JP2015060443A - Traffic volume grasping device and method - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a system in which, using a vehicle imaging device, a plurality of images of one vehicle are taken, and a speed is acquired from differential information of the images, while determining an advancing direction.SOLUTION: A traffic volume grasping device includes an imaging part for imaging a monitor region, control means for controlling the imaging part, and a frame memory for storing an image from the imaging part, for processing the image, so that type, passage, speed, and an advancing direction of a vehicle are detected. It includes means for reading a number plate of a vehicle with one or a plurality of images as a target image, means for particulating a type from the number plate, means for acquiring moving speed from a plurality of images having a time lag, and means for acquiring advancing direction of the vehicle from the plurality of images having a time lag.

Description

  The present invention relates to a traffic flow grasping device and method for grasping a traffic volume and a passage route on a road based on information from monitoring devices arranged on an intersection and on a road.

  As a background art in this technical field, there is JP-A-2000-322684 (Patent Document 1). This Patent Document 1 states that “a roadside apparatus transmits an inquiry signal to a communication area via an antenna, and a response signal to the inquiry signal is continuously returned from the vehicle-mounted device of the vehicle that has passed through this communication area. ”Measures the number of responses of the response signal that is continuously sent back through the vehicle and measures the passing speed of the vehicle.”

JP 2000-322684 A

  In Patent Document 1, since the vehicle traveling speed is obtained and the traveling direction is not grasped at an intersection or the like, comprehensive traffic flows (traffic volume and passage route) at a plurality of intersections in a predetermined area are described. I can't figure it out.

  Therefore, the present invention provides a mechanism for taking a plurality of images for one vehicle using a vehicle imaging device, obtaining the speed from the image difference, and determining the traveling direction.

  In order to solve the above-described problem, the image information processing unit calculates the vehicle movement amount from the difference between a plurality of images obtained for one vehicle, obtains the speed, and combines the change of the difference. The traveling direction is determined. In addition, comprehensive traffic flow is grasped based on traffic flow information at a plurality of intersections in a predetermined area.

  According to the present invention, it is possible to grasp not only the speed of the vehicle traveling on the road but also the passing vehicle amount and the traveling direction of the vehicle on the intersection, so that it is possible to grasp the total traffic flow.

It is a figure which shows the system configuration | structure in a present Example. It is a figure which shows the arrangement | positioning condition of the monitoring apparatus in each intersection. It is a figure which shows the example of the movement condition of the vehicle in an intersection. It is a figure which shows the advancing direction determination table based on the code | symbol of the change of the coordinate value of a vehicle. It is a figure which shows the example of the movement condition management table | surface of each vehicle for every intersection. It is a figure which shows the positional relationship between the shape and location of each intersection, and intersections. It is a figure which shows the condition where a specific vehicle passes the intersection sequentially. It is a figure which shows the example of the table | surface which manages the movement condition of each vehicle. It is a figure which shows the relationship between the direction in the screen display of each intersection, and an actual direction. It is a figure which shows management of the information of the intersection in a five-way road. It is a figure which shows the example in case a vehicle merges at the intersection. It is a figure which shows the confluence | merging pattern of the vehicle in the nearby intersection.

  In the present embodiment, an example will be described in which imaging devices are arranged in four directions on an intersection, and the vehicle traveling speed is calculated, vehicle license plate information is read, and the progress of the vehicle such as going straight, turning left, turning right, etc. is grasped.

(System configuration)
FIG. 1 is a diagram illustrating a configuration of a system for grasping road traffic volume and a passage route according to the present embodiment.

  The imaging unit 100 is equipped with an LED illumination device capable of photographing even at night, and is configured from non-storage type CMOS cameras 100-1 to 100-n capable of high-speed continuous shooting and aperture adjustment, and provides vehicle shooting. Four cameras are provided at one intersection.

  The image information processing apparatus 200 is connected to the imaging unit 100 and includes an image information reception unit 201, an image information processing unit 202, and a passing vehicle information transmission unit 203. The image information processing apparatus 200 processes images at several intersections at the same time, and a plurality of images are provided in a predetermined area.

  The imaging unit 100 and the image information processing apparatus 200 are paired, and a plurality of pairs can be installed corresponding to each intersection. The image information processing apparatus 200 generates passing vehicle information at the intersection at predetermined time intervals based on image information from the four cameras.

