JP2001101571A - Method, system, and device for decoding traffic congestion and recording medium - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a method, system, and device for deciding traffic congestion by which an abrupt change of a traffic flow such as the occurrence of a traffic congestion, etc. in a wide detecting range can be recognized quickly and precisely and to provide a recording medium. SOLUTION: The processing section 12 of a space sensor 1 measures space parameters including the space occupying ratio and average spatial speed of vehicles running in a measuring range contained in the picture taken with a CCTV camera 11 overlooking a road and transmits the measured results to a central transportation controller 2 through a line and the congestion deciding means 22 of the controller 2 decides the state of the traffic flow on the road by deciding that the region in which the correlation between the measured space occupying ratio and average space speed exists out of a plurality of regions in accordance with a plurality of states of the traffic flow which are decided by dividing the correlation distributing region between the space occupying ratio and average space speed by the combinations of the threshold values of the space occupying ratio and average space speed by referring to a threshold combination table 21.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention measures spatial parameters such as a space occupancy, a spatial average speed, and a traffic density of a vehicle existing in a measurement range in an image obtained by overhead photographing a road,
A traffic congestion determination method, a traffic congestion determination system, a traffic congestion determination device, and a computer program for causing a computer to perform traffic congestion determination, based on spatial parameters, for determining the state of traffic flow on a road (normal flow, congestion flow, stagnant flow, etc.) Related to a recording medium on which is recorded.

[0002]

2. Description of the Related Art In a current traffic congestion determination system, traffic congestion is determined using time parameters such as traffic volume and time occupancy collected by ultrasonic sensors arranged at intervals of 500 m above a road. The ultrasonic sensor is installed in an overtaking lane where the stability of the occupancy data is excellent without any change in the road shape due to the on-ramp and the off-ramp, and data is collected.

In the current traffic congestion determination system, an occupancy-traffic characteristic (OV characteristic) is created from the collected data, and the congestion is determined by dividing the vicinity of 35% of the occupancy into a non-congestion area and a congestion area by using a branch point. However, collected data for 5 minutes is used to create OV characteristics. The reason is that 1-minute collected data causes a sharp change following the change in traffic flow, such as compression waves during congestion, and 15-minute collected data delays the response to changes in traffic flow, but 5-minute collection This is because changes in traffic flow are skillfully captured in the data, and stable changes are seen.

[0004]

It is said that the determination of traffic congestion can be better understood by using spatial parameters such as traffic density and spatial average speed. However, since the ultrasonic sensor is a point measurement sensor, it can only measure time parameters such as traffic volume and time occupancy. Further, since the 5-minute collected data is used for measuring the time parameter, the time for accumulating the data causes a time delay until the traffic jam determination. Furthermore, since the accumulated data for a predetermined period such as 5 minutes is averaged, rapid changes in traffic flow such as occurrence of traffic congestion cannot be quickly or precisely captured.

[0005] Further, since the ultrasonic sensor has a narrower detection range than the vehicle body, only discrete data can be obtained at a data collection interval of 500 m. In addition, when the ultrasonic sensor is arranged in the passing lane, the vehicle passes. Since we only collect lane data,
The traffic jam outside the detection range cannot be captured.

The present invention has been made in order to solve such a problem, and measures a space parameter such as a space occupancy, a space average speed, and a traffic density of a vehicle from an image at the moment when a road is overlooked. The correlation between the space occupancy rate and the spatial average speed, or the correlation between the traffic density and the spatial average speed, is included in any of a plurality of areas obtained by dividing the correlation distribution area of these parameters according to each of a plurality of states of the traffic flow. A traffic congestion determination method, a traffic congestion determination system, a traffic congestion determination device, and a computer for determining traffic congestion in a wide detection range by quickly or precisely determining a sudden change in traffic flow by determining a traffic flow state based on the traffic flow. The purpose of the present invention is to provide a recording medium on which a computer program for performing the above is recorded.

[0007]

A traffic congestion determination method according to the present invention is a traffic congestion determination method for judging the state of traffic flow on a road using an image obtained by photographing the road from a bird's-eye view. The space occupancy and the space average speed of the vehicle are measured, and the correlation between the measured space occupancy and the space average speed is used to determine the correlation distribution area between the space occupancy and the space average speed. The method is characterized in that the state of the traffic flow on the road is determined based on which of the plurality of areas corresponding to the plurality of states of the traffic flow, which are classified according to the combination.

