JP2022023863A - Congestion prediction system, terminal, congestion prediction method, and congestion prediction program - Google Patents

Abstract

PROBLEM TO BE SOLVED: To predict congestion at an arbitrary location by collecting images of mobile objects on a road.

SOLUTION: A congestion prediction device 100 collects images of mobile objects on a road captured by a plurality of terminals 110 and position information indicating positions at which the images were captured. An image processing unit 101 performs image processing on the collected images, and extracts traffic information, which is information relating to the mobile objects existing in the images. A traffic information processing unit 103 calculates information relating to congestion prediction for the road corresponding to the captured position using the traffic information extracted by the image processing unit 101, and generates congestion prediction information, which is then transmitted to the terminals 110.

SELECTED DRAWING: Figure 1

COPYRIGHT: (C)2022,JPO&INPIT

Description

The present invention relates to a traffic jam prediction system, a terminal device, a traffic jam prediction method, and a traffic jam prediction program for predicting traffic jams using images captured from a moving body. However, the use of the present invention is not limited to the traffic jam prediction system, the terminal device, the traffic jam prediction method, and the traffic jam prediction program.

Conventionally, as a technique for predicting traffic congestion on a road, a server that determines a traffic congestion situation at a vehicle position based on a speed from a vehicle and an average speed of the vehicle position has been disclosed (see, for example, Patent Document 1 below). In this technique, it is also described that an image captured by a vehicle is analyzed and used for determining traffic congestion.

In addition, there is a system for determining traffic congestion using images taken by a camera fixedly installed on the road.

Japanese Unexamined Patent Publication No. 2011-134195

However, with the above-mentioned conventional technique, it is not possible to predict a traffic jam at an arbitrary place. Further, in the configuration of the traffic jam judgment using the image, there is a problem that the installation cost is high and there is a limitation that only the traffic jam judgment of the place where the camera is fixedly installed can be performed.

In addition, from information such as speed received from the vehicle, it is determined whether or not the section of the road on which the vehicle is traveling is congested, and it is not possible to predict that the vehicle will be congested in the future. There is. The above-mentioned problems are given as an example of the problems to be solved by the present invention.

In order to solve the above-mentioned problems and achieve the object, the traffic jam prediction system according to the invention of claim 1 has an acquisition unit that acquires an image of a moving object on a road and position information indicating the position where the image was taken. Using the image processing unit that extracts traffic information, which is information about moving objects existing in the image, and the traffic information extracted by the image processing unit, traffic congestion prediction on the road corresponding to the photographed position. It is characterized by having a traffic information processing unit that calculates information about the vehicle.

Further, the terminal device according to the invention of claim 6 is a terminal that transmits the image to a traffic jam prediction device that predicts the traffic jam of the moving body on the road based on an image of a moving body on the road. An image pickup unit that captures an image of a moving object on a road, a processing unit that adds the captured position information to the image captured by the image pickup unit and transmits the captured position information to the congestion prediction device. It is characterized by having.

Further, the terminal device according to the invention of claim 7 is a terminal device connected to a traffic jam prediction device that predicts a traffic jam of the moving body on the road based on an image of a moving body on the road. It is characterized by having a processing unit that requests a traffic jam prediction device including a desired position and time to the traffic jam prediction device and outputs information related to the traffic jam prediction received from the traffic jam prediction device.

Further, the traffic jam prediction method according to the invention of claim 8 is a traffic jam prediction method carried out by a traffic jam prediction device, in which an acquisition step of acquiring an image of a moving object on a road and position information indicating a shot position is obtained. Using the image processing step of extracting traffic information, which is information about a moving object existing in the image, and the traffic information extracted by the image processing step, the traffic congestion prediction on the road corresponding to the photographed position is used. It is characterized by including a traffic information processing process for calculating information about the vehicle.

Further, the traffic jam prediction program according to the invention of claim 9 is characterized in that the computer executes the traffic jam prediction method described above.

FIG. 1 is a block diagram showing a configuration example of a traffic congestion prediction system according to an embodiment. FIG. 2 is a flowchart showing an example of the processing procedure of the congestion prediction system according to the embodiment. FIG. 3 is a block diagram showing an example of the hardware configuration of the navigation device. FIG. 4 is a block diagram showing an example of the hardware configuration of the server. FIG. 5 is a flowchart showing the processing contents of image transmission by the navigation device. FIG. 6 is a flowchart showing the processing contents of traffic information generation by the server. FIG. 7 is a chart showing an example of the contents of traffic information. FIG. 8 is a flowchart showing the processing contents of the traffic jam prediction using the traffic information. FIG. 9 is a flowchart showing the processing contents from the traffic jam prediction request to the transmission of the traffic jam prediction information. FIG. 10 is an explanatory diagram showing a state in which natural traffic congestion occurs.

