Mobile object detection system and mobile object detection method

Abstract

To provide means capable of highly accurately detecting a movement of a mobile body.SOLUTION: A mobile body detection system includes a vibration sensor and a control section. The control section detects a peak in data from the sensor. If other peaks exist within a predetermined time from an object peak for determining whether the detected peak corresponds to a movement of a mobile body, the control section detects one of the peaks as the movement of the mobile body. If the other peaks exist within the predetermined time from the object peak, the control section does not detect the other peaks as the movement of the mobile body but detects the object peak as the movement of the mobile body.SELECTED DRAWING: Figure 1

Description

Technical Field

[0004] , , , , , , ، ,

[0006] , , , ,

[0005] , , , , , , ,

[0003]

[0001] The present invention relates to a moving object detection system for detecting the movement of a moving object and a moving object detection method.

Background Art

[0002] <0OOOO09>The applicant has filed an invention for detecting a moving object such as a vehicle using a vibration sensor (Japanese Patent Application No. 2020-117429). In the invention according to Japanese Patent Application No. 2020-117429, in the data based on vibration data, a mark is placed on a peak value exceeding a predetermined threshold value, and when there are two or more marks within a predetermined time, vibration data for a predetermined time centered on the high peak value is cut out, and based on the cut-out vibration data, the movement of the moving object is detected. According to the invention according to Japanese Patent Application No. 2020-117429, when a vibration peak of a passing vehicle is detected, particularly in the case of a vehicle with a long vehicle length, as shown in FIG. 3, for one passing vehicle, two vibration peaks (the portions indicated by arrows in FIG. 3) may be detected. In the above invention, since the determination as to whether it is a vehicle or noise is made for both of the two detected peaks, the counted number of vehicles may be larger than the actual number of passing vehicles, which is a problem.

[0003] Note that Patent Document 1 discloses a vehicle classification device that classifies the type of vehicle (passenger car, truck, tractor, etc.) using a vibration sensor.

Prior Art Documents

Patent Documents

[0004]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0005] As described above, the above conventional invention has a problem that a moving object may be erroneously detected. <00000зо> The object of the present invention is to provide a means capable of detecting the movement of a moving object with high accuracy. [Means for solving the problem]

[0007] The first invention's moving object detection system comprises a sensor and a control unit, wherein the control unit detects a peak in the data from the sensor and determines whether the detected peak is the movement of a moving object. If other peaks exist within a predetermined time from the target peak, the control unit detects one of the peaks as the movement of a moving object.

[0008] In this invention, if the control unit detects a peak as the movement of a moving object if other peaks exist within a predetermined time from the target peak used to determine whether or not it is the movement of a moving object, it detects one of the peaks as the movement of the moving object. Therefore, for example, even if the moving object is a long vehicle and multiple peaks indicating the movement of a moving object exist within a predetermined time, it will be detected as the movement of a single moving object. As a result, false detection of moving objects is prevented, and the movement of moving objects can be detected with high accuracy.

[0009] The second invention's moving object detection system is characterized in that, in the moving object detection system of the first invention, if another peak exists within a predetermined time from the target peak, the control unit does not detect the other peak as the movement of a moving object, but detects the target peak as the movement of a moving object.

[0010] The third invention's moving object detection system is characterized in that, in the moving object detection system of the first invention, if one other peak exists within a predetermined time period from the target peak, the control unit does not detect the other peak as the movement of a moving object, but detects the target peak as the movement of a moving object.

[0011] The fourth invention's moving object detection system is characterized in that, in the moving object detection system of the second or third invention, the control unit disables other peaks so as not to detect other peaks as the movement of a moving object.

[0012] The fifth invention's moving object detection system comprises a sensor and a control unit, wherein the control unit detects peaks in the data from the sensor, counts the number of peaks as the number of moving objects, and when counting the number of peaks as the number of moving objects, if other peaks exist within a predetermined time from the target peak for determining whether or not to count the detected peak as a moving object, it counts one of the peaks as a moving object.

[0013] In this invention, when the control unit counts the number of peaks as the number of moving objects, if other peaks exist within a predetermined time from the target peak for determining whether or not to count a detected peak as a moving object, it counts one of the peaks as the moving object. Therefore, for example, even if the moving object is a long vehicle and there are multiple peaks indicating the movement of a moving object within a predetermined time, it will be counted as one moving object. As a result, false detection (counting) of moving objects is prevented, and the movement (number) of moving objects can be detected (counted) with high accuracy.

