Traveling vehicle detection device, traveling vehicle detection method, and program

Abstract

Provided is a traveling vehicle detecting device (100) that comprises: a measurement unit (208) for measuring a continuous physical amount in a prescribed place by using optical fiber sensors laid along a road; a detection unit (209) for detecting a temporal change pattern of the physical amount from a measurement result of the measurement unit; and a determination unit (210) for determining the presence or absence of a traveling vehicle on the basis of the change pattern of the physical amount. Here, the physical amount is vibration intensity, and the change pattern is change in absolute value, change in difference, change in proportion, or change in shape of a graph.

Description

Moving vehicle detection device, moving vehicle detection method, and non-transitory computer-readable medium

[0001] The present invention relates to a moving vehicle detection device, a moving vehicle detection method, and a non-transitory computer-readable medium.

[0002] A method for detecting moving vehicles using optical fiber sensing has been developed. Patent Document 1 discloses providing an optical fiber sensor for traffic monitoring, comprising a mold including an elongated plate and an optical fiber wound on at least one surface of the elongated plate. In Patent Document 1, the elongated plate is flexible in a direction perpendicular to at least one surface, and traffic passes over the optical fiber sensor so as to cause a change in at least one predetermined characteristic of the optical signal transmitted by the optical fiber sensor. Patent Document 1 discloses that the sensor has a narrow width that allows it to be easily positioned within the road surface, high flexibility that allows it to conform to the road surface, and good cross-axis sensitivity rejection.

[0003] Japanese Patent Application Laid-Open No. 2004-527030

[0004] Patent Document 1 discloses improving the sensitivity of optical fibers from a physical aspect. However, there are problems that cannot be solved by simply increasing the sensitivity. Therefore, there is room for improving the sensitivity of optical fibers from a software aspect.

[0005] An object of the present disclosure is to provide a moving vehicle detection device, a moving vehicle detection method, and a non-transitory computer-readable medium that improve sensitivity from a software perspective in optical fiber sensing of moving vehicles.

[0006] The moving vehicle detection device of the present disclosure is a moving vehicle detection device that includes a measurement unit that measures continuous physical quantities at a predetermined location using optical fiber sensors laid along a road, a detection unit that detects a change pattern in the physical quantities over time from the measurement results of the measurement unit, and a determination unit that determines the presence or absence of a moving vehicle based on the change pattern in the physical quantities.

[0007] The moving vehicle detection method of the present disclosure comprises the steps of: measuring continuous physical quantities at a predetermined location using optical fiber sensors laid along a road; detecting a pattern of change in the physical quantities over time from the measurement results of the measuring step; and determining the presence or absence of a moving vehicle based on the pattern of change in the physical quantities.

[0008] The non-transitory computer-readable medium of the present disclosure is a non-transitory computer-readable medium that records a program that causes a moving vehicle detection device to execute the following steps: measuring continuous physical quantities at a predetermined location using optical fiber sensors laid along a road; detecting a pattern of change in the physical quantities over time from the measurement results of the measuring step; and determining the presence or absence of a moving vehicle based on the pattern of change in the physical quantities.

[0009] The present disclosure makes it possible to provide a moving vehicle detection device that improves sensitivity from a software perspective in optical fiber sensing of moving vehicles.

[0010] 1 is a schematic diagram of a running vehicle detection device according to an embodiment. FIG. 1 is a block diagram of a running vehicle detection device according to an embodiment. FIG. 2 is a diagram showing measurement results at one measurement point according to an embodiment. FIG. 3 is a diagram showing measurement points at multiple locations according to an embodiment. FIG. 4 is a block diagram of a configuration for determining a threshold value for a feature quantity according to a first embodiment. FIG. 5 is a block diagram of a configuration for determining a threshold value for a feature quantity according to a second embodiment. FIG. 6 is a block diagram of a configuration for modeling and utilizing a feature quantity using a learning device according to a third embodiment. FIG. 7 is a block diagram of a configuration for modeling and utilizing a feature quantity using a learning device according to a fourth embodiment. FIG. 8 is a diagram showing Table 1 according to an example. FIG. 9 is a diagram showing Table 1 according to an example. FIG. 10 is a block diagram of a running vehicle detection device according to the present disclosure. FIG. 11 is a schematic diagram of a related optical fiber sensing device. FIG. 12 is a diagram showing measurement results using optical fiber sensing when no vehicles pass. FIG. 13 is a diagram showing measurement results using optical fiber sensing when vehicles of different models pass. FIG. 14 is a diagram showing measurement results using optical fiber sensing when vehicles of different models pass. FIG. 15 is a diagram showing measurement results using optical fiber sensing when vehicles of different lanes pass. FIG. 16 is a diagram showing measurement results of frequency characteristics using optical fiber sensing when vehicles of different models pass.