(Embodiment)
Hereinafter, preferred embodiments of the traffic jam prediction system, the terminal device, the traffic jam prediction method, and the traffic jam prediction program according to the present invention will be described in detail with reference to the accompanying drawings.

FIG. 1 is a block diagram showing a configuration example of a traffic congestion prediction system according to an embodiment. The traffic jam prediction system according to the embodiment can be configured by a traffic jam prediction device 100 and a terminal device 110 that are communication-connected to each other.

The traffic jam prediction device 100 includes an image processing unit 101, an information coupling unit 102, a traffic information processing unit 103, a storage unit 104, and a traffic jam prediction information management unit 105. The terminal device 110 includes an image pickup unit 111, a positioning unit 112, and a processing unit 113.

The congestion prediction device 100 can be configured by using, for example, a single server, but is not limited to this, and may be configured by using a plurality of servers distributed according to the functions of the image processing unit 101 and the traffic information processing unit 103.

The terminal device 110 captures an image of the moving front by the image pickup unit 111 and transmits it to the traffic jam prediction device 100. At this time, the processing unit 113 transmits the shooting ID and the position information at the time of shooting positioned by the positioning unit 112 together with the shot image of the imaging unit 111. In addition, speed and time information at the time of movement may be transmitted. Further, the processing unit 113 requests the traffic jam prediction device 100 for the traffic jam prediction by a user operation or the like, processes the traffic jam prediction information transmitted from the traffic jam prediction device 100, and outputs the information to a display unit or the like (not shown).

The terminal device 110 transmits an image used for traffic jam prediction to the traffic jam prediction device 100 by configuring the navigation device provided on the mobile body, a smartphone or mobile phone owned by the user, a portable personal computer, or the like. Further, it is possible to request the traffic jam prediction device 100 to predict the traffic jam.

Not limited to these, the terminal device 110 may be configured by using a personal computer or the like fixedly installed in a home, a facility (store), or the like. In this case, the terminal device 110 omits the function of transmitting an image used for the traffic jam prediction to the traffic jam prediction device 100, and requests the traffic jam prediction device 100 to predict the traffic jam. In the case of such a terminal device 110, the functions of the image pickup unit 111 and the positioning unit 112 can be omitted, and the processing unit 113 of the terminal device 110 requests a traffic jam prediction by a user operation or the like, and is transmitted from the traffic jam prediction device 100. The information of the traffic jam prediction is output to the display unit or the like.

The image processing unit 101 of the traffic jam prediction device 100 processes the captured image transmitted from the terminal device 110. In this image processing, the road and the vehicle shown in the captured image are extracted (detected). Then, the traveling lane (own lane) in which the terminal device 110 (moving body) is traveling, the number of lanes, the number of vehicles for each lane, and the distance between vehicles are detected. In addition, it is also possible to detect the vehicle type of each vehicle included in the captured image.

The information coupling unit 102 stores each information of the shooting ID, the shooting time, and the shooting position of the shot image transmitted from the terminal device 110 in the storage unit 104 in association with each other. When the time of the traffic jam prediction device 100 is used, the time when the captured image is received can be used, and it is not necessary to transmit the time information on the terminal device 110 side. Further, when the traffic jam prediction device 100 grasps the speed of the moving body provided with the terminal device 110, the traffic jam prediction device 100 may, for example, determine the movement amount and the movement required time of the latitude and longitude of the terminal device 110. The velocity of the moving body can be obtained by the calculation based on.

Then, the information combining unit 102 associates (combines) the own lane information, which is the result of image processing by the image processing unit 101, with the same shooting ID, and stores the information in the storage unit 104 as traffic information. Further, the information coupling unit 102 stores the vehicle information (number of vehicles) for each lane for each shooting ID in the storage unit 104 as vehicle information for each lane. The traffic information is stored in the storage unit 104 in a state of being classified by road, position, and time zone.

The traffic information processing unit 103 calculates the probability of congestion occurrence for each position and time zone using a statistical function for each traffic information classified by the position and time zone of the road stored in the storage unit 104, and predicts the congestion. It is stored in the storage unit 104 as information. The storage unit 104 stores traffic information based on changes in a plurality of captured images on the same road, and the traffic information processing unit 103 uses the traffic information at the time of request from the terminal device 110 as the past traffic information ( For example, by comparing with the same time and place), the occurrence of traffic congestion is predicted.

The traffic information is not limited to information about the same road, but can be information within a predetermined range (other roads).

At this time, since traffic information for each lane is obtained based on the captured image, the degree of congestion is determined based on the number of vehicles for each lane (corresponding to the distance between vehicles and the density of vehicles), and the congestion prediction for each lane is made. It can be carried out.

The traffic jam prediction information includes future traffic jam prediction near the current position of the terminal device 110, traffic jam prediction on the travel route to the destination of the terminal device 110, and traffic jam prediction near the destination. The calculation process of the congestion prediction information may be calculated for each road (location) in advance, or may be calculated for the corresponding road (location) at the time of request from the terminal device 110.

