory. Based on the historical information, the detection results are integrated for each location to determine whether or not there is an abnormality in the moving object. Equipped with, An abnormal sound detection method, wherein the determination includes determining whether the moving body is abnormal if, based on the integrated results, the number of times the most recent abnormal sound was detected relative to the number of times the body passed through substantially the same location exceeds a first threshold, or if the number of times the abnormal sound was detected at substantially the same location over a predetermined period exceeds a second threshold.

12. A function that collects the operating sound of the mobile unit using a microphone installed on the mobile unit. A function to detect abnormal sounds among the aforementioned operating sounds, A function for inputting the position information of the aforementioned moving object. A function to store in memory historical information that associates the detection result of the abnormal sound with the location information. Based on the aforementioned history information, a function is provided to determine whether or not there is an abnormality in the moving object by integrating the detection results for each location information. A program that enables a computer to implement this, The program includes a function that determines whether there is an abnormality in the moving object based on the integrated results, if the number of times the most recent abnormal sound was detected relative to the number of times the object passed through substantially the same location exceeds a first threshold, or if the number of times the abnormal sound was detected at substantially the same location over a predetermined period exceeds a second threshold.

Images