[0011] Embodiments Hereinafter, embodiments of the present invention will be described with reference to the drawings. However, the invention according to the claims is not limited to the following embodiments. Furthermore, not all of the configurations described in the embodiments are necessarily essential as means for solving the problems. For clarity of explanation, the following description and drawings have been omitted and simplified as appropriate. In each drawing, the same elements are given the same reference numerals, and repeated explanations are omitted as necessary.

[0012] (Related Optical Fiber Sensing Device) Fig. 12 is a schematic diagram of a related optical fiber sensing device. Fig. 13 is a diagram showing measurement results using optical fiber sensing when no vehicles are passing. Fig. 14 is a diagram showing measurement results using optical fiber sensing when vehicles are passing. Fig. 15 is a diagram showing measurement results using optical fiber sensing when vehicles of different models are passing. Fig. 16 is a diagram showing measurement results using optical fiber sensing when vehicles in different lanes are passing. Fig. 17 shows measurement results of frequency characteristics using optical fiber sensing when vehicles of different models are passing. Related optical fiber sensing devices will be described with reference to Figs. 12 to 17.

[0013] 12, an optical fiber sensing device 1200 includes an optical fiber cable 1201 laid along a road and a sensing device 1202 connected to the end of the optical fiber cable. Probe light, such as laser light, emitted from the sensing device 1202 passes through the optical fiber cable.

[0014] Now, consider the case where a car is traveling on a road. When the car passes a measurement point of interest, vibrations occur in the optical fiber cable 1201. This causes changes in the phase and intensity of the backscattered light passing through the optical fiber cable 1201 at the measurement point of interest. By detecting the phase and intensity of this backscattered light, the optical fiber sensing device 1200 senses that vibrations have occurred in the optical fiber. Such an optical fiber sensor is called a distributed acoustic sensor (DAS).

[0015] As shown in Fig. 13, when no vehicle passes at the measurement point of interest, the intensity of the vibration of the optical fiber does not exceed the threshold and fluctuates as noise. As shown in Fig. 14, when a vehicle passes at the measurement point of interest, the vibration of the vehicle causes the intensity of the vibration of the optical fiber to exceed the threshold, and the optical fiber sensing device 1200 detects the passage of a vehicle.

[0016] However, as shown in Figure 15, the signal strength varies depending on the vehicle type, and the optical fiber sensing device 1200 has the problem that the signal strength of a large vehicle exceeds the threshold and can detect the vehicle, but the signal strength of a small vehicle does not exceed the threshold and cannot detect the vehicle.

[0017] As shown in Figure 16, the optical fiber sensing device 1200 has the problem that when the same vehicle passes through a lane close to the optical fiber, the signal strength exceeds the threshold and the vehicle can be detected, but when the vehicle passes through a lane far from the optical fiber, the signal strength does not exceed the threshold and the vehicle cannot be detected.

[0018] 17, the frequency characteristics when passing through the measurement differ depending on the vehicle type. Therefore, the optical fiber sensing device 1200 could not identify the vehicle type by measuring whether the intensity exceeds the threshold value.

[0019] (Explanation of Optical Fiber Sensing Device According to an Embodiment) Fig. 1 is a schematic diagram of a running vehicle detection device according to an embodiment. Fig. 2 is a block diagram of an optical fiber sensor according to an embodiment. Fig. 3 is a diagram showing measurement results at one measurement point according to an embodiment. Fig. 4 is a diagram showing measurement points at multiple locations according to an embodiment. A running vehicle detection device according to an embodiment will be described with reference to Figs. 1 to 4.