[0014] The moving object detection system of the sixth invention is characterized in that, in the moving object detection system of the fifth invention, if other peaks exist within a predetermined time from the target peak, the control unit does not count the other peaks as moving objects, but counts the target peak as a moving object.

[0015] The seventh invention's moving object detection system is the fifth invention's moving object detection system, characterized in that, if another peak exists within a predetermined time period from the target peak, the control unit does not count the other peak as a moving object, but counts the target peak as a moving object.

[0016] The eighth invention's moving object detection system is characterized in that, in the sixth or seventh invention's moving object detection system, the control unit disables other peaks and does not count other peaks as moving objects.

[0017] The ninth invention's moving object detection system comprises a sensor and a control unit, wherein the control unit detects peaks in the data from the sensor, classifies the detected peaks into types of moving objects, and, if other peaks exist within a predetermined time from the target peak for determining whether or not to detect a peak classified as a type of moving object, it detects any one of the peaks of the same type as a moving object.

[0018] In this invention, the control unit, when a peak classified as a type of moving object exists within a predetermined time from the target peak used to determine whether or not to detect it as a moving object, detects one of the peaks of the same type as the moving object. Therefore, for example, if the moving object is a vehicle, even if moving objects of different types, such as a large vehicle and a small vehicle, are detected consecutively, neither will be invalidated, and each will be detected as a moving object. As a result, false detection of moving objects is prevented, and the movement of moving objects can be detected with high accuracy.

[0019] The tenth invention's moving object detection system is characterized in that, in the moving object detection system of the ninth invention, if another peak exists within a predetermined time from the target peak and the other peak is of the same type as the target peak, the control unit does not detect the other peak as a moving object, but detects the target peak as a moving object.

[0020] The 11th invention is a moving object detection system of the 9th invention, characterized in that, if one other peak exists within a predetermined time period from the target peak, and the other peak is of the same type as the target peak, the control unit does not detect the other peak as a moving object, but detects the target peak as a moving object.

[0021] The moving object detection system of the 12th invention is characterized in that, in the moving object detection system of the 10th or 11th invention, the control unit disables other peaks so as not to detect other peaks as moving objects.

[0022] The mobile object detection system of the 13th invention includes a sensor and a control unit. The control unit detects peaks in the data from the sensor, classifies the detected peaks into types of mobile objects, counts the peaks classified into types of mobile objects for each type of mobile object, and when counting the peaks classified into types of mobile objects for each type of mobile object, if there are other peaks within a predetermined time from the target peak to be determined whether to count the peak classified into the type of mobile object as a mobile object, one of the peaks with the same type is counted as a mobile object.

[0023] In the present invention, when the control unit counts the peaks classified into types of mobile objects for each type of mobile object, if there are other peaks within a predetermined time from the target peak to be determined whether to count the peak classified into the type of mobile object as a mobile object, one of the peaks with the same type is counted as a mobile object. Therefore, for example, when different types of mobile objects such as large vehicles and small vehicles are continuously detected when the mobile object is a vehicle, none of them are invalidated, and each is detected (counted) as a mobile object. For this reason, false detection (counting) of mobile objects is prevented, and the movement (number) of mobile objects can be detected (counted) with high accuracy.

[0024] The mobile object detection system of the 14th invention is the mobile object detection system of the 13th invention, wherein when there are other peaks within a predetermined time from the target peak and the other peaks are of the same type as the target peak, the control unit does not count the other peaks as mobile objects and counts the target peak as a mobile object.

[0025] The mobile object detection system of the 15th invention is the mobile object detection system of the 13th invention, wherein when there is one other peak within a predetermined time after the target peak and the other peak is of the same type as the target peak, the control unit does not count the other peak as a mobile object and counts the target peak as a mobile object.

[0026] The moving object detection system of the 16th invention is the moving object detection system of the 14th or 15th invention, wherein the control unit invalidates other peaks, and does not count other peaks as moving objects.

[0027] The moving object detection system of the 17th invention is the moving object detection system of any one of the 9th to 16th inventions, wherein the moving object is a vehicle, and the control unit classifies the detected peak into a large vehicle or a small vehicle based on the detected peak.

[0028] The moving object detection system of the 18th invention is the moving object detection system of any one of the 1st to 17th inventions, wherein the sensor is a vibration sensor.

[0029] The moving object detection system of the 19th invention is the moving object detection system of any one of the 1st to 18th inventions, wherein the control unit classifies the data from the vibration sensor into vibrations that are peaks and noise.