A traffic congestion determination system according to the present invention is a traffic congestion determination system for determining the state of traffic flow on a road using an image obtained by photographing a road from a bird's-eye view. And a terminal device having a unit for measuring a spatial parameter including the spatial average speed and a unit for transmitting the measured spatial parameter, a unit for receiving the spatial parameter transmitted from the terminal device, a correlation between the space occupancy and the spatial average speed. The distribution area is divided into a plurality of areas corresponding to a plurality of states of the traffic flow, a means for storing a combination of thresholds of the space occupancy and the space average speed, and the correlation between the measured space occupancy and the space average speed, A central device having means for determining the state of traffic flow on the road according to which of the plurality of areas is included.

A traffic congestion determining apparatus according to the present invention is a traffic congestion determining system for determining the state of traffic flow on a road by using an image obtained by photographing a road from a bird's-eye view. Means for measuring the spatial occupancy and the spatial average velocity, and a threshold for the spatial occupancy and the spatial average velocity, which divides the correlation distribution area of the spatial occupancy and the spatial average velocity into a plurality of areas corresponding to a plurality of states of the traffic flow, respectively. And means for determining the state of traffic flow on the road based on which of the plurality of areas contains the correlation between the measured space occupancy and the space average speed.

[0010] The recording medium of the present invention uses traffic parameters on a road by utilizing a space parameter including a space occupancy and a space average speed of a vehicle existing in a measurement range in the image which is measured from a bird's-eye image of the road. A computer program for causing the computer to determine the flow state is recorded, and on a computer-readable recording medium, program code means for causing the computer to acquire the measured space parameter; and Program code means for dividing the correlation distribution region of the spatial average speed into a plurality of regions corresponding to a plurality of states of the traffic flow, and referring to data of a combination of a threshold of the spatial occupancy and the spatial average speed; The correlation between the occupied space occupancy and the spatial average velocity is divided into a plurality Wherein the computer program comprising program code means for determining the state of the road traffic flow by either included in any of the range are recorded.

Further, according to the present invention, the space occupancy and the space occupancy can be changed by combining different thresholds in a case where a traffic congestion occurs when the traffic flow is switched to a traffic congestion state and in a case where the traffic congestion is switched from a traffic congestion state to a normal state. It is characterized in that the correlation distribution region of the spatial average velocity is divided.

Further, the present invention is characterized in that traffic density is used instead of space occupancy.

In the present invention, the space parameters such as the space occupancy, the space average speed, and the traffic density of the vehicles existing in the measurement range in the image at the moment when the road is overlooked are measured, and the measured space occupancy and space are measured. The correlation of the average speed is calculated based on the combination of the threshold of the space occupancy and the threshold of the space average velocity. The state of the traffic flow on the road is determined based on which of the plurality of areas is classified according to.

[0014] Thus, by taking a bird's-eye view of an image of a relatively long multi-lane road extension, the vehicle detection range is widened and continuous data on the road can be collected. In addition, since spatial parameters can be collected from instantaneous images in principle, data accumulation such as 5 minutes is not required, and temporal averaging is not required, as compared to congestion determination using ultrasonic sensors. It is. Therefore, the state of traffic flow such as traffic jam can be immediately grasped. Also, a rapid change in traffic flow at the time of congestion can be quickly and precisely captured.

Further, according to the present invention, the space occupancy rate and the space distribution for dividing the correlation distribution area are determined when the traffic flow changes from a normal flow to a congested flow and when the traffic flow changes from a congested flow to a normal flow. The congestion flow is determined by using different combinations of the spatial average speed or the threshold value of the traffic density and the threshold value of the spatial average speed.

Therefore, for example, by setting the threshold value at the time of switching from the congested flow to the normal flow to a value lower than the threshold value at the time of switching from the normal flow to the congested flow and giving the threshold value hysteresis, the space occupancy and the spatial average When the correlation between the speeds or the correlation between the traffic density and the spatial average speed is near the change point of the traffic flow, it is possible to prevent the traffic jam determination result from being repeatedly inverted in a short time.

In the present invention, traffic density is used instead of space occupancy. Since the traffic density has a first-order linear relationship with the space occupancy, and the correlation distribution with the spatial average speed is similar, it is possible to determine the congestion using either of the spatial parameters.