  The passing vehicle information is information on the vehicle number, the passing speed of the vehicle, the traveling direction of the vehicle, and the color / model of the vehicle.

  The central processing unit 300 is connected to the image information processing device 200 via a network, and includes a communication processing unit 301, a passing vehicle information storage device 302, a passing vehicle information database 303, and a passing vehicle information operating device 311. Based on the passing vehicle information at each intersection, comprehensive traffic flow (road traffic volume and passing route of each vehicle) is grasped.

  The communication processing device 301 can receive passing vehicle information from the plurality of image information processing devices 200 by communication.

  An example in which an image acquired by the imaging unit 100 of the present embodiment is converted into passing vehicle information by the image information processing apparatus 200 and stored in the passing vehicle information DB (database) 303 is shown below.

  FIG. 2 shows an installation example when a pair of the imaging unit 100 and the image information processing apparatus 200 is installed at an intersection.

  First, a continuous image captured by the camera 100-1 with a fixed shutter speed is received by the image information receiving unit 201 and transferred to the image information processing unit 202. The image information processing unit 202 detects a change in images continuously taken with respect to one camera, and if there is a change greater than a set threshold, it determines that the vehicle has entered, performs image processing, and moves the image and the shutter. Calculate the vehicle speed from the speed relationship.

  Further, the position of the license plate of the vehicle is determined from the photographed image, and the description content of the license plate is read from the character recognition of the image.

  Also, the direction of travel of the vehicle, that is, when the vehicle image moves forward, due to a change in each image, goes straight, turns right when the vehicle image moves to the left, and turns left when the vehicle image moves to the right. And determine the direction of travel.

  Further, the color of the vehicle is extracted based on the color of the photographed image, and the contour is extracted from the photographed image by image processing to determine the vehicle type such as a truck, van, or passenger car.

  The passing vehicle information obtained by the above processing is passed to the passing vehicle information transmission unit 203.

  The passing vehicle information transmission unit 203 passes the passing vehicle information obtained by the image information processing unit 202 to the central processing unit 300.

  The central processing unit 300 is a communication processing unit 301 that receives information on traffic flow (road traffic volume and passing route of each vehicle) at each intersection from the image information processing devices 200 provided at a plurality of locations in a predetermined area. One vehicle based on the information of the passing vehicle information storage device 302 that stores the flow information in the passing vehicle information DB 303 and the traffic flow (the road traffic volume and the passing route of each vehicle) stored in the passing vehicle information DB 303 The travel direction is determined, and the overall traffic flow is grasped and displayed based on the traffic flow information at a plurality of intersections in a predetermined area.

(1) Grasp of traffic flow at one intersection (1-1) Pattern of vehicle movement status at an intersection FIG. 3 shows a pattern of vehicle movement status at one intersection. If the upper direction of the intersection display screen in FIG. 3 is “north”, for example, the lower direction is “south”, the right direction is “east”, and the left direction is “west”. (1) in FIG. 3 shows a situation where the vehicle approaches the intersection in the upward direction (north) and goes straight (north), turns right (east), or turns left (west). FIG. 3 (2) shows a situation where the vehicle approaches the intersection in the downward direction (south) and goes straight (south), right turn (west), or left turn (east). (3) in FIG. 3 shows a situation where the vehicle is approaching the intersection toward the right (east) and goes straight (east), turns right (south), or turns left (north). (4) in FIG. 3 shows a situation in which the vehicle approaches the intersection toward the left (west) and goes straight (west), turns right (north), or turns left (south).

  As will be described later, the correspondence between the vertical and horizontal directions on the display screen and the actual direction of east, west, south, and north is different for each intersection, and when the information on the vehicle movement status is linked between adjacent intersections, It is necessary to perform an operation of rotating the other display screen by 90 degrees clockwise, for example, so as to match the one display screen so that the actual east, west, north, and south directions coincide.