[0030] The moving object detection method of the 20th invention is to detect a peak in the data from the sensor, and when there is another peak within a predetermined time from the target peak for determining whether the detected peak is the movement of a moving object, detect any one of the peaks as the movement of the moving object.

[0031] The moving object detection method of the 21st invention is to detect a peak in the data from the sensor, count the number of detected peaks as the number of moving objects, and when counting the number of detected peaks as the number of moving objects, if there is another peak within a predetermined time from the target peak for determining whether to count the detected peak as a moving object, count any one of the peaks as a moving object.

[0032] The 22nd invention's method for detecting moving objects is characterized by detecting a peak in the data from a sensor, classifying the detected peak into the type of moving object, and if other peaks exist within a predetermined time from the target peak for determining whether or not to detect a peak classified as a moving object, detecting one of the peaks of the same type as the moving object.

[0033] The 23rd invention provides a method for detecting moving objects, which involves detecting a peak in the data from a sensor, classifying the detected peak into the type of moving object, counting the peaks classified into the type of moving object for each type of moving object, and, when counting the peaks classified into the type of moving object for each type of moving object, if other peaks exist within a predetermined time from the target peak for determining whether or not to count the peak classified into the type of moving object as a moving object, then one of the peaks of the same type is counted as a moving object. [Effects of the Invention]

[0034] According to the present invention, the movement of a moving object can be detected with high accuracy. [Brief explanation of the drawing]

[0035] [Figure 1] A flowchart illustrating the processing operation of the moving object detection system (first embodiment). [Figure 2] A flowchart illustrating the processing operation of the moving object detection system (second embodiment). [Figure 3] This is a diagram showing vibration data, etc. [Modes for carrying out the invention]

[0036] Embodiments of the present invention will be described below. The moving object detection system according to an embodiment of the present invention detects the movement of moving objects such as vehicles and counts the number of detected moving objects. The moving object detection system includes a vibration sensor, a control unit such as a CPU (Central Processing Unit), etc. The moving object detection system may consist of one device or two or more devices. Furthermore, although the details will be described later, the detection of moving objects based on vibration data detected by the vibration sensor may be performed by, for example, a (cloud) server. Note that the processes described below are basically executed by the control unit, but in processes that say "the control unit does...", the phrase "the control unit does..." may be omitted.

[0037] Vibration sensors are installed, for example, on roads. The vibration sensors detect vibrations transmitted to the road from vehicles passing in front of them and output vibration data. The control unit extracts feature quantities based on the vibration data output from the vibration sensors and classifies the extracted feature quantities as either noise or vibrations caused by passing vehicles. The control unit then counts the number of vibrations classified as caused by passing vehicles as the number of passing vehicles. For details on feature extraction, please refer to the applicant's patent application, Japanese Patent Application No. 2020-117429.

[0038] (First Embodiment) If the control unit detects two vibration peaks that are determined to be caused by a passing vehicle and they are adjacent to each other within one second, it invalidates the later vibration peak. The control unit does not count the invalidated vibration peak as a passing vehicle. In other words, the counting result for the invalidated vibration peak is invalid.

[0039] Specifically, if two or more vibration peaks identified as vehicle vibrations are adjacent, the control unit compares the occurrence times of the first vibration peak with that of the next vibration peak. If the difference in occurrence times between the two peaks is within 1 second, the control unit invalidates the later vibration peak. Therefore, the counting results from the invalidated vibration peak are invalid. Furthermore, the control unit does not compare the occurrence times of the invalidated vibration peak with any adjacent peaks.

[0040] In other words, the control unit detects vibration peaks in the vibration data from the vibration sensor. The control unit then determines whether the detected vibration peak is due to a vehicle passing by. If other vibration peaks exist within a predetermined time from the target vibration peak, the control unit detects one of the vibration peaks as the vehicle passing (movement of a moving object). For example, if there is one vibration peak before or after a predetermined time from the target vibration peak, and one vibration peak before or after the predetermined time, that is, if there are three vibration peaks from before to after a predetermined time from the time of the target vibration peak, the control unit detects one of the vibration peaks as the vehicle passing by.