[0020] As shown in FIG. 1, a running vehicle detection device 100 includes an optical fiber cable 101 laid along a road and a sensing device 102 connected to the end of the optical fiber cable.

[0021] As shown in FIG. 2, the sensing device 102 includes a light source 202 , an optical modulator 203 , a circulator 204 , a photodetector 205 , a data processing unit 206 , a control unit 212 , and an output unit 213 .

[0022] The optical fiber cable 101 is a linear cable that detects backscattered light (Rayleigh scattered light) of emitted laser light. The optical fiber cable 101 transmits light. The optical fiber cable is made of a material capable of transmitting light, such as fibrous quartz glass or plastic, and has a two-layer structure consisting of a central core and a cladding that surrounds the core. One end of the optical fiber cable 101 is connected to the circulator 204. The optical fiber cable 101 is installed where vibration detection is desired, in this case, near a road.

[0023] The light source 202 is a laser light source with high coherence and narrow linewidth. The laser light source may be a solid-state laser such as a ruby ​​laser or a YAG laser, a liquid laser such as a dye laser, an excimer laser, or a CO ２ A gas laser, a semiconductor laser, etc. can be used. The light source 202 oscillates pulsed light at a constant cycle toward the optical modulator 203 under the control of the control unit 212.

[0024] The optical modulator 203 is a device that modulates the pulsed light output from the light source 202. The optical modulator 203 can change the wavelength, frequency, intensity, phase, etc. of the light under the control of the control unit 212. The pulsed light modulated by the optical modulator 203 is output to the circulator 204.

[0025] The circulator 204 is a device that emits the pulsed light output from the optical modulator 203 toward the optical fiber cable 101. The circulator 204 also outputs the backscattered light that has returned from the optical fiber cable 101 toward the photodetector 205.

[0026] The photodetector 205 is a device that converts the received backscattered light into analog data. For example, a photodiode can be used as the photodetector. The analog data output from the photodetector 205 is output to the data processing unit 206.

[0027] The data processing unit 206 includes an analog-to-digital conversion unit 207, a detection unit 209, a determination unit 210, and a storage unit 211. The data processing unit 206 is a device configured with a semiconductor integrated circuit or the like. The analog-to-digital conversion unit 207 is a device that converts analog data output from the photodetector 205 into digital data. The analog-to-digital conversion unit 207, together with the control unit 212, processes analog data related to the backscattered light output from the photodetector 205. The photodetector 205 and the analog-to-digital conversion unit 207 correspond to the measurement unit 208.

[0028] The detection unit 209 is a device that detects feature amounts from the data measured by the measurement unit 208. The storage unit 211 is a device that stores feature amount data or thresholds obtained by machine learning, which are to be compared with the feature amount data measured by the measurement unit 208. The determination unit 210 is a device that compares the feature amount data measured by the measurement unit 208 with the feature amount data or thresholds stored in the storage unit 211, and determines the presence or absence of a traveling vehicle.

[0029] The control unit 212 is a device configured with a semiconductor integrated circuit, just like the data processing unit 206. The control unit 212 is a central processing unit capable of executing programs. Here, the control unit 212 controls and processes the digital data of the data processing unit 206.

[0030] The output unit 213 is a device such as a display device and a speaker, and outputs data as a display, a sound, or the like.

[0031] The method for detecting vibrations of a moving vehicle using optical fiber sensing is the same as the related technology described above, and therefore will not be described here.

[0032] The running vehicle detection device 100 according to the embodiment measures changes over time in a physical quantity at a measurement point of interest. Measurement points 1, 2, 3, and 4 are arranged at multiple locations along a road, and the running vehicle detection device 100 measures changes in each physical quantity. Here, the physical quantity is preferably the intensity of vibration.

[0033] Different measurement points can be created by increasing the number of optical fibers (channels) and synchronizing the sensing devices. However, different measurement points can also be created by measuring the time it takes for backscattered light from emitted laser light to return, and simultaneously measure the different points.