Further, the traffic jam prediction information is not limited to the traffic jam prediction for the road corresponding to the photographed position. For example, it may be a road closest to the captured position information or a road on the map information that is map-matched at the time of capture.

The traffic jam prediction information management unit 105 manages the traffic jam prediction information. The traffic jam prediction information management unit 105 transmits the traffic jam prediction information of the position or area corresponding to the traffic jam prediction request from the terminal device 110 to the terminal device 110.

For example, the terminal device 110 requests the traffic jam prediction device 100 for traffic jam prediction information in a predetermined area near the travel route and the destination when the moving body is moving. In addition, when searching for a route by a moving body, traffic congestion prediction information in a predetermined area near the travel route or the destination is similarly requested. Correspondingly, the traffic jam prediction device 100 can extract the traffic jam prediction information in the predetermined area near the terminal device 110 (moving body) traveling route and the destination and transmit it to the terminal device 110.

FIG. 2 is a flowchart showing an example of the processing procedure of the congestion prediction system according to the embodiment. The processing content of the traffic jam prediction performed by the traffic jam prediction device 100 is shown. First, the traffic jam prediction device 100 acquires a photographed image and a position information, a speed, and a time at the time of photographing from the terminal device 110 (step S201). As described above, the speed and time can be acquired without receiving from the terminal device 110.

Next, the traffic jam prediction device 100 processes the captured image (step S202). By this image processing, the traveling lane (own lane) in which the terminal device 110 (moving body) is traveling, the number of lanes, the number of vehicles for each lane, the vehicle type of each vehicle, and the distance between vehicles are detected. The detected information is associated (combined) with the information (position information, speed, time) acquired in step S201 to generate traffic information and is stored in the storage unit 104.

The same or different plurality of terminal devices 110 transmit a plurality of captured images to the traffic jam prediction device 100 at different places and at different times. As a result, the traffic jam prediction device 100 repeatedly executes steps S201 and S202 to store a large amount of traffic information in a database for each road location and time. The traffic jam prediction device 100 uses this traffic information to perform the following traffic jam prediction processing.

Next, the traffic jam prediction device 100 waits for a traffic jam prediction request from the terminal device 110 (step S203: No loop). When the traffic jam prediction request is received (step S203: Yes), the traffic jam prediction process is performed (step S204). In this traffic jam prediction process, the traffic information generated in step S202 is referred to, and this place (range) and time are based on the occurrence of past traffic jams at the same place and time as the place (range) and time included in the request. Predict the occurrence of traffic congestion in.

The place where the occurrence of traffic congestion is predicted does not have to be exactly the same as the place included in the request, but the same road as the place included in the request or the past traffic information and the occurrence of traffic congestion within a predetermined range. You may make a prediction based on this. The same applies to the time, and a predetermined time zone including the time included in the request may be used.

After that, the traffic jam prediction information is transmitted to the terminal device 110 (step S205), and a series of traffic jam prediction processes are completed.

According to the above embodiment, in order to predict traffic congestion using images taken at a plurality of places and times on the road taken by a camera of the terminal device, a fixed installation such as a camera or a sensor on the road is used. It is possible not only to predict traffic congestion limited to the installation location using the above, but also to predict traffic congestion at any location on the road.

In addition, since a large amount of traffic information for determining a traffic jam state of a road can be acquired from images transmitted from a large number of terminal devices, the accuracy of traffic jam prediction at an arbitrary place can be improved.

As a terminal device, a general-purpose device such as a mobile phone, a smartphone, or a car navigation system equipped with a camera capable of taking an image and transmitting location information can be used, so that the traffic jam prediction device can easily acquire a large number of images and the system cost. Can also be reduced.

In addition, when generating congestion prediction information, it was decided to use the distance between vehicles and the number of vehicles, but in addition to this, the weather, the day of the day, the traffic information of other roads connected to the road at the requested location, and the requested. Other data that may cause congestion, such as the presence or absence of an event near the location, may be added as a parameter for congestion prediction. This makes it possible to further improve the accuracy of traffic congestion prediction.

Then, the traffic jam was determined based on the number of vehicles (inter-vehicle distance), but the images were processed by image processing the similarity between the image before the traffic jam at any place on the road and the current image transmitted from the terminal device. Congestion can be predicted in the same way by comparing the above.

Further, the image captured by the terminal device is not limited to a still image, but may be a moving image. If it is a video, it is possible to determine the speed of a moving object in another lane, so that more detailed traffic information can be obtained.

In addition, statistics such as multiple regression analysis may be applied to the generation of congestion prediction data. At this time, an appropriate optimum algorithm can be rearranged for each road.

