System and method

Abstract

To enable easy labeling of measured data. [Solution] System 1 comprises a vibration sensor 3 that outputs vibration data, a switch 4 having an operator, and a PC 2 that synchronously records information corresponding to the operation of the operator on the switch 4 to the labeling vibration data output from the vibration sensor 3. The PC 2 generates labeling data based on the labeling vibration data for which information has been recorded in synchronous order. The PC 2 generates a learning model based on the measurement data for the learning model output from the vibration sensor 3 and the labeling data.

Description

Technical Field

[0001] The present invention relates to a system and a method for counting moving objects such as vehicles and detecting their types.

Background Art

[0002] The applicant has filed an invention related to a system that detects the movement of a moving object (e.g., a vehicle) based on vibration data measured by a vibration sensor and counts the number of detected moving objects (see, for example, Patent Document 1). When classifying a detected vehicle into, for example, a large vehicle or a small vehicle (vehicle type discrimination), it is necessary to give a feature amount and correct data corresponding to the feature amount in advance to make it possible to classify from the feature amount. In this case, as the correct data, it is necessary to photograph the moving vehicle with a camera and create labeling data indicating when and what type of vehicle has passed from the photographed video. Hereinafter, adding information to the measured data (e.g., vibration data) is called labeling.

Prior Art Documents

Patent Documents

[0003]

Patent Document 1

Summary of the Invention

Problems to be Solved by the Invention

[0004] When performing the above-described labeling using a camera, labor and cost are required.

[0005] An object of the present invention is to enable easy labeling of measured data.

Means for Solving the Problems

[0006] The system of the first invention is characterized by comprising a sensor that outputs measurement data, a switch having an operator, and a control unit that synchronously records information corresponding to the operation of the operator on the switch to the labeling measurement data output from the sensor.

[0007] In this invention, information corresponding to the operation of the switch's control element (for example, the type of moving object) is recorded synchronously with the labeling measurement data output from the sensor. This makes it easy to label the measured data.

[0008] Furthermore, since there is no need to use a camera for labeling, labor and costs can be reduced.

[0009] The system of the second invention is characterized in that, in the system of the first invention, the control unit generates labeling data based on the labeling measurement data recorded synchronously with the information.

[0010] The system of the third invention is characterized in that, in the system of the second invention, the control unit generates a learning model based on the measurement data for the learning model output from the sensor and the labeling data.

[0011] The system of the fourth invention is characterized in that, in the system of the third invention, the control unit generates the learning model for detecting a moving object.

[0012] The system of the fifth invention is characterized in that, in the system of the fourth invention, the control unit detects the moving object based on the measurement data for detecting the moving object output from the sensor and the learning model.

[0013] The system of the sixth invention is characterized in that, in the system of the fifth invention, the control unit counts the number of detected moving objects.

[0014] The system of the seventh invention is characterized in that, in the system of the third invention, the control unit generates the learning model for detecting the moving object and the type of the moving object.

[0015] The system of the eighth invention is characterized in that, in the system of the seventh invention, the control unit detects the moving object and the type of the moving object based on the measurement data for moving object detection output from the sensor and the learning model.

[0016] The system of the ninth invention is characterized in that, in the system of the eighth invention, the control unit counts the number of detected moving objects and the number of types of moving objects.

[0017] The system of the tenth invention is characterized in that, in the system of the eighth invention, the number of operators is equal to the number of types of moving bodies.

[0018] The system of the eleventh invention is characterized in that, in the system of the eighth invention, the moving body is a vehicle, and the type of the moving body is the size of the vehicle.

[0019] The system of the twelfth invention is characterized in that, in the system of the first invention, the sensor measures vibration and outputs the measured vibration as measurement data.

[0020] The method of the twelfth invention is characterized by synchronously recording information corresponding to the operation of an operator on a switch with the labeling measurement data output from the sensor. [Effects of the Invention]

[0021] According to the present invention, measured data can be easily labeled. [Brief explanation of the drawing]

[0022] [Figure 1] This diagram illustrates the configuration of a system according to an embodiment of the present invention.

Embodiment for Carrying Out the Invention

[0023] Hereinafter, embodiments of the present invention will be described. FIG. 1 is a diagram showing a configuration example of a system 1 according to an embodiment of the present invention. The system 1 is composed of, for example, a personal computer (hereinafter referred to as "PC") 2, a vibration sensor 3, a switch 4, etc. The system 1 detects the movement and type of a vehicle (moving body) based on vibration (measurement) data detected by the vibration sensor 3, and counts the number of detected vehicles and the types of vehicles. The PC 2 (control unit) executes generation of labeling (correct) data, generation of a learning model, vehicle counting, vehicle classification, etc.

[0024] The vibration sensor 3 is installed, for example, on a road or the like. The vibration sensor 3 detects vibration transmitted from a vehicle passing by the periphery of the vibration sensor 3 to the road, and outputs vibration data. The switch 4 has a plurality (for example, n) of operators 4a to 4n. The operator is, for example, a button. The number of operators matches the number of types of vehicles to be classified. For example, if the types of vehicles to be classified are four types: large four-wheel vehicle, medium four-wheel vehicle, small four-wheel vehicle, and two-wheeled vehicle, the number of operators is four.