[0034] Next, the running vehicle detection device 100 detects patterns of change in physical quantities at different times or different locations. One of the patterns of change in physical quantities is, as shown in FIG. 3, a change in the vibration intensity (change in the absolute value of the intensity, change in the difference in intensity) before and after the passage of a vehicle, and a change in the vibration intensity ratio. These are used as feature quantities. Another pattern of change in physical quantities is, as shown in FIG. 4, a change in the vibration intensity and change in the vibration intensity ratio between a measurement point of interest and measurement points before and after it. These are used as feature quantities. These changes in vibration intensity and intensity ratio are detected. Furthermore, a change in the shape of a graph of vibration intensity may be detected as a feature quantity.

[0035] These feature amounts may be detected for each vibration frequency, such as 0 to 10 Hz, 10 to 30 Hz, 30 to 50 Hz, 50 to 70 Hz, 70 to 100 Hz, 0 to 50 Hz, 50 to 100 Hz, and 0 to 100 Hz.

[0036] As described above, these feature amounts may be detected for the current time and the past. Alternatively, these feature amounts may be detected for the current time, the past, and the future. When future time is used, a delay occurs for the time that references the future. Using feature amounts at multiple times makes it possible to express the characteristics of a traveling vehicle.

[0037] As described above, these feature quantities may be detected for the measurement point of interest and the measurement points before and after it. Using feature quantities from multiple measurement points makes it possible to express the characteristics of a moving vehicle. Furthermore, these feature quantities may use changes in vibration, which is the vehicle speed.

[0038] The change pattern of this physical quantity is compared with a spatiotemporal signal feature model that reflects the characteristics of moving vehicles measured in advance, and the presence or absence of a moving vehicle is determined. The feature model is preferably created using supervised machine learning such as gradient boosting or random forest. However, unsupervised machine learning such as deep learning may also be used. Although comparison with a feature model created by machine learning enables highly accurate detection, a determination may also be made simply based on whether a threshold is exceeded.

[0039] Such a running vehicle detection device can detect running vehicles with high accuracy.

[0040] (Method for determining whether a vehicle is a traveling vehicle by a traveling vehicle detection device according to the first embodiment) Fig. 5 is a block diagram of a form for determining a feature amount based on a threshold value according to the first embodiment. The method for determining whether a vehicle is a traveling vehicle by a traveling vehicle detection device according to the first embodiment will be described with reference to Fig. 5 .

[0041] As shown in Fig. 5, first, the sensing device 102 according to the embodiment acquires time waveform data of vibration intensity for each measurement point (S501). Next, the acquired time waveform data is subjected to a Fast Fourier Transform (FFT) to acquire spectrum data for each measurement point (S502). Next, the data is divided into data for each evaluation band of vibration frequency (S503). Next, each feature amount is calculated (S504). Next, threshold processing is performed for each feature amount (S505). Finally, the presence or absence of a traveling vehicle is determined based on whether the feature amount exceeds a threshold (S506).

[0042] In this way, the running vehicle detection device can detect running vehicles with high accuracy.

[0043] (Method for determining whether a vehicle is a traveling vehicle by a traveling vehicle detection device according to a second embodiment) Fig. 6 is a block diagram of a form for determining a feature amount based on a threshold value according to a second embodiment. A method for determining whether a vehicle is a traveling vehicle by a traveling vehicle detection device according to a second embodiment will be described with reference to Fig. 6 .

[0044] As shown in Fig. 6, first, the sensing device 102 according to the embodiment acquires time waveform data of vibration intensity for each measurement point (S601). Next, the acquired time waveform data is divided into processing time intervals (S602). Next, each feature amount is calculated (S603). Next, threshold processing is performed for each feature amount (S604). Finally, the presence or absence of a traveling vehicle is determined based on whether the threshold is exceeded (S605).

[0045] In this way, the time waveform data of the vibration intensity can be easily processed without performing a fast Fourier transform.

[0046] (Method for determining a running vehicle by a running vehicle detection device according to the third embodiment) Fig. 7 is a block diagram of a form in which a feature is modeled and used by a learning device according to the third embodiment. A method for determining a running vehicle by a running vehicle detection device according to the third embodiment will be described with reference to Fig. 7 .

[0047] 7, first, time waveform data of vibration intensity for each measurement point is acquired as past data by the sensing device 102 according to the embodiment (S701). Next, the acquired time waveform data is subjected to a fast Fourier transform (FFT) to acquire spectrum data for each measurement point (S702). Next, the data is divided into data for each evaluation band of the vibration frequency (S703). Next, each feature amount is calculated (S704).

[0048] The camera data and the reference data obtained by the loop coil, which indicates the presence or absence of a running vehicle, are used as label data, and the obtained features are input into a learning device for supervised learning to create a feature model (S705). These steps are executed in advance, and the feature model is created before actual measurement.

[0049] Similarly, time waveform data of vibration intensity for each measurement point is acquired by the sensing device 102 according to the embodiment as data to be inferred (S706). Next, the acquired time waveform data is subjected to a fast Fourier transform (FFT) to acquire spectrum data for each measurement point (S707). Next, the data is divided into data for each vibration frequency band (S708). Next, each feature amount is calculated (S709).

[0050] Next, the feature quantities of the data to be inferred are compared with the feature quantity model to perform inference (S710). Finally, the presence or absence of a running vehicle is determined and output (S711). In this way, the running vehicle detection device can detect running vehicles more accurately than by threshold determination.

[0051] (Method for determining a running vehicle by a running vehicle detection device according to a fourth embodiment) FIG. 8 is a block diagram of a form in which feature quantities are modeled and used by a learning device according to a fourth embodiment. A method for determining a running vehicle by a running vehicle detection device according to the fourth embodiment will be described with reference to FIG. 8. The fourth embodiment differs from the third embodiment in that it determines the presence or absence of a running vehicle for each vehicle type and each lane. Therefore, only the differences between the fourth embodiment and the third embodiment will be described.

[0052] In the fourth embodiment, lane and vehicle type information is added to the label data, and supervised learning is performed in which the camera data, label data obtained by the loop coil, and the obtained features are input to a learning device to create a feature model (S805).

[0053] For this purpose, the feature quantities of the data to be inferred are compared with the feature quantity model to perform inference (S810), and the presence or absence of a traveling vehicle is determined for each traveling lane and each vehicle type (S811).

[0054] In this way, a highly accurate running vehicle detection device and a running vehicle detection method are provided by using threshold determination or a learning device.

[0055] Furthermore, some or all of the processing in the running vehicle detection device 100 described above can be implemented as a computer program. Such a program can be stored on various types of non-transitory computer-readable media and supplied to a computer. Non-transitory computer-readable media include various types of tangible recording media. Examples of non-transitory computer-readable media include magnetic recording media (e.g., flexible disks, magnetic tapes, hard disk drives), magneto-optical recording media (e.g., magneto-optical disks), CD-ROMs (Read Only Memory), CD-Rs, CD-R / Ws, and semiconductor memories (e.g., mask ROMs, PROMs (Programmable ROMs), EPROMs (Erasable PROMs), flash ROMs, and RAMs (Random Access Memory)). The program may also be supplied to a computer by various types of temporary computer-readable media. Examples of temporary computer-readable media include electrical signals, optical signals, and electromagnetic waves. The temporary computer-readable medium can supply the program to the computer via a wired communication path such as an electric wire or an optical fiber, or via a wireless communication path.

[0056] Such a non-transitory computer-readable medium can enable a vehicle detection device to perform a vehicle detection method using thresholding or a learner.

[0057] (Example) In this example, measurements were taken at three measurement points: 668 (-1), 669, and 670 (+1). In this example, the vibration intensity, intensity difference, and intensity ratio for each frequency were used as feature quantities. In addition, in this example, machine learning using gradient boosting was performed. In addition, in this example, measurements were taken with a time resolution of 100 msec (10 data points every 10 msec were used as one), a distance resolution of 3.2 m, and three channels. The results are shown in Table 1 of FIG. 9.