(1-2) Determination of Traveling Direction of Vehicle at Intersection In the passing vehicle information DB 303 shown in FIG. 1, each vehicle for each intersection is shown in the movement status management table for each vehicle shown in FIG. In order to hold the vehicle number, vehicle type, shooting date and time (t _k ), vehicle coordinate values (X _i , Y _i ), and to determine the traveling direction and speed of each vehicle, the previous time t _k−1 , the current The above information is held for three times of time t _k and next time t _{k + 1} . The intervals of these three times are not necessarily the same as the time intervals of the photographing times, but may be any time intervals necessary for obtaining the traveling direction and speed of each vehicle.

The traveling direction and speed of the vehicle are obtained by comparing the coordinate values at the two times. However, the orientation of the vehicle traveling in the previous time t _k-1 and the current time t _k is the same (straight), by comparing the coordinate values at the previous time t _k-1 and the current time t _k, straight of the vehicle Decide the direction. That is, when δX _i = X _i −X _i−1 and δY _i = Y _i −Y _i−1 , in the case of straight traveling, δX _i = 0 (up / down) or δY _i = 0 (right / Left)),
If δX _i = 0 and Y _i > Y _i−1, then up; if Y _i <Y _i−1, then down;
If δY _i = 0 and X _i > X _i−1 , the direction is right, and if X _i <X _i−1 , the direction is left.

Then, by comparing the coordinates of the vehicle at the current time t _k and the next time t _{k + 1} Tokyo, determine the change in the traveling direction. That is, when ΔX _i = X _{i + 1} −X _i and ΔY _i = Y _{i + 1} −Y _i , based on the signs of ΔX _i and ΔY _i , as shown in the traveling direction determination table shown in FIG. Ask for. The upper part of FIG. 4 shows the case of up or down, the lower part shows the case of facing right or left, and either δX _i = 0 or δY _i = 0, the previous time t _k−1 and the current time It used depending on the comparison result of the coordinate values in the time t _k.

  In the traveling direction determination table shown in FIG. 4, the traveling direction of the vehicle can be determined by a combination of symbols, so that the processing is easy.

In the movement status management table of each vehicle shown in FIG. 5, the vehicle number (number read from the license plate) and the vehicle type (white private car, black truck, etc.), the shooting date / time (t _k ) and the vehicle The coordinate values (X _i , Y _i ) are also shown, but depending on the shooting situation of the shooting unit 100 shown in FIG. 1, a part of the vehicle number cannot be read due to the shadow of another vehicle, or shooting is performed. In some cases, the vehicle type cannot be discriminated from the image. Even in such a case, only the information that is found is stored in the management table of FIG. For example, information such as “012-34 **” as the vehicle number and “** color private car” as the vehicle type is stored. Such partially unknown information can be estimated on the basis of a partial match in comparison with other shooting dates and times or information at intersections near the intersection. Based on such an estimation result, partially unknown information can be supplemented.

The time t _i-1 corresponding to the time t _i of the _vehicle, or the time t _{i + 1} if the coordinate data is not present in the moving condition management table of intersection C _n is the intersection C _n in the vicinity of the intersection C _{n -1} or information on the vehicle is extracted from the movement status management table at the intersection Cn _{+ 1} , and the direction of vehicle travel is determined based on the travel direction determination table.

(1-3) speed V of the speed the vehicles of each vehicle, the distance from the coordinate value at the current time _{t k} and the next time _{t k + 1} Metropolitan _{L = √ ((X i +} 1 -X i) 2 + (Y i + 1 -Y _i ) ² ) is obtained, and it is obtained by V = L / (t _{k + 1} −t _k ). Alternatively, assuming that the vehicle is either turning at a right angle or going straight at the intersection, the distance L is calculated from the coordinate values at the current time t _k and the next time t _{k + 1} by L = | X _{i + 1} − X _i | + | Y _{i + 1} −Y _i | (= | ΔX _i | + | ΔY _i |), and the speed V when the vehicle turns at the intersection is expressed as V = L / (t _{k + 1} −t _k ) It can also be obtained as follows.

  In the first embodiment, the case of obtaining the traveling direction and speed of each vehicle at one intersection has been described. However, in the second embodiment, traffic is not limited to one intersection, and traffic is referred to at an intersection near the intersection. Intersection processing for understanding the flow will be described. That is, in the present embodiment, a process for estimating the influence of traffic flow at other intersections adjacent to the north side, south side, west side, and east side of the intersection on the traffic flow on the road to the intersection will be described.

  Since there are T-junctions as well as ordinary crossroads at intersections, the shape of the intersection that distinguishes them, the position of the intersection represented by the center coordinates of the intersection, and the number of other intersections adjacent to the intersection are listed for each intersection. FIG. 6 shows a table related to the shape and location of each intersection and the positional relationship between the intersections.

In the shape of the intersection, “10” at the intersection C ₁ indicates a crossroad, and “T (S)” at the intersection C ₂ indicates a T-shaped road with a dead end on the south side. In the following, it is assumed that the intersection is a crossroad unless otherwise specified.

FIG. 7 shows the positional relationship between the intersections shown in FIG. The north side of the intersection C ₁ intersection adjacent without there is an intersection C ₄ on the south side, there is an intersection C ₆ to west, shows that there is an intersection C ₂ on the east side, the intersection C ₂ is the south dead end in the T-junction, it shows that the intersection C ₁ is adjacent to the west.

  As for the shape of the intersection, even if the two roads are not orthogonal to each other, if the shape is close to a cross, the shape is set to “ten”, and the T-junction is similarly T If the shape is close to a letter, the shape is “T (S)” or the like. Rather than a T-junction, in the case of a three-way road that is an intersection where there is no straight-ahead part, if the direction where the angle between two roads is 90 degrees or more is on the south side, “Y (S)” is used for convenience. Are assigned directions such as up / down, right / left, and straight / right / left turn, and processing is performed in the same manner as the above T-junction.

  The five-way road shown in FIG. 10 is processed by the following method.

  Method 1: The traffic volume is small, for example, the road (4) is ignored, and the above processing is performed as an intersection of the cross roads including the roads (1), (2), (3) and (5).

  Method 2: For example, roads (3) and (4) with a small amount of traffic are combined into a virtual road (3) ′ as shown by a broken line to form a cross road, and virtual road (3) ′ is formed. The route direction shown in Fig. 4 is determined separately for A and B corresponding to the original roads (3) and (4), and if the roads are divided into the original roads (3) and (4) after processing The processing procedure for a normal crossroad in Example 1 can be used as it is. However, in the traveling direction determination table shown in FIG. 4, straight / right / left turn is divided into straight A / right turn A / left turn A and straight B / right turn B / left turn B. That is, in the method 2, the virtual road (3) ′ as shown by the broken line is regarded as a one-way, two-lane road, and the processing procedure for the normal cross road in the first embodiment is used.

In the display screen of the present embodiment, the positional relationship between the intersections is distinguished by the direction of east, west, south, and north, and the traveling direction of each vehicle is distinguished by the direction of up / down / right / left. Therefore, when the traveling direction of each vehicle is associated with the east, west, south, and north indicating the positional relationship between the intersections, the correspondence is as shown in FIG. 9 showing the relationship between the direction in the screen display of each intersection and the actual direction. It depends on the intersection. Therefore, when associating information about traffic flow at two adjacent intersections, the positional relationship between the intersections must match the directions of east, west, south, and north based on FIG. For example, when the vehicle advances in the downstream direction an intersection C ₁ on the screen, because the C ₁ in the downlink direction is S (South), the intersection C ₄ adjacent to the intersection C _1, on the display screen, the same S It is necessary to associate the leftward direction corresponding to the (south) direction. At that time, which intersection is adjacent to the intersection can be determined from the positional relationship between the intersections shown in FIG. By associating the direction of the road with the traveling direction of the vehicle using the above information and identifying the traveling direction of the road passing through two adjacent intersections, the moving path of the specific vehicle can be identified.

  FIG. 8 is traffic flow information showing the movement status of each vehicle moving while passing through an adjacent intersection based on the relationship between the direction in the screen display of each intersection and the actual direction shown in FIG. is there.

In FIG. 8, for example, a vehicle with the vehicle number “012-3456” turns right at the intersection C ₁ at the speed of 45 km / h from the up direction (north) to the right (east), and about 5 minutes later, the intersection C ₂ becomes 35 km / h. It shows that you are turning right (east) and left (north) at a speed of h.