[0025] Information such as the type of vehicle is added (labeled) to the vibration data output from the vibration sensor 3. Hereinafter, the vibration data before labeling is referred to as the vibration data for labeling. The vibration data after labeling is referred to as the labeling data. To generate the labeling data, the operator operates one of the controls on switch 4 in response to the passage of a vehicle. For example, when a large four-wheeled vehicle passes, the operator operates the corresponding control. A signal source is connected to switch 4, for example, and a signal corresponding to the control is output to PC2. When the operator operates the control for a large four-wheeled vehicle, a signal indicating a large four-wheeled vehicle is output to PC2. The signal may be, for example, a sound (do-re-mi), and is output to PC2 by the operation of the control. The sound may be a chord, not just a single note. In this case, multiple chords can be represented by combinations of single notes, thus reducing the number of sound-producing devices.

[0026] PC2 records the vibration data for labeling output from vibration sensor 3 in synchronization with the signal (information) corresponding to the operation of the control element on switch 4. In other words, PC2 records the vibration data for labeling output from vibration sensor 3 and the signal from switch 4 (signal source) at the same time. PC2 generates labeling data based on the vibration data for labeling recorded in synchronization with the signals. That is, PC2 generates labeling data by adding information such as the vehicle type to the vibration data for labeling.

[0027] PC2 generates a learning model for detecting vehicles and vehicle types based on measurement data for the learning model output from vibration sensor 3 and labeling data. The generation of the learning model is performed, for example, before a traffic volume survey in which vehicles are actually counted. PC2 also detects vehicles and vehicle types based on vibration data for vehicle detection output from vibration sensor 3 (for example, vibration data detected at the actual traffic volume survey site) and the learning model. PC2 counts the number of detected vehicles and vehicle types.

[0028] (Variation 1) In the embodiment described above, sound is output to PC2 as a signal corresponding to the operation of the control element on switch 4. Alternatively, a high / low signal may be input to the input / output port (GPIO: General Purpose Input / Output) of the microcomputer as a signal corresponding to the operation of the control element. In this case, the same number of input / output ports as the number of control elements are used to receive the high / low signals from switch 4. The high / low signals are recorded in the microcomputer's memory as text data, etc., at the same time as the vibration data. PC2 generates labeling data based on this information and the vibration data for labeling. Note that the input to the input / output port may not be divided into multiple parts, and the content of the information (for example, information indicating a large four-wheeled vehicle) may be determined by the high / low pattern.

[0029] (Modification 2) In the embodiment described above, the number of operators on switch 4 is equal to the number of vehicle types. Alternatively, the number of operators on the switch may correspond to the number of lanes and the number of vehicle types. For example, if there are 2 lanes and 2 vehicle types, 4 operators may be provided. In this case, the operators correspond to, for example, large vehicles at the back of the lane, small vehicles at the back of the lane, large vehicles at the front of the lane, and small vehicles at the front of the lane. For example, when a large vehicle passes at the back of the lane, the operator operates the corresponding operator. A signal source is connected to switch 4, for example, and a signal having a frequency corresponding to the operator is output to PC2. When the operator operates the operator for a four-wheeled large vehicle, a signal indicating a four-wheeled large vehicle is output to PC2. PC2 generates labeling data based on this information and the vibration data for labeling.

[0030] As described above, in this embodiment, information corresponding to the operation of the control element of the switch 4 (for example, the type of moving object) is recorded synchronously with the labeling measurement data output from the vibration sensor 3. This makes it easy to label the measured data.

[0031] Furthermore, since there is no need to use a camera for labeling, labor and costs can be reduced.

[0032] 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]

[0033] The present invention can be suitably adopted in systems and methods for counting, detecting the type of, and so on, moving objects such as vehicles. [Explanation of Symbols]

[0034] 1 System 2 PC (Control Unit) 3. Vibration sensor 4 switches

Claims

1. A sensor that outputs measurement data, A switch having an operator, A control unit that synchronizes and records information corresponding to the operation of the switch's control element with the labeling measurement data output from the sensor, A system characterized by comprising the following features.

2. The system according to claim 1, characterized in that the control unit generates labeling data based on the labeling measurement data recorded in synchronization with the information.

3. The control unit, The system according to claim 2, characterized in that it generates a learning model based on the measurement data for the learning model output from the sensor and the labeling data.

4. The control unit, The system according to claim 3, characterized in that it generates the learning model for detecting moving objects.

5. The control unit, The system according to claim 4, characterized in that it detects the moving object based on the measurement data for detecting the moving object output from the sensor and the learning model.

6. The control unit, The system according to claim 5, characterized in that it counts the number of detected moving objects.

7. The control unit, The system according to claim 3, characterized in that it generates a learning model for detecting a moving object and the type of the moving object.

8. The control unit, The system according to claim 7, characterized in that it detects the moving object and the type of the moving object based on the measurement data for detecting a moving object output from the sensor and the learning model.

9. The control unit, The system according to claim 8, characterized in that it counts the number of detected moving objects and the number of types of moving objects.

10. The system according to claim 8, characterized in that the number of operators is equal to the number of types of the moving body.

11. The aforementioned moving object is a vehicle, The system according to claim 8, characterized in that the type of the moving body is the size of the vehicle.

12. The system according to claim 1, characterized in that the sensor measures vibration and outputs the measured vibration as measurement data.

13. A method characterized by synchronizing and recording information corresponding to the operation of an operator on a switch with the labeling measurement data output from a sensor.

Images