[0041] For example, if other vibration peaks exist within a predetermined time interval from the target vibration peak, the control unit will not detect the other vibration peaks as the passing of a vehicle, but will instead detect the target vibration peak as the passing of a vehicle. In other words, as described above, if there is one vibration peak before a predetermined time and one vibration peak before a predetermined time interval centered around the target vibration peak, that is, if there are three vibration peaks between a predetermined time interval and a predetermined time interval centered around the time of the target vibration peak, the control unit will detect the target vibration peak (the vibration peak in the middle of the time) as the passing of a vehicle.

[0042] As described above, if there is another vibration peak within a predetermined time (1 second) after the target vibration peak (i.e., there are two adjacent vibration peaks), the control unit will not detect the other vibration peak (the vibration peak that is later in time) as the passing of a vehicle, but will instead detect the target vibration peak (the vibration peak that is earlier in time) as the passing of a vehicle.

[0043] The control unit disables other vibration peaks, thereby preventing them from being detected as vehicle passing. The control unit then detects the target vibration peak as vehicle passing (passing vehicle).

[0044] The following describes the processing operation of the moving object detection system based on the flowchart shown in Figure 1. The control unit of the moving object detection system sets N, which indicates which peak it is, to 1 (N←1) (S1). Next, the control unit determines whether N is less than the total number of peaks (N < total number of peaks) (S2). If the control unit determines that N is less than the total number of peaks (S2:Yes), it checks the inference result for the Nth peak (S3). Next, the control unit determines whether the Nth peak is a peak that has been determined to be a valid vehicle passage (S4).

[0045] If the control unit determines that the Nth peak is a peak that has been determined to be a valid vehicle passage (S4:Yes), it checks the N+1th peak (S5). Next, the control unit determines whether the N+1th peak exists within 1 second of the Nth peak and is a peak that has been determined to be a vehicle passage (S6). If the control unit determines that the N+1th peak exists within 1 second of the Nth peak and is a peak that has been determined to be a vehicle passage (S6:Yes), it invalidates the N+1th peak (S7).

[0046] After processing S7, the control unit determines that the Nth peak is not a peak determined to be a valid vehicle passage (S4: No), or that the N+1th peak does not exist within 1 second of the Nth peak, or is not a peak determined to be a vehicle passage (S6: No), then it sets N to N+1 (N←N+1) (S8) in order to check the N+1th peak. After processing S8, the control unit executes processing S2.

[0047] If the control unit determines that N is not less than the total number of peaks, i.e., N is the total number of peaks (N = total number of peaks) (S2: No), it counts the peaks that were determined to be valid vehicle passages (S9).

[0048] The mobile object detection system according to this embodiment enables more accurate counting of passing vehicles. Therefore, it enables more accurate traffic volume surveys. Field tests using the mobile object detection system according to this embodiment showed that the counting accuracy (the ratio of the number of vehicles measured by machine learning divided by the true value) improved by approximately 4%, enabling more accurate counting of passing vehicles. When the vehicle counting invalidation described in this embodiment was not applied, the true value was 3247 vehicles, the count was 3389 vehicles, and the counting accuracy was 104%. However, when the vehicle counting invalidation was applied, the true value was 3247 vehicles, the count was 3260 vehicles, and the counting accuracy was 100%.

[0049] As explained above, in this embodiment, if the control unit of the moving object detection system detects one of the peaks as the movement of a moving object if other peaks exist within a predetermined time (e.g., 1 second) from the target peak used to determine whether or not it is the movement of a moving object. Therefore, for example, even if the moving object is a long vehicle and multiple peaks indicating the movement of a moving object exist within a predetermined time, it will be detected as the movement of a single moving object. As a result, false detection of moving objects (vehicles) is prevented, and the movement of moving objects can be detected with high accuracy.

[0050] In this embodiment, after confirming whether each vibration peak is invalid or not, vibration peaks that are determined to be valid are counted as vehicles (moving objects). Alternatively, as described below, the determination of whether a vibration peak is invalid or not may be made when counting vehicles.

[0051] The control unit detects vibration peaks in the vibration data from the vibration sensor. When the control unit counts the detected vibration peaks, it determines whether or not to count the detected vibration peak as a vehicle (moving object). If other vibration peaks exist within a predetermined time from the target vibration peak, the control unit counts one of the vibration peaks as a vehicle (moving object). For example, if there is one vibration peak before a predetermined time and one vibration peak before a predetermined time from the target vibration peak, that is, if there are three vibration peaks between a predetermined time and a predetermined time from the target vibration peak, the control unit counts one of the vibration peaks as a vehicle.

[0052] For example, if the control unit has other vibration peaks within a predetermined time interval from the target vibration peak, it will not count the other vibration peaks as vehicles, but will count the target vibration peak as a vehicle. In other words, as described above, if there is one vibration peak before or after a predetermined time and one vibration peak before or after a predetermined time interval centered on the target vibration peak, that is, if there are three vibration peaks between a predetermined time interval and a predetermined time interval centered on the time of the target vibration peak, the control unit will count the target vibration peak (the vibration peak in the middle in terms of time) as a vehicle.

[0053] As described above, if there is another vibration peak within a predetermined time (1 second) after the target vibration peak (i.e., there are two adjacent vibration peaks), the control unit will not count the other vibration peak (the vibration peak that is later in time) as a vehicle, but will count the target vibration peak (the vibration peak that is earlier in time) as a vehicle.

[0054] The control unit disables other vibration peaks, thereby preventing them from being counted as part of the vehicle.

[0055] As explained above, when the control unit counts the number of detected peaks as the number of moving objects, if other peaks exist within a predetermined time (e.g., 1 second) from the target peak used to determine whether or not to count a detected peak as a moving object, it counts one of the peaks as the moving object. Therefore, for example, even if the moving object is a long vehicle and multiple peaks indicating the movement of a moving object exist within a predetermined time, it will be counted as one moving object. As a result, false detection (counting) of moving objects (vehicles) is prevented, and the movement (number) of moving objects can be detected (counted) with high accuracy.

[0056] (Second Embodiment) In the second embodiment, similar to the first embodiment, data from vibration data is classified into two categories: vehicles (vibration peaks) and noise. Subsequently, vibration peaks identified as vehicles are classified into two categories: large vehicles and small vehicles. In this case, if the invalidation of vehicle counting shown in the first embodiment is applied, peaks from different vehicle types may be invalidated. For example, a peak identified as a small vehicle may invalidate a peak identified as an adjacent large vehicle. In this case, the vehicle-specific counting results and the measured values ​​may not match, which is problematic. Regarding the classification of large and small vehicles, machine learning is used to learn whether a vibration peak is from a large or small vehicle, and the classification is performed based on the results of the machine learning.

[0057] The control unit invalidates the later vibration peak if two vibration peaks identified as being caused by passing vehicles are located adjacent to each other within one second. In this case, the control unit invalidates the later peak only if the vehicle type identification result for both peaks is the same (both are either large vehicles or small vehicles). The control unit does not count the invalidated vibration peaks as large vehicles or small vehicles. In other words, the count result for the invalidated vibration peaks becomes invalid.

[0058] Specifically, if two or more vibration peaks identified as vehicle vibrations are adjacent, the control unit compares the occurrence times of the first vibration peak with the next vibration peak. The control unit invalidates the later vibration peak only if the difference in occurrence times between the two peaks is within 1 second and the identification results for both peaks are the same. Therefore, the counting results from the invalidated vibration peaks are invalid. Furthermore, the control unit does not compare the occurrence times of the invalidated vibration peaks with adjacent peaks.

[0059] In other words, the control unit detects vibration peaks in the vibration data from the vibration sensor. Next, the control unit classifies the detected vibration peaks into large vehicles or small vehicles (type of vehicle (mobile object)). Then, if other vibration peaks exist within a predetermined time from the target vibration peak used to determine whether or not to detect a peak classified as a mobile object, the control unit detects one of the vibration peaks of the same type as a vehicle (mobile object). For example, if there is one vibration peak before or after a predetermined time from the target vibration peak, and one vibration peak before or after the predetermined time, that is, if there are three vibration peaks from before to after a predetermined time from the time of the target vibration peak, and all vibration peaks are of the same type of vehicle (for example, all are large vehicles), the control unit will detect one of the vibration peaks as a vehicle.

[0060] For example, if the control unit detects the target vibration peak as a vehicle and there are other vibration peaks within a predetermined time interval from the target vibration peak (for example, a vibration peak classified as a large vehicle), and the other vibration peaks (for example, vibration peaks classified as large vehicles) are of the same type as the target vibration peak (for example, the same as a large vehicle), the control unit will not detect the other vibration peaks as vehicles, but will detect the target vibration peak as a vehicle. In other words, as described above, if there is one vibration peak before or after a predetermined time and one vibration peak before or after a predetermined time interval centered on the target vibration peak, that is, if there are three vibration peaks between a predetermined time interval and a predetermined time interval centered on the time of the target vibration peak, and all vibration peaks are of the same type of vehicle, the control unit will detect the target vibration peak (the vibration peak in the middle of the time interval) as a vehicle.