[0018]

FIG. 1 is a block diagram showing the configuration of a traffic control system to which a traffic jam judging system according to the present invention is applied. A CCTV (Closed-Circuit TV) camera 11 is a device (hereinafter, referred to as a space sensor) 1 that measures spatial parameters of vehicles existing on a road, such as spatial occupancy, spatial average speed, and traffic density, using image processing. It is an imaging unit. The CCTV camera 11 is mounted on an upper portion of a pole such as a T-type illumination lamp erected on a median strip of a road. At this time, for example, CCTV cameras 11 are installed at 120 m intervals,
Measures the angle of view of the TV camera 11 in the traveling direction of the lane, 120 m
In addition, it is possible to collect continuous data on the road by setting the measurement range of all lanes (four lanes up / down) in the direction of lane crossing. Lane data can be collected.

The processing section 12 of the space sensor 1 is realized by a microcomputer having an image processing function. The processing unit 12 captures images taken by the plurality of CCTV cameras 11 from the bird's-eye view via a cable or the like, and performs image processing.
The space occupancy of vehicles existing in the measurement range of the CCTV camera 11,
Calculate spatial parameters such as spatial average speed and traffic density.

FIG. 2 is a diagram showing an example of the measurement range of one CCTV camera 11. In addition, as for the vehicle existing on the measurement boundary, the case where the rear part (in the case of rear measurement) or the vehicle head (in the case of front measurement) exists within the measurement range is set as the measurement target.

The traffic control center for managing and controlling the traffic condition of the road is provided with a traffic control central device for determining the condition of the traffic flow (normal flow, congested flow, stagnant flow, etc.) from the spatial parameters measured by the spatial sensor 1. The traffic control central device 2 and the processing unit 12 of the space sensor 1 are connected by a line.

The traffic control central unit 2 assigns two spatial parameters, for example, a correlation distribution area between the average spatial velocity and the spatial occupancy, to each of a plurality of states of the traffic flow (normal flow, congested flow, stagnant flow, etc.). A threshold combination table 21 is provided which stores a combination of threshold values of two spatial parameters that are divided into a plurality of corresponding areas, for example, a combination of threshold values of a space occupancy and a space average velocity.

These thresholds are set on a daily (or monthly) basis.
A time series graph of two spatial parameters, for example, a spatial average speed and a space occupancy rate, is created in units of years), and a time zone in which congestion occurs is obtained from a correlation graph of the two parameters created for each time zone from the time series graph. Is set to distinguish between the correlation distribution area of the normal flow and the correlation distribution area of the normal flow time zone.

FIG. 3 is an example of a correlation graph between the spatial average speed and the space occupancy. FIG. 4 is a conceptual diagram of the threshold combination table 21, and FIG. 5 is a conceptual diagram in which a correlation distribution area of the space occupancy and the space average speed is divided into a plurality of areas according to the state of traffic flow. FIG. 5 shows that the thresholds are a = 35%, b = 20%, α = 20
An example in the case of setting km / h and β = 30 km / h is shown.

The traffic congestion determining means 22 of the traffic control central device 2
With reference to the threshold value combination table 21, the state of the traffic flow on the road is determined based on which distribution area contains the correlation between the two spatial parameters transmitted from the spatial sensor 1 via the line.

Next, the principle of the traffic congestion determination method will be described. Traffic characteristics such as traffic congestion can be expressed by the following equation. Traffic flow = traffic density × average spatial speed In other words, traffic conditions such as traffic congestion can be calculated by using two factors: “congestion degree of vehicles (traffic density)” and “movement speed of vehicle group (average spatial speed)”.
It is said that the traffic phenomena can be accurately grasped if the determination is made using the correlation between the two spatial parameters.

Further, the traffic density has a linear relationship with the space occupancy as shown in the following equation. Traffic density = space occupancy / average vehicle length Accordingly, the correlation of spatial average speed-space occupancy and the correlation of spatial average speed-traffic density show the same distribution (FIG. 6 (a)).
(b)). That is, it can be said that the congestion determination can be made using any of the correlation between the space occupancy and the spatial average speed and the correlation between the traffic density and the spatial average speed.

Next, an example of a procedure for measuring the spatial parameters in the processing unit 12 of the spatial sensor 1 will be described with reference to the flowchart of FIG. A vehicle as a detection target is extracted from an image captured from the CCTV camera 11. At this time, the presence candidate of the vehicle is extracted by combining the background difference method and the spatial differentiation method (step S7-1). For each vehicle presence candidate, the amount of movement is calculated by the time difference method, and the speed is calculated.
(Step S7-2).

A space parameter is calculