Traffic volume estimation device
The traffic volume estimation device accurately estimates traffic volume by combining image analysis for vehicle types with connected vehicle data and traffic counter measurements, addressing the limitations of existing methods.
Patent Information
- Application Number
- JP2023071153
- Authority / Receiving Office
- JP · JP
- Patent Type
- Applications
- Current Assignee / Owner
- Filing Date
- 2023-04-24
- Publication Date
- 2024-11-06
- Estimated Expiration
- Not applicable · inactive patent
AI Technical Summary
Existing traffic volume estimation methods lack accuracy due to the absence of GPS-equipped vehicles, which limits the estimation of vehicle types and overall traffic volume.
A traffic volume estimation device that calculates vehicle type ratios through image analysis and combines this with real-time data from connected vehicles and traffic counters to estimate traffic volume accurately.
Enables accurate and real-time estimation of traffic volume by integrating vehicle type ratios derived from image analysis with connected vehicle data and traffic counter measurements.
Smart Images

Figure 2024156569000001_ABST
Abstract
Description
[Technical field]
[0001] The present invention relates to a traffic volume estimation device that accurately estimates traffic volume based on data from vehicles and measurement results from observation equipment such as a traffic counter. [Background technology]
[0002] Patent Document 1 discloses an invention that uses driving data from a vehicle equipped with an on-board unit that collects location data using a Global Positioning System (GPS) and transmits it to a server together with a vehicle code to calculate traffic volume in a specified area of a road and estimates traffic volume for each vehicle type by associating it with the vehicle's master information. [Prior art documents] [Patent documents]
[0003] [Patent Document 1] JP 2019-091170 A Summary of the Invention [Problem to be solved by the invention]
[0004] However, since not all vehicles traveling on public roads are equipped with on-board devices capable of transmitting GPS location information and vehicle master information, the invention described in Patent Document 1 had a problem in terms of the accuracy of estimating traffic volume.
[0005] In consideration of the above, an object of the present invention is to provide a traffic volume estimation device that accurately estimates traffic volume based on data from vehicles and measurement results from observation equipment such as a traffic counter. [Means for solving the problem]
[0006] In order to achieve the above object, the traffic volume estimation device described in claim 1 includes: a vehicle type ratio determination unit that calculates a vehicle type ratio indicating the ratio of vehicle types at an observation point by image analysis of image data captured from above the observation point; an estimated traffic volume calculation unit that calculates an estimated traffic volume at the observation point by multiplying a total traffic volume measured at the observation point by a rate of change in the number of connected vehicles over time at the observation point, the rate of change being calculated based on driving data acquired in time series from connected vehicles capable of communicating with the outside world; and a vehicle type estimated traffic volume calculation unit that calculates an estimated traffic volume by vehicle type by multiplying the estimated traffic volume by the vehicle type ratio.
[0007] According to the traffic volume measuring device of claim 1, the traffic volume can be estimated with high accuracy in real time based on data from vehicles and measurement results from observation equipment such as a traffic counter. Effect of the Invention
[0008] As described above, the traffic volume estimation device according to the present invention can estimate traffic volume with high accuracy based on data from vehicles and measurement results from observation equipment such as a traffic counter. [Brief description of the drawings]
[0009] [Figure 1] 1 is a schematic diagram showing an example of a specific configuration of a traffic volume estimation device according to an embodiment of the present invention. [Diagram 2] FIG. 2 is a functional block diagram showing an example of processing of the traffic volume estimation device according to the present embodiment. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS
[0010] Hereinafter, a traffic volume estimation device 100 according to this embodiment will be described with reference to Fig. 1. Fig. 1 is a schematic diagram showing an example of a specific configuration of the traffic volume estimation device 100 according to this embodiment. As shown in Fig. 1, the traffic volume estimation device 100 according to this embodiment includes an air vehicle 10 such as an artificial satellite or an airplane that acquires image data of an observation point from the sky, an image DB (database) 12 that stores image data acquired by the air vehicle 10, a vehicle type determination server 14 that determines the type of vehicle recorded in the image data by referring to the image DB 12, a vehicle type ratio DB 16 that stores the vehicle type ratio that is the determination result by the vehicle type determination server 14, a vehicle 20 that is a so-called connected vehicle capable of communicating with the outside world, and vehicle 20 information acquired from the vehicle 20 through communication. the vehicle data DB22 that accumulates driving data of the vehicle, an observation device 30 that measures traffic volume such as a traffic counter, a traffic volume DB32 that stores information on the traffic volume measured by the observation device 30, an estimated traffic volume calculation server 40 that calculates estimated traffic volumes by referring to the vehicle data DB22 and the traffic volume DB32, an estimated traffic volume DB42 that stores the calculated estimated traffic volumes, a vehicle type estimated traffic volume calculation server 50 that estimates traffic volumes by vehicle type by referring to the vehicle type ratio DB16 and the estimated traffic volume DB42, and a vehicle type estimated traffic volume DB52 that stores the estimated traffic volumes by vehicle type.
[0011] 2 is a functional block diagram showing an example of the process of the traffic volume estimation device 100 according to this embodiment. In step S1, ground image data is acquired from the flying object 10, and the acquired image data is stored in the image DB 12.
[0012] In step S2, the vehicle type determination server 14 determines the type of vehicle, such as a car, truck, or bus, contained in the data of the single image stored in the image DB 12 by image analysis. The determination is made based on the shape and size of pixels indicating the vehicle recorded in the data of the single image. As an example, the vehicle type determination server 14 determines the type of vehicle in the image data by using a machine-learned model constructed by machine learning using a mathematical model such as a CNN (Convolutional Neural Network) using image data in which the type of vehicle has been specified in advance as training data.