[0061] As described above, if there is one other vibration peak within a predetermined time (1 second) after the target vibration peak (i.e., two vibration peaks are adjacent), and the other vibration peak is of the same type as the target vibration peak, the control unit will not detect the other vibration peak (the vibration peak that is later in time) as a vehicle, but will detect the target vibration peak (the vibration peak that is earlier in time) as a vehicle.

[0062] The control unit disables other vibration peaks, thereby preventing them from being detected as vehicles. The control unit then detects the target vibration peak as a vehicle.

[0063] The following describes the processing operation of the moving object detection system based on the flowchart shown in Figure 2. The control unit of the moving object detection system sets N, which indicates which peak it is, to 1 (N←1) (S11). Next, the control unit determines whether N is less than the total number of peaks (N < total number of peaks) (S12). If the control unit determines that N is less than the total number of peaks (S12:Yes), it checks the inference result for the Nth peak (S13). Next, the control unit determines whether the Nth peak is a peak that has been determined to be a valid vehicle passage (S14).

[0064] If the control unit determines that the Nth peak is a peak that has been determined to be a valid vehicle passing (S14: Yes), it checks the N+1th peak (S15). Next, the control unit determines whether the N+1th peak existed within 1 second of the Nth peak, was determined to be a vehicle passing, and is of the same vehicle type as the Nth peak (S16). If the control unit determines that the N+1th peak existed within 1 second of the Nth peak, was determined to be a vehicle passing, and is of the same vehicle type as the Nth peak (S16: Yes), it invalidates the N+1th peak (S17).

[0065] After processing S17, the control unit determines that the Nth peak is not a peak determined to be a valid vehicle passage (S14: No), or that the N+1th peak does not exist within 1 second of the Nth peak, is not a peak determined to be a vehicle passage, or is not the same vehicle type as the Nth peak (S16: No), then it sets N to N+1 to check the N+1th peak (N←N+1) (S18). After processing S18, the control unit executes the process in S12.

[0066] If the control unit determines that N is not less than the total number of peaks, i.e., N is the total number of peaks (N = total number of peaks) (S12: No), it counts the peaks that were determined to be invalid large vehicles or small vehicles (S19).

[0067] The mobile object detection system according to this embodiment can reduce cases where peaks are invalidated between different vehicle types. In field tests using the mobile object detection system according to this embodiment, two cases were reduced where vibration peaks identified as large vehicles invalidated vibration peaks identified as small vehicles, and one case was reduced where vibration peaks identified as small vehicles invalidated vibration peaks identified as large vehicles.

[0068] As explained above, in this embodiment, the control unit of the moving object detection system, when determining whether or not to detect a peak classified as a type of moving object (type of vehicle, e.g., large vehicle or small vehicle), detects one of the peaks of the same type as the moving object if other peaks exist within a predetermined time from the target peak. Therefore, for example, if the moving object is a vehicle, even if moving objects of different types, such as a large vehicle and a small vehicle, are detected consecutively, neither will be invalidated, and each will be detected as a moving object. As a result, false detection of moving objects (vehicles) is prevented, and the movement of moving objects can be detected with high accuracy.

[0069] In this embodiment, after confirming whether each vibration peak is invalid or not, vibration peaks that are determined to be valid are counted as vehicles (moving objects). Alternatively, as described below, the determination of whether a vibration peak is invalid or not may be made when counting vehicles.

[0070] The control unit detects vibration peaks in the vibration data from the vibration sensor. Next, the control unit classifies the detected vibration peaks into large vehicles or small vehicles (type of vehicle (mobile object)). When the control unit counts the vibration peaks classified by vehicle type for each vehicle type, if there are other vibration peaks within a predetermined time from the vibration peak that determines whether or not to count a peak classified as a mobile object, the control unit counts one of the vibration peaks of the same type as a vehicle (mobile object). For example, if there is one vibration peak before or after a predetermined time from the target vibration peak, and one vibration peak before the predetermined time, that is, if there are three vibration peaks from before to after a predetermined time from the time of the target vibration peak, the control unit counts one of the vibration peaks of the same type as a vehicle.