  As described above, by using the positional relationship between the intersections in FIG. 6 and the relationship between the direction in the screen display of each intersection in FIG. 9 and the actual azimuth, as shown by the dashed arrows in FIG. The moving route of a specific vehicle can be displayed on the screen.

  In the first and second embodiments, information on the traffic flow of a specific (focused) vehicle has been described. In the third embodiment, a description will be given of a traffic flow, particularly a traffic volume estimation process, when a plurality of vehicles traveling from various directions join at an intersection and travel on a straight road.

  FIG. 11 shows a situation where, at the intersection B, the right-turning left turn, the straight going up, and the left-turning right turn join each other and proceed on a straight road toward the intersection A. FIG. 12 shows a merging pattern of vehicles at a nearby intersection as shown in FIG. FIG. 12 shows a correspondence between a straight direction on a straight road and a set of traveling directions of a vehicle at a nearby intersection related to merging with the straight road.

  For example, the traveling direction of the vehicle related to the merging to the straight road in the upward direction toward the intersection A is ascending straight, turning to the right and turning to the left, and turning to the left. The number of vehicles traveling on the straight road in the upward direction.

  Further, at the intersection A, as shown in FIG. 3, in particular, as shown in (1), the number of vehicles of straight / left / right turn is obtained, and by repeating such processing for each intersection, each vehicle in a predetermined area is obtained. It is possible to grasp the traffic flow (road traffic volume and route of each vehicle) at the intersection.

  In addition, when there is no image capturing unit 100 and image information processing apparatus 200 shown in FIG. 1 at the intersection, and there is an image capturing unit 100 and image information processing apparatus 200 shown in FIG. 3, the information on the movement status of the vehicle at the adjacent intersection as shown in FIG. 3 is acquired, and based on the information, the traffic flow at the intersection where there is no monitoring facility such as the intersection can be estimated.

For example, the information on the positional relationship between the intersections shown in FIG. 6 also holds information on intersections that do not have monitoring equipment such as the intersection, and based on the information on the positional relationship between the intersections in FIG. The traffic flow at the intersection is estimated using the average value of the traffic flow (road traffic volume) corresponding to each traveling direction at a plurality of intersections adjacent to the intersection. For example, the average value of the straight traffic flow in the upward direction at the intersection adjacent to the north side of the intersection and the straight traffic flow in the upward direction at the intersection adjacent to the south side of the intersection is the average value of the straight traffic flow in the upward direction at the intersection. Estimated as traffic flow. (In this estimation method, influences from intersections adjacent to the west and east sides of the intersection are ignored.)
By the processing shown in the first to third embodiments, not only the speed of the vehicle traveling on the road but also the amount of passing vehicles and the traveling direction of the vehicle on the intersection can be grasped, so that the comprehensive traffic flow is grasped. Is possible.

DESCRIPTION OF SYMBOLS 100: Imaging part, 200: Image information processing apparatus, 201: Image information receiving part, 202: Image information processing part, 203: Passing vehicle information transmission part, 300: Central processing unit, 301: Communication processing unit, 302: Passing vehicle Information storage device, 303: passing vehicle information database, 311: passing vehicle information operating device

Claims (12)