Hereinafter, examples of the present invention will be described. In this embodiment, a configuration in which a navigation device 300 is mounted on a mobile body (vehicle) and each user's navigation device 300 accesses the server as the above-mentioned congestion prediction device 100 will be described. Here, the navigation device 300 has a function as the terminal device 110 described above, and photographs the road when the moving body is moving and transmits it to the traffic jam prediction device 100. Further, the navigation device 300 requests the traffic jam prediction device 100 (server) for traffic jam prediction information at a desired location.

(Hardware configuration of navigation device 300)
FIG. 3 is a block diagram showing an example of the hardware configuration of the navigation device. In FIG. 3, the navigation device 300 includes a CPU 301, a ROM 302, a RAM 303, a magnetic disk drive 304, a magnetic disk 305, an optical disk drive 306, an optical disk 307, an audio I / F (interface) 308, a microphone 309, a speaker 310, and an input device 311. It is equipped with a video I / F 312, a display 313, a communication I / F 314, a GPS unit 315, various sensors 316, and a camera 317. Each component 301 to 317 is connected by a bus 320.

The CPU 301 controls the entire navigation device 300. The ROM 302 records a boot program, a traffic jam prediction program, and the like. The RAM 303 is used as a work area of the CPU 301. That is, the CPU 301 controls the entire navigation device 300 by executing various programs recorded in the ROM 302 while using the RAM 303 as a work area.

The magnetic disk drive 304 controls reading / writing of data to the magnetic disk 305 according to the control of the CPU 301. The magnetic disk 305 records the data written under the control of the magnetic disk drive 304. As the magnetic disk 305, for example, an HD (hard disk) or an FD (flexible disk) can be used.

Further, the optical disk drive 306 controls reading / writing of data to the optical disk 307 according to the control of the CPU 301. The optical disc 307 is a detachable recording medium from which data is read out under the control of the optical disc drive 306. The optical disc 307 can also use a writable recording medium. As the removable recording medium, in addition to the optical disk 307, an MO, a memory card, or the like can be used.

Examples of the information recorded on the magnetic disk 305 and the optical disk 307 include map data, vehicle information, road information, travel history, and the like. Map data is used when searching for routes in car navigation systems, and background data that represents features such as buildings, rivers, ground surfaces, and energy supply facilities, and road shapes that represent road shapes with links and nodes. Vector data including data.

The voice I / F 308 is connected to a microphone 309 for voice input and a speaker 310 for voice output. The voice received by the microphone 309 is A / D converted in the voice I / F 308. The microphones 309 are installed, for example, on the dashboard of a vehicle, and the number of microphones 309 may be singular or plural. From the speaker 310, a voice obtained by D / A-converting a predetermined voice signal in the voice I / F 308 is output.

Examples of the input device 311 include a remote controller, a keyboard, and a touch panel having a plurality of keys for inputting characters, numerical values, various instructions, and the like. The input device 311 may be realized by any one form of the remote controller, the keyboard, and the touch panel, but it can also be realized by a plurality of forms.

The video I / F 312 is connected to the display 313. Specifically, the video I / F 312 is output from a graphic controller that controls the entire display 313, a buffer memory such as VRAM (Video RAM) that temporarily records image information that can be displayed immediately, and a graphic controller. It is configured by a control IC or the like that controls the display 313 based on the image data to be generated.

The display 313 displays various data such as icons, cursors, menus, windows, characters and images. As the display 313, for example, a TFT liquid crystal display, an organic EL display, or the like can be used.

The camera 317 captures an image of the road outside the vehicle. The video may be either a still image or a moving image. By photographing the road outside the vehicle with the camera 317, other vehicles traveling along with the road are photographed. This captured image is image-processed by the CPU 301 and then transmitted to the traffic jam prediction device 100 (server 400 described later) together with the position information of the GPS unit 315.

The communication I / F 314 is wirelessly connected to the network and functions as an interface between the navigation device 300 and the CPU 301. Communication networks that function as networks include in-vehicle communication networks such as CAN and LIN (Local Interconnect Network), public line networks, mobile phone networks, DSRC (Dedicated Short Range Communication), LAN, and WAN. The communication I / F 314 is, for example, a connection module for a public line, an ETC (non-stop automatic toll collection system) unit, an FM tuner, a VICS (Vehicle Information and Communication System: registered trademark) / beacon receiver, and the like.

The GPS unit 315 receives radio waves from GPS satellites and outputs information indicating the current position of the vehicle. The output information of the GPS unit 315 is used in calculating the current position of the vehicle by the CPU 301 together with the output values of various sensors 316 described later. The information indicating the current position is information that identifies one point on the map data, such as latitude / longitude and altitude.

The various sensors 316 output information for determining the position and behavior of the vehicle, such as a vehicle speed sensor, an acceleration sensor, an angular velocity sensor, and an inclination sensor. The output values of the various sensors 316 are used by the CPU 301 to calculate the current position of the vehicle and the amount of change in speed and direction.

The CPU 301 shown in FIG. 3 realizes the function of the processing unit 113 of the terminal device 110 shown in FIG. 1 by executing a program stored in the ROM 302 or the like. The camera 317 of FIG. 3 realizes the function of the image pickup unit 111 of FIG. 1, and the GPS unit 315 of FIG. 3 realizes the function of the positioning unit 112 of FIG.

(Server configuration example)
FIG. 4 is a block diagram showing an example of the hardware configuration of the server. The server 400 that constitutes the traffic congestion prediction device 100 also has the same configuration as the navigation device 300 shown in FIG. In the server 400, the GPS unit 315, various sensors 316, the camera 317, and the like shown in FIG. 3 are unnecessary.

The traffic jam prediction device 100 shown in FIG. 1 realizes a traffic jam prediction function by executing a predetermined program by the CPU 401 using a program or data recorded in a ROM 402 or the like provided in the server 400. The above traffic information and traffic congestion prediction information are stored in the magnetic disk 405 or the like of the server 400. Further, it communicates with the navigation device 300 via the communication I / F 414, and outputs the congestion prediction information corresponding to the request from the navigation device 300 to the navigation device 300.

Further, the server 400 may have a function of calculating the transmission time of the image transmitted from each navigation device 300 and the speed of the moving body on which the navigation device 300 is mounted. The server 400 can calculate the speed of the moving body by the calculation based on the movement amount of the latitude and longitude of the navigation device 300 and the movement required time.

The CPU 401 shown in FIG. 4 has an image processing unit 101, an information coupling unit 102, a traffic information processing unit 103, and a traffic jam prediction information management unit 105 of the traffic jam prediction device 100 shown in FIG. 1 by executing a program stored in the ROM 402 or the like. To realize the function of. The magnetic disk 305, the optical disk 307, and the like of FIG. 4 realize the function of the storage unit 104 shown in FIG.

(Image transmission process by navigation device)
FIG. 5 is a flowchart showing the processing contents of image transmission by the navigation device. The navigation device 300 always positions the position by the GPS unit 315 or the like when the moving body moves (step S501). Then, for example, the road is photographed by the camera 317 at predetermined timings (for example, every minute) (step S502). For example, by photographing the road ahead, the moving body on the road is photographed together with the road.

Further, the trigger (timing) for photographing the road by the camera 317 can be a request output from the server 400 side to the navigation device 300, an instruction from the user of the navigation device 300, or the like.

Then, the navigation device 300 transmits the position information at the time of shooting together with the captured image to the server 400 every time the image is captured (step S503). The navigation device 300 may assign a time to the image for each shooting and transmit a plurality of images to the server 400 at a predetermined time.

The navigation device 300 repeatedly captures an image as described above and transmits the image to the server 400. As a result, the server 400 can collect images of a plurality of roads taken by the plurality of navigation devices 300.

(Traffic information generation processing by the server)
FIG. 6 is a flowchart showing the processing contents of traffic information generation by the server. First, the server 400 receives the image transmitted from the navigation device 300 and the position information (shooting position) (step S601). Next, the speed and time of the moving body (navigation device 300) at the time of receiving the image are acquired (step S602).

The server 400 may acquire the speed and time of these moving objects transmitted by the navigation device 300, or may use those generated inside the server 400. When generating on the server 400 side, the time inside the server 400 at the time of receiving the image is used as the time. For the movement of the moving body, the speed of the moving body calculated by the calculation based on the movement amount of the latitude and longitude of the moving body (navigation device 300) and the movement required time is used.

Next, the server 400 detects the vehicle, the traveling lane, and the number of lanes from the acquired image by the image processing of the image processing unit 101 (step S603). At this time, for example, the captured image can be converted into an ortho image, the position and distance of the vehicle can be specified, and the number of vehicles for each lane can be detected. In addition, in the ortho image, the width of the own lane is derived, and the number of lanes and the distance between vehicles on the driving road are determined.

Next, the server 400 associates each information of position information, speed, time, number of vehicles for each lane, and own lane for each acquired image (ID) by the information coupling unit 102, and stores the information in the storage unit 104 as traffic information. Store (step S604). The above steps S601 to S604 are performed every time an image is received from the navigation device 300.

(Example of traffic information)
FIG. 7 is a chart showing an example of the contents of traffic information. FIG. 7A is photographed image processing information 701 obtained by image processing one image transmitted from one user. The captured image processing information 701 includes lane information (number of lanes), own lane information, and vehicle information ID for each lane. In the illustrated example, it is shown that the own vehicle is the first from the left (own lane information) of the three-lane road (lane information).

The details of the vehicle information ID for each lane in FIG. 7A are shown in 702 in FIG. 7B. The vehicle information ID 702 for each lane is assigned an ID for each image, and the position for each ID obtained by image processing (the lane position, which is the lane position of the moving body existing in the image) and each lane. It has each information of the distance between vehicles (the distance between a moving body existing in the image and the own vehicle). In the illustrated example, when ID = 1, a moving body exists at a position (lane 1) in front of the vehicle position with a vehicle distance of 45 m, and a vehicle distance of 30 m and a vehicle distance of 60 m at a position in front of the vehicle position (lane 3). Indicates that each moving object exists. In the illustrated example, of the three lanes, the position of the left lane is indicated by "1", the position of the center lane is indicated by "2", and the position of the right lane is indicated by "3".

The shooting condition information 703 shown in FIG. 7 (c) includes the shot image processing information 701 of FIG. 7 (a), the shooting position (latitude / longitude in the illustrated example), the speed, and the time, which are the conditions at the time of shooting. This is information associated with the shooting ID.

FIG. 7D is traffic information 704. This is information obtained by accumulating the shooting condition information 703 shown in FIG. 7 (c) for different places and times of the same user, and for different users. By increasing the number of lines of the traffic information 704, it is possible to construct a database of own lane information, lane information, and vehicle information ID (lane position, inter-vehicle distance) for each lane for a desired shooting position.

(Congestion prediction processing using traffic information)
FIG. 8 is a flowchart showing the processing contents of the traffic jam prediction using the traffic information. First, the traffic information processing unit 103 of the server 400 classifies the traffic information 704 by position, date, and time zone (step S801), and creates a statistical function for each classified data (step S802). In this state, the traffic information processing unit 103 waits for the input of the request condition included in the traffic jam prediction request from the navigation device 300 (step S803: No loop).

If the traffic information processing unit 103 has input the requirement condition (step S803: Yes), the traffic information processing unit 103 calculates the congestion occurrence value by using the statistical function corresponding to the requirement condition (step S804). Then, the traffic information processing unit 103 determines whether the congestion occurrence value is high (positive value) (step S805), and if the congestion occurrence value is higher than the predetermined value (step S805: Yes), the congestion occurrence value is used as the congestion occurrence value. It is generated as prediction information and stored in the storage unit 104 (step S806). If the congestion occurrence value is low (step S805: No), it is determined that there is no congestion occurrence, and the current process is terminated.

A specific example of the above processing will be described. In the above example, the occurrence of traffic congestion is determined by comparing the occurrence value of traffic congestion with a predetermined threshold value as the reliability. For example, the case where the requirement condition of the traffic jam prediction transmitted (input) from the navigation device 300 is specified as the left lane traffic volume, the right lane traffic volume, the date (day of the week), and the time zone will be described.

The navigation device 300 transmits the captured image, and the server 400 analyzes the image to determine the left lane traffic volume, the right lane traffic volume, the date (day of the week), and the time zone. In addition, the date and time will be calculated based on the data reception time of the server. An example of traffic is the number of cars per unit time.

First, the traffic information processing unit 103 refers to the traffic information 704 corresponding to the date (day of the week) and the time zone designated by the storage unit 104. For example, the requirement condition is x road, Tuesday, and the traffic information 704 that aggregates the data from 12:00 to 13:00 is referred to.

Then, the traffic information processing unit 103 uses the statistical function fn regarding the occurrence of traffic congestion and the requirement conditions X and Y as input values, and calculates the traffic congestion probability using the following calculation formula.
fn (X, Y) = an (X) + bn (Y)
fn (xn, yn) = an (xn) + bn (yn)
Congestion occurrence value = max (1 / fn (| X-xn |, | Y-yn |))-tn

However, the fn: statistical function (n = 0,1, ...) Statistical function fn is a function for comparing the past congestion forecast data with the current traffic volume to obtain the congestion occurrence value. X: Left lane input traffic volume, Y: Right lane input traffic volume, xn: Left lane statistical traffic volume, yn: Right lane statistical traffic volume, an: Left lane prediction coefficient, bn: Right lane prediction coefficient, tun: Threshold The value. The traffic volume is obtained from the speed of the moving body at the time of taking an image and the number of moving bodies taken.

When it is determined that a traffic jam has occurred on a certain road, the traffic jam forecast data goes back a predetermined time from the time when the traffic jam is judged on that road (the time to go back differs for each road), and the traffic information until the traffic jam occurs (for example, each lane). Traffic volume, speed and time). If this statistical data is significantly different from the usual traffic information, it will be the traffic congestion prediction data on this road. However, it has a large number of forecast data for one road for each parameter such as date and time. The above-mentioned determination of the occurrence of traffic congestion may be made by the congestion prediction device 100 of the system, and if the information is acquired from another external system, it may be determined as congestion when the distance between vehicles is narrow or the speed is low. to decide.

In addition, regarding the above statistical function fn, in the above example (two-lane road), there are two characteristics such as "when the right lane is extremely crowded, then traffic jam occurs" or "when both right and left are crowded to some extent, then traffic jam occurs". In the case of a road with a road, fn is n = 2 because there are two patterns of congestion occurrence. This feature is calculated from the congestion forecast data.