[0071] For example, if another vibration peak exists within a predetermined time from the target vibration peak, and that other vibration peak (for example, a vibration peak classified as a large vehicle) is of the same type as the target vibration peak (for example, a vibration peak classified as a large vehicle), the control unit will not count the other vibration peak as a vehicle, but will count the target vibration peak as a vehicle. In other words, as described above, if there is one vibration peak before or after a predetermined time from the target vibration peak, and one vibration peak before or after a predetermined time, that is, if there are three vibration peaks from before or after a predetermined time from the time of the target vibration peak, and all of them are vibration peaks of the same type, the control unit will count the target vibration peak (the vibration peak in the middle in terms of time) as a vehicle.

[0072] As described above, if there is one other vibration peak within a predetermined time (1 second) after the target vibration peak (i.e., two vibration peaks are adjacent), and the other vibration peak is of the same type as the vibration peak classified as a vehicle type, the control unit will not count the other vibration peak (the vibration peak that is later in time) as a vehicle, but will count the target vibration peak (the vibration peak that is earlier in time) as a vehicle.

[0073] The control unit disables other vibration peaks, thereby preventing them from being counted as part of the vehicle.

[0074] As explained above, when the control unit counts peaks classified by type of moving object for each type of moving object, if other peaks exist within a predetermined time from the target peak for determining whether or not to count a peak classified by type of moving object as a moving object, the control unit counts one of the peaks of the same type as the moving object. Therefore, for example, if the moving object is a vehicle, even if moving objects of different types, such as a large vehicle and a small vehicle, are detected consecutively, neither will be invalidated, and each will be detected (counted) as its respective moving object. As a result, false detection (counting) of moving objects is prevented, and the movement (number) of moving objects can be detected (counted) with high accuracy.

[0075] Furthermore, the classification of vibration and noise, as well as the classification of large and small vehicles, in the vibration data mentioned above are based on the magnitude of the vibration data.

[0076] Although embodiments of the present invention have been described above, the applicable forms of the present invention are not limited to those described above, and modifications can be made as appropriate without departing from the spirit of the invention. [Industrial applicability]

[0077] The present invention can be suitably used in a mobile object detection system and a mobile object detection method for detecting the movement of a moving object.

Claims

1. A vibration sensor that detects vibrations transmitted to an object to be installed and outputs vibration data, It comprises a control unit and, The control unit, Based on the vibration data output from the vibration sensor, feature quantities are extracted, and the extracted feature quantities are classified as vibrations caused by the movement of a moving object or other factors. A moving object detection system characterized by detecting one of the candidates classified as vibrations caused by the movement of a moving object as the movement of a moving object if other candidates exist within a predetermined time.

2. The moving object detection system according to claim 1, characterized in that, if other candidates exist within a predetermined time from the candidate to be determined as the moving object, the control unit does not detect the other candidates as the moving object, but detects the candidate to be determined as the moving object.

3. The moving object detection system according to claim 1, characterized in that if, within a predetermined time period from the candidate to be determined as a moving object, there is one other candidate, the control unit does not detect the other candidate as moving object, but detects the candidate to be determined as moving object.

4. The moving object detection system according to claim 2 or 3, characterized in that the control unit does not detect other candidates as moving objects by disabling other candidates.

5. A vibration sensor that detects vibrations transmitted to an object to be installed and outputs vibration data, It comprises a control unit and, The control unit, Based on the vibration data output from the vibration sensor, feature quantities are extracted, and the extracted feature quantities are classified as vibrations caused by the movement of a moving object or other factors. The number of vibrations classified as being caused by the movement of a moving object is counted as the number of moving objects. A mobile object detection system characterized in that, when counting the number of vibrations classified as being caused by the movement of a mobile object as the number of mobile objects, if other candidates exist within a predetermined time from the candidates classified as vibrations caused by the movement of a mobile object, one of those candidates is counted as a mobile object.

6. The moving object detection system according to claim 5, characterized in that, if other candidates exist within a predetermined time from the candidate to be determined as a moving object, the control unit does not count the other candidates as moving objects, but counts the candidate to be determined as a moving object.

7. The moving object detection system according to claim 5, characterized in that if, within a predetermined time period from the candidate to be determined as a moving object, there is one other candidate, the control unit does not count the other candidate as a moving object, but counts the candidate to be determined as a moving object.

8. The mobile object detection system according to claim 6 or 7, characterized in that the control unit does not count other candidates as mobile objects by disabling them.

9. A vibration sensor that detects vibrations transmitted to an object to be installed and outputs vibration data, It comprises a control unit and, The control unit, Based on the vibration data output from the vibration sensor, feature quantities are extracted, and the extracted feature quantities are classified as vibrations caused by the movement of a moving object or other factors. Vibrations caused by the movement of a moving object are classified by the type of moving object. A mobile object detection system characterized by detecting one candidate of the same type as the mobile object if other candidates exist within a predetermined time from a candidate classified as a type of mobile object.

10. The moving object detection system according to claim 9, characterized in that, if, within a predetermined time from the candidate to be determined as a moving object, another candidate exists and the other candidate is of the same type as the candidate to be determined, the control unit does not detect the other candidate as a moving object, but detects the candidate to be determined as a moving object.

11. The moving object detection system according to claim 9, characterized in that, if, within a predetermined time period from the candidate to be determined as a moving object, there is one other candidate and the other candidate is of the same type as the candidate to be determined, the control unit does not detect the other candidate as a moving object, but detects the candidate to be determined as a moving object.

12. The moving object detection system according to claim 10 or 11, characterized in that the control unit does not detect other candidates as moving objects by disabling other candidates.

13. A vibration sensor that detects vibrations transmitted to an object to be installed and outputs vibration data, It comprises a control unit and, The control unit, Based on the vibration data output from the vibration sensor, feature quantities are extracted, and the extracted feature quantities are classified as vibrations caused by the movement of a moving object or other factors. Vibrations caused by the movement of a moving object are classified by the type of moving object. The candidates classified by type of mobile object are counted for each type of mobile object. A mobile object detection system characterized in that, when counting candidates classified by type of mobile object, if other candidates exist within a predetermined time from the candidate classified by type of mobile object, one of the candidates of the same type is counted as the mobile object.

14. The moving object detection system according to claim 13, characterized in that, if, within a predetermined time from the candidate to be determined for determining whether or not it is the movement of a moving object, another peak exists and the other peak is of the same type as the candidate to be determined, the control unit does not count the other candidate as a moving object, but counts the candidate to be determined as a moving object.

15. The mobile object detection system according to claim 13, characterized in that if, within a predetermined time period from the candidate to be determined, one other candidate exists and the other candidate is of the same type as the candidate to be determined, the control unit does not count the other candidate as a mobile object, but counts the candidate to be determined as a mobile object.

16. The moving object detection system according to claim 14 or 15, characterized in that the control unit does not count other candidates as moving objects by invalidating them.

17. The moving object is a vehicle. The mobile body detection system according to any one of claims 9 to 16, characterized in that the control unit classifies vibrations caused by the movement of a mobile body into large vehicles or small vehicles.

18. Based on the vibration data output from a vibration sensor that detects vibrations transmitted to an object to be installed and outputs vibration data, feature quantities are extracted, and the extracted feature quantities are classified as vibrations caused by the movement of a moving object or other, A method for detecting moving objects, characterized in that if other candidates exist within a predetermined time from a candidate classified as vibration caused by the movement of a moving object, one of the candidates is detected as the movement of a moving object.

19. Based on the vibration data output from a vibration sensor that detects vibrations transmitted to an object to be installed and outputs vibration data, feature quantities are extracted, and the extracted feature quantities are classified as vibrations caused by the movement of a moving object or other, The number of vibrations classified as being caused by the movement of a moving object is counted as the number of moving objects. A method for detecting moving objects, characterized in that when counting the number of vibrations classified as being caused by the movement of a moving object as the number of moving objects, if other candidates exist within a predetermined time from the candidates classified as being caused by the movement of a moving object, one of the candidates is counted as a moving object.

20. Based on the vibration data output from a vibration sensor that detects vibrations transmitted to an object to be installed and outputs vibration data, feature quantities are extracted, and the extracted feature quantities are classified as vibrations caused by the movement of a moving object or other, Vibrations caused by the movement of a moving object are classified by the type of moving object. A mobile object detection system characterized by detecting one candidate of the same type as the mobile object if other candidates exist within a predetermined time from a candidate classified as a type of mobile object.

21. Based on the vibration data output from a vibration sensor that detects vibrations transmitted to an object to be installed and outputs vibration data, feature quantities are extracted, and the extracted feature quantities are classified as vibrations caused by the movement of a moving object or other, Vibrations caused by the movement of a moving object are classified by the type of moving object. The candidates classified by type of mobile object are counted for each type of mobile object. A method for detecting moving objects, characterized in that when counting candidates classified into the type of moving object for each type of moving object, if other candidates exist within a predetermined time from the candidate classified into the type of moving object, one of the candidates of the same type is counted as the moving object.

Images