An image information processing apparatus that is connected to an imaging unit that images a road including an intersection as a monitoring area at predetermined time intervals and generates passing vehicle information including a vehicle number and a vehicle type of the vehicle from an image from the imaging unit via a network. Connected traffic volume monitoring device
Means for reading the vehicle number from the license plate of the vehicle displayed on the image with one or a plurality of images as a target image;
Means for identifying the vehicle type including the color of the vehicle body from the shape of the vehicle displayed in the image;
Means for determining the moving speed from a plurality of images with time differences;
Means for determining the traveling direction of the vehicle at the intersection from a plurality of images having a time difference;
A traffic flow grasping device characterized by comprising: Means for determining the traveling direction of the vehicle at the intersection,
Means for determining whether the straight direction of the vehicle is an upward direction, a downward direction, a right direction, or a left direction on the screen by comparing the coordinate values at the previous time t _k-1 and the current time t _k ;
Based on the comparison of the coordinate values of the vehicle at the current time t _k and the next time t _{k + 1} and the direction of straight travel of the vehicle, the change in the travel direction of the vehicle is straight, right turn, or left turn on the screen. Means for deciding either
The traffic flow grasping device according to claim 1, comprising: The traffic flow grasping device is
If any part of the vehicle number or vehicle type of the vehicle cannot be determined from the captured image, partial coincidence in comparison with other captured date / time or information from the captured image at the intersection near the intersection The traffic flow grasping device according to claim 1, wherein a portion that cannot be determined is estimated based on the information, and partially unknown information is supplemented based on the estimation result. The traffic flow grasping device is a device that acquires and processes passing vehicle information related to a plurality of intersections provided in a predetermined area from each of a plurality of image information processing devices, through the network,
First information for managing the positional relationship with other intersections adjacent to the intersection based on the direction with respect to the intersection, and the upward direction, the downward direction, the right direction, or the left direction on the image including each intersection Holds the second information that manages the correspondence with the actual direction,
The movement of a specific vehicle by specifying the traveling direction of the road passing through two adjacent intersections by associating the direction of the road with the traveling direction of the vehicle using the first information and the second information. The traffic flow grasping device according to claim 2, further comprising means for specifying a route.   Based on the first information and the second information, in a straight road from the first intersection to the second intersection, go straight from the first intersection to the straight road, turn right, and turn left to merge 5. The traffic flow grasping device according to claim 4, further comprising means for determining the traffic volume of the straight road based on the number of vehicles.   Based on the first information and the second information, when the image information processing apparatus to which the imaging unit is connected is not installed at the intersection, information on the traffic volume at the intersection adjacent to the intersection 5. The traffic flow grasping device according to claim 4, further comprising means for estimating a traffic volume at the intersection based on the basis. An image information processing apparatus that is connected to an imaging unit that images a road including an intersection as a monitoring area at predetermined time intervals and generates passing vehicle information including a vehicle number and a vehicle type of the vehicle from an image from the imaging unit via a network. The method of grasping the traffic flow in the connected processing equipment
Read the vehicle number from the license plate of the vehicle displayed in the image with one or more images as the target image,
Identify the vehicle type including the color of the vehicle body from the shape of the vehicle displayed in the image,
Find the moving speed from multiple images with time differences,
Finding the traveling direction of the vehicle at the intersection from a plurality of images with a time difference,
Traffic flow grasping method characterized by this. When determining the traveling direction of the vehicle at the intersection,
Comparison of the coordinate values in the previous time t _k-1 and the current time t _k, decided straight direction of the vehicle, on the screen, uplink, downlink, rightward, or any of left,
Based on the comparison of the coordinate values of the vehicle at the current time t _k and the next time t _{k + 1} and the direction of straight travel of the vehicle, the change in the travel direction of the vehicle is straight, right turn, or left turn on the screen. Decide either
The traffic flow grasping method according to claim 7. In the traffic flow grasping method,
If any part of the vehicle number or vehicle type of the vehicle cannot be determined from the captured image, partial coincidence in comparison with other captured date / time or information from the captured image at the intersection near the intersection The traffic flow grasping method according to claim 7, wherein a part that cannot be determined is estimated based on the information, and partially unknown information is supplemented based on the estimation result. The processing device is a device that acquires and processes passing vehicle information related to a plurality of intersections provided in a predetermined area from each of a plurality of image information processing devices, and
First information for managing the positional relationship with other intersections adjacent to the intersection based on the direction with respect to the intersection, and the upward direction, the downward direction, the right direction, or the left direction on the image including each intersection Holds the second information that manages the correspondence with the actual direction,
The movement of a specific vehicle by specifying the traveling direction of the road passing through two adjacent intersections by associating the direction of the road with the traveling direction of the vehicle using the first information and the second information. 9. The traffic flow grasping method according to claim 8, further comprising means for specifying a route.   Based on the first information and the second information, in a straight road from the first intersection to the second intersection, go straight from the first intersection to the straight road, turn right, and turn left to merge The traffic flow grasping method according to claim 10, wherein the traffic volume of the straight road is obtained based on the number of vehicles.   Based on the first information and the second information, when the image information processing apparatus to which the imaging unit is connected is not installed at the intersection, information on the traffic volume at the intersection adjacent to the intersection 11. The traffic flow grasping method according to claim 10, wherein the traffic volume at the intersection is estimated based on the traffic volume.

Images