The right lane input traffic volume is the traffic volume in the right lane calculated from the image taken by the navigation device 300. It is indicated by the number of lines at position "3" (that is, the number of vehicles existing in the right lane of the three-lane road). The left lane input traffic volume is indicated by the number of lines at the position "1" in FIG. 7 (b) vehicle information ID 702 "1" for each lane (that is, the number of vehicles existing in the left lane of a three-lane road).

The right (left) lane statistical traffic volume is a statistical value of the average or mode of the traffic volume at position "3" in the corresponding pattern in the congestion prediction data. The right (left) lane prediction coefficient is a coefficient that weights the difference between the right (left) lane statistical traffic volume and the right lane input traffic volume. For example, if there is a pattern on the road that "when the right lane is extremely crowded, traffic jam will occur after that", the right lane prediction coefficient is increased. As a result, the difference in the right lane is greatly reflected in the final congestion occurrence value.

A specific example of the above-mentioned congestion prediction will be described. The traffic jam prediction device 100 says that "the right lane is extremely crowded" and "the right and left lanes are slightly crowded" in the statistical data (data of the storage unit 104) 30 minutes before the traffic jam occurs at the time t of a certain road. Suppose there are two patterns. It can be seen that a sign of traffic jam appears 30 minutes before this road. These two patterns are stored in the storage unit 104 as t-time congestion prediction data of this road.

Then, it is assumed that the image is transmitted from the user of the navigation device 300 at time t. It is assumed that the traffic jam prediction device 100 determines that the right lane is very crowded by image analysis of this image. It is assumed that the traffic jam prediction device 100 is very similar to the pattern of "extremely crowded right lane" when the similarity with the traffic jam prediction data is calculated by the statistical function for the image analysis result. In this case, the traffic information processing unit 103 of the traffic jam prediction device 100 determines that the traffic jam will occur 30 minutes after the image is sent, and tells the traffic jam prediction information management unit 105 that the road is congested 30 minutes later. Then it will be output.

A specific example of calculation by the statistical function of the traffic jam prediction device 100 will be described. This road has the following two patterns of traffic congestion forecasts, so
Pattern 1: "Congestion occurs when the right lane is extremely crowded"
Pattern 2: "Congestion occurs when both the right and left lanes are slightly crowded"
Therefore, two statistical functions fn are prepared.
f1 (X, Y) = a1 (x1) + b1 (y1)
f2 (X, Y) = a2 (x2) + b1 (y2)

x1: Left lane traffic volume in pattern 1 forecast data (average or most frequent) → Left lane statistical traffic volume y1: Right lane traffic volume in pattern 1 forecast data (average or most frequent)
x2: Left lane traffic volume in pattern 2 forecast data (average or mode)
y2: Right lane traffic volume in pattern 2 forecast data (average or mode)
a1: Weight of the left lane of pattern 1, b1: Weight of the right lane of pattern 1 (In pattern 1, we want to focus only on the right lane, so increase the value of a1 and decrease the value of b1).
a2: Weight of the left lane of pattern 2, b2: Weight of the right lane of pattern 2 (in pattern 2, we want to pay attention to both lanes, so a2 = b2 is sufficient).

Then, when the traffic jam forecast information management unit 105 inputs the traffic volumes X and Y of each lane, the traffic jam prediction information management unit 105
1 / f1 (| x1-X |, | y1-Y |) and 1 / f2 (| x2-X |, | y2-Y |) are calculated. The one with the larger value is adopted and compared with the threshold value tn. If it is larger than tun, congestion prediction information is generated.

The traffic jam prediction information management unit 105 stores and manages the traffic jam prediction information in the storage unit 104, and transmits the traffic jam prediction information to the navigation device 300 when the traffic jam prediction request is requested from the navigation device 300.

If a requirement condition at the time of a traffic jam prediction request from the navigation device 300, for example, a weather or the like is specified, the traffic information processing unit 103 refers to the traffic information 704 including this parameter. As the requirements increase, it becomes possible to predict traffic congestion with high accuracy for the requirements.

Further, in the above processing example, the traffic jam prediction information is calculated after waiting for the traffic jam prediction request from the navigation device 300, but the present invention is not limited to this. For example, the traffic information processing unit 103 may be configured to generate congestion prediction information for each position and time in advance based on the traffic information 704. In this case, if there is a traffic jam prediction request from the navigation device 300, the traffic jam prediction information of the corresponding position and time can be immediately transmitted to the navigation device 300.

(About traffic jam prediction request and transmission of traffic jam prediction information)
FIG. 9 is a flowchart showing the processing contents from the traffic jam prediction request to the transmission of the traffic jam prediction information. First, the navigation device 300 transmits a congestion prediction request for a designated position or a designated range to the server 400 (step S901).

The traffic jam prediction information management unit 105 of the server 400 receives this traffic jam prediction request (step S902), and transmits (delivers) the traffic jam prediction of the corresponding designated position and designated range to the navigation device 300 (step S903).