[0013] In step S3, the vehicle type determination server 14 marks the vehicle type determined in the image data, and in step S4, the vehicle type determination server 14 calculates the vehicle type ratio. Then, in step S5, the calculated vehicle type ratio is stored in the vehicle type ratio DB 16.
[0014] In step S6, data on the total vehicle traffic volume is acquired by the observation equipment 30, such as a traffic counter. Then, in step S7, the data on the total vehicle traffic volume acquired in step S6 is stored in the traffic volume DB 32.
[0015] In step S8, at time t1, the total vehicle traffic volume in the target range, which is the observation point, is extracted from the traffic volume DB 32. Then, in step S9, the total vehicle traffic volume in the target range is stored in the traffic volume DB 32. As an example, the target range is set by dividing a region into a predetermined mesh unit.
[0016] In step S10, travel data of the vehicle 20 capable of communicating with the outside world is acquired. The travel data of the vehicle 20 acquired in step S10 is, for example, information indicating the position information, speed, traveling direction, and type of the vehicle 20 of the vehicle 20.
[0017] In step S11, the driving data of the vehicle 20 is stored in the vehicle data DB 22. Then, in step S12, the driving data of the vehicle 20 within the target range is extracted from the vehicle data DB 22 at time t1.
[0018] In step S13, the travel data of the vehicles 20 in the target range extracted at time t1 is stored in the vehicle data DB 22.
[0019] In step S14, at time t2, the travel data of the vehicles 20 in the target range is extracted from the vehicle data DB 22. Then, in step S15, the travel data of the vehicles 20 in the target range extracted at time t2 is stored in the vehicle data DB 22.
[0020] In step S16, the estimated traffic volume calculation server 40 starts a target range traffic volume estimation process that calculates an estimated traffic volume in the target range by referring to the vehicle data DB 22 and the traffic volume DB 32. The target range traffic volume estimation process calculates a rate of change in traffic volume over time from the change between the traffic volume in the target range of the vehicle 20 at time t1 and the traffic volume in the target range of the vehicle 20 at time t2, each of which is acquired by wireless communication. Then, the latest traffic volume in the target area is estimated by multiplying the calculated rate of change by the total vehicle traffic volume in the target range detected by the observation device 30 such as a traffic counter.
[0021] In step S17, the estimated traffic volume calculation server 40 calculates the estimated traffic volume of the target range at time t3 by processing such as multiplying the total vehicle traffic volume of the target range extracted at time t1 in step S8 by the aforementioned rate of change.
[0022] In step S18, the vehicle type estimated traffic volume calculation server 50 multiplies the estimated traffic volume in the target area calculated in step S17 by the vehicle type number ratio calculated in step S5 to calculate the estimated number of vehicles by vehicle type.
[0023] In step S19, the calculated estimated number of vehicles by vehicle type is stored in the vehicle type estimated traffic volume DB 52 at time t4, and the process ends.
[0024] As described above, according to this embodiment, the traffic volume in the target range is calculated by multiplying the total vehicle traffic volume in the target range detected by the observation device 30 such as a traffic counter by the rate of change in the number of vehicles 20 in the target range over time calculated over time based on vehicle driving data acquired from vehicles 20 capable of wireless communication with the outside world. Furthermore, the estimated number of vehicles by vehicle type in the target range can be accurately estimated by multiplying the calculated traffic volume in the target range by the vehicle type ratio calculated from ground images acquired by the flying object 10.
[0025] Although the observation equipment 30 such as a traffic counter can grasp the total number of vehicles in the target area, it takes some time to count the number of vehicles, so it is not suitable for grasping the latest information in real time. Also, although the running data of the vehicles 20 equipped with communication functions can be acquired in real time, it only constitutes a part of the total number of vehicles.
[0026] In this embodiment, the traffic volume in the target range is estimated in real time based on the rate of change over time in the number of vehicles 20 capable of wireless communication, which allows the number of vehicles to be determined in real time, in the target range, and the total vehicle traffic volume in the target range determined by the observation equipment 30.
[0027] Furthermore, by multiplying the estimated traffic volume by the vehicle type number ratio in the target range detected from the image, the estimated number of vehicles by vehicle type in the target range can be calculated.
[0028] In this embodiment, the vehicle type ratio is calculated from the image acquired by the flying object 10, but the present invention is not limited to this. For example, the vehicle type ratio in the target range may be calculated based on data acquired from a vehicle 20 with which communication is possible. In addition, each process in the vehicle type ratio determination server 14, the estimated traffic volume calculation server 40, and the vehicle type estimated traffic volume calculation server 50 may be realized by one server. [Explanation of symbols]
[0029] 10 Projectile 14 Vehicle type determination server 20 Vehicles 30 Observation Equipment 40 Traffic volume calculation server 50 Vehicle type estimated traffic volume calculation server 100 Traffic volume estimation device
Claims
[Claim 1] a vehicle type ratio determination unit that calculates a vehicle type ratio indicating a ratio of a vehicle type at an observation point by image analysis of image data obtained by capturing an image of the observation point from above; an estimated traffic volume calculation unit that calculates an estimated traffic volume at the observation point by multiplying a rate of change in the number of connected vehicles over time at the observation point, the rate of change being calculated based on driving data acquired in time series from connected vehicles capable of communicating with the outside world, by a total traffic volume measured at the observation point; a vehicle type estimated traffic volume calculation unit that calculates a vehicle type estimated traffic volume by multiplying the estimated traffic volume by the vehicle type ratio; A traffic volume estimation device comprising:
Citation Information
Patent Citations
Management system
JP2019091170A