[0058] This was used to evaluate the degree to which passing vehicles on two lanes could be detected. The location was a highway, the time was 15 minutes, and the number of passing vehicles was 189. The results are shown in Table 2 in Figure 10.

[0059] It can be seen that the method of the present disclosure has a higher detection rate and a lower false positive rate than the comparative method. In particular, the method of the present disclosure was able to improve the detection rate in distant lanes with weak traffic lights, such as oncoming lanes.

[0060] Fig. 9 is a block diagram of a running vehicle detection device according to the present disclosure. The running vehicle detection device according to the present disclosure will be described with reference to Fig. 11. The running vehicle detection device 100 according to the present disclosure includes a measurement unit 208 that measures continuous physical quantities at a predetermined location using optical fiber sensors laid along a road, a detection unit 209 that detects a change pattern in the physical quantities over time from the measurement results of the measurement unit, and a determination unit 210 that determines the presence or absence of a running vehicle based on the change pattern in the physical quantities.

[0061] Although the embodiments of the present invention have been described above, the present invention includes appropriate modifications that do not impair the objects and advantages thereof, and is not limited to the above-described embodiments. For example, although the second embodiment uses a threshold value for judgment, a learning device may be used for judgment as in the third and fourth embodiments.

[0062] Some or all of the above embodiments can be described as, but are not limited to, the following supplementary notes. (Supplementary Note 1) A running vehicle detection device comprising: a measurement unit that measures continuous physical quantities at predetermined locations using optical fiber sensors laid along a road; a detection unit that detects a pattern of change in the physical quantities over time from the measurement results of the measurement unit; and a determination unit that determines the presence or absence of a running vehicle based on the pattern of change in the physical quantities. (Supplementary Note 2) The running vehicle detection device according to Supplementary Note 1, wherein there are a plurality of the predetermined locations that are arranged along the road. (Supplementary Note 3) The running vehicle detection device according to Supplementary Note 1 or 2, wherein the physical quantity is vibration intensity, and the change pattern is a change in absolute value, a change in difference, a change in ratio, or a change in graph shape. (Supplementary Note 4) The running vehicle detection device according to Supplementary Note 3, wherein the detection unit detects the vibration intensity for each predetermined frequency band using a fast Fourier transform. (Supplementary Note 5) The running vehicle detection device according to any one of Supplements 1 to 4, wherein the determination unit compares the pattern of change in the physical quantities with a feature quantity model of the physical quantities created using machine learning. (Supplementary Note 6) The running vehicle detection device according to Supplementary Note 5, wherein the machine learning is supervised machine learning. (Supplementary Note 7) The running vehicle detection device according to any one of Supplements 1 to 6, wherein the determination unit determines the vehicle type and driving lane of the running vehicle. (Supplementary Note 8) A running vehicle detection method comprising the steps of measuring continuous physical quantities at predetermined locations using optical fiber sensors laid along a road, detecting a pattern of change in the physical quantities over time from measurement results of the measuring step, and determining the presence or absence of a running vehicle based on the pattern of change in the physical quantities. (Supplementary Note 9) The running vehicle detection method according to Supplementary Note 8, wherein there are multiple predetermined locations and the predetermined locations are arranged along the road. (Supplementary Note 10) A non-transitory computer-readable medium having recorded thereon a program causing a running vehicle detection device to execute the steps of measuring continuous physical quantities at predetermined locations using optical fiber sensors laid along the road, detecting a pattern of change in the physical quantities over time from measurement results of the measuring step, and determining the presence or absence of a running vehicle based on the pattern of change in the physical quantities.(Supplementary Note 11) The running vehicle detection method according to Supplementary Note 8 or 9, wherein the physical quantity is vibration intensity, and the change pattern is a change in absolute value, a change in difference, a change in ratio, or a change in graph shape. (Supplementary Note 12) The running vehicle detection method according to Supplementary Note 11, wherein the step of detecting a change pattern in the physical quantity comprises a step of detecting the vibration intensity for each predetermined frequency band using a fast Fourier transform. (Supplementary Note 13) The running vehicle detection method according to Supplementary Note 8, 9, 11, or 12, wherein the step of determining the presence or absence of a running vehicle comprises a step of comparing the change pattern in the physical quantity with a feature quantity model of the physical quantity created using machine learning. (Supplementary Note 14) The running vehicle detection method according to Supplementary Note 13, wherein the machine learning is supervised machine learning. (Supplementary Note 15) The running vehicle detection method according to any one of Supplements 8, 9, 11 to 14, wherein the step of determining the presence or absence of a running vehicle determines the vehicle type and driving lane of the running vehicle. (Supplementary Note 16) The non-transitory computer-readable medium of Supplementary Note 10, wherein there are a plurality of the predetermined locations and the predetermined locations are arranged along the road. (Supplementary Note 17) The non-transitory computer-readable medium of Supplementary Note 10 or 16, wherein the physical quantity is vibration intensity, and the change pattern is a change in absolute value, a change in difference, a change in ratio, or a change in graph shape. (Supplementary Note 18) The non-transitory computer-readable medium of Supplementary Note 17, wherein the step of detecting the change pattern of the physical quantity comprises a step of detecting the vibration intensity for each predetermined frequency band using a fast Fourier transform. (Supplementary Note 19) The non-transitory computer-readable medium of any one of Supplements 10, 16 to 18, wherein the step of determining the presence or absence of a traveling vehicle comprises a step of comparing the change pattern of the physical quantity with a feature quantity model of the physical quantity created using machine learning. (Supplementary Note 20) The non-transitory computer-readable medium of claim 19, wherein the machine learning is supervised machine learning. (Supplementary Note 21) The non-transitory computer-readable medium according to any one of Supplementary Notes 10, 16 to 20, wherein the step of determining the presence or absence of a traveling vehicle determines a type of vehicle and a traveling lane of the traveling vehicle.

[0063] REFERENCE SIGNS LIST 100 Running vehicle detection device 101 Optical fiber cable 102 Sensing device 202 Light source 203 Optical modulator 204 Circulator 205 Photodetector 206 Data processing unit 207 Analog-to-digital conversion unit 208 Measuring unit 209 Detecting unit 210 Determining unit 211 Storage unit 212 Control unit 213 Output unit 1000 Optical fiber sensing device 1001 Optical fiber cable 1002 Sensing device

Claims

1. A measurement unit that measures continuous physical quantities at a predetermined location using optical fiber sensors installed along a road; a detection unit that detects a change pattern of the physical quantity over time from a measurement result of the measurement unit; a determination unit that determines the presence or absence of a traveling vehicle based on a change pattern of the physical quantity.

2. The moving vehicle detection device according to claim 1 , wherein the predetermined locations are a plurality of locations arranged along the road.

3. 2. The running vehicle detection device according to claim 1, wherein the physical quantity is a vibration intensity, and the change pattern is a change in absolute value, a change in difference, a change in ratio, or a change in graph shape.

4. The running vehicle detection device according to claim 3 , wherein the detection unit detects the intensity of the vibration for each predetermined frequency band by using a fast Fourier transform.

5. The running vehicle detection device according to claim 1 , wherein the determination unit compares the change pattern of the physical quantity with a feature quantity model of the physical quantity that is created using machine learning.

6. The running vehicle detection device according to claim 5 , wherein the machine learning is supervised machine learning.

7. The traveling vehicle detection device according to claim 1 , wherein the determination unit determines a vehicle type and a traveling lane of the traveling vehicle.

8. Measuring continuous physical quantities at a predetermined location with optical fiber sensors installed along a road; detecting a pattern of change in the physical quantity over time from a measurement result of the measuring step; and determining whether or not a traveling vehicle is present based on a change pattern of the physical quantity.

9. The method of claim 8 , wherein the predetermined locations are a plurality of locations arranged along the road.

10. Measuring continuous physical quantities at a predetermined location with optical fiber sensors installed along a road; detecting a pattern of change in the physical quantity over time from a measurement result of the measuring step; and determining whether or not a traveling vehicle is present based on a change pattern of the physical quantity.