The navigation device 300 receives the traffic jam prediction information transmitted from the server 400 (step S904), and displays the traffic jam prediction information on the display unit (step S905).

The traffic jam prediction information management unit 105 of the server 400 may be configured to monitor the presence or absence of a change in the traffic jam prediction based on the possibility that the traffic jam state may change with the passage of time after step S903. (Step S906). If the congestion prediction change does not occur (step S906: No), the process ends, but if the congestion prediction change occurs (step S906: Yes), the congestion prediction information is recalculated and the congestion prediction information is calculated again in step S903. Send.

(About the occurrence of traffic jams)
FIG. 10 is an explanatory diagram showing a state in which natural traffic congestion occurs. As shown in FIG. 10A, as the number of traveling moving objects increases, the distance between the moving objects becomes narrower. Next, as shown in FIG. 10B, when the speed of some of the moving bodies 1001 slows down, the following moving bodies 1011 and 1012 step on the brakes one after another to reduce the speed so as not to collide. As a result, as shown in FIG. 10 (c), a completely stopped moving body 1021 appears, and subsequent moving bodies 1031, 1032, 1033, 1034 also stop one after another, causing natural congestion.

(About traffic jam forecast)
In the congestion prediction of the embodiment, the point that the distance between the moving bodies becomes narrower as the number of moving bodies increases is referred to the photographed image of the navigation device 300 and the past database (traffic information 704). It is a prediction. By referring to the past traffic information 704 corresponding to the required condition, the server 400 determines which of the states (a) to (c) corresponds to in the example of the natural traffic jam shown in FIG. Corresponding to, the traffic jam prediction information corresponding to the requested date and time included in the traffic jam prediction request is transmitted to the navigation device 300.

In addition, how the congestion changes after the requested time (for example, when the request corresponds to the time shown in FIG. 9A, for example, 10 minutes later, the natural congestion shown in FIG. 9C occurs. Can also be transmitted to the navigation device 300 as congestion prediction information.

(About application of traffic jam prediction information to route search)
When the above-mentioned congestion prediction information indicates that there is a high possibility of congestion occurrence (congestion occurrence value is high), the navigation device 300 assigns congestion prediction information to the road on the corresponding congestion occurrence candidate route and routes. I can guide you.

The navigation device 300 of the embodiment performs a route search in consideration of the above-mentioned congestion prediction information. As a result, since the average travel time of the road (link) where congestion is expected is long, a route avoiding this congestion point is searched for.

However, in the route search using the traffic jam prediction information in this way, the presence or absence of the actual traffic jam at the corresponding place becomes clear at the predicted time. Therefore, it is conceivable that the predicted traffic jam does not occur (traffic jam does not occur) at this predicted time. In such a case, when the predicted time is reached, the navigation device 300 re-searches the route and does not use the traffic jam prediction information used last time.

The route search is not limited to the navigation device 300, but may be performed by the server 400. The server 400 may acquire position information such as the current position and destination of the navigation device 300, and search for a route including other search conditions such as time.

In the above embodiment, the number of lanes, the number of vehicles, the distance between vehicles, etc. are image-processed from the captured image to detect the traffic volume, but it is also possible to acquire other parameters by image processing. For example, if there is a guide sign under construction on the lane of the road in the photographed image, there is a possibility that sudden congestion may occur due to the construction. In addition, vehicle types (buses, police cars), pedestrian crossings, etc. that can cause traffic congestion can also be used as parameters.

Therefore, the accuracy of the traffic information 704 can be improved by statistically processing the common points and differences between the captured images themselves and the images before and after the occurrence of the traffic jam.

Further, in the above embodiment, the traffic jam prediction device 100 such as a server performs the traffic jam prediction, but the traffic jam prediction device 110 such as one car navigation device or a personal computer predicts the traffic jam. Can also be configured to transmit to another terminal device 110 that has requested.

The method related to traffic congestion prediction described in the present embodiment can be realized by executing a program prepared in advance on a computer such as a personal computer or a workstation. This program is recorded on a computer-readable recording medium such as a hard disk, flexible disk, CD-ROM, MO, or DVD, and is executed by being read from the recording medium by the computer. Further, this program may be a transmission medium that can be distributed via a network such as the Internet.

100 Congestion prediction device 101 Image processing unit 102 Information coupling unit 103 Traffic information processing unit 104 Storage unit 105 Congestion prediction information management unit 110 Terminal device 111 Imaging unit 112 Positioning unit 113 Processing unit

Claims (1)

An acquisition unit that acquires an image of a moving object on the road and position information indicating the position where the image was taken, and an acquisition unit.
An image processing unit that extracts traffic information that is information about a moving body existing in the image, and an image processing unit.
Using the traffic information extracted by the image processing unit, the traffic information processing unit that calculates information on the traffic congestion prediction on the road corresponding to the photographed position, and the traffic information processing unit.
Congestion prediction system characterized by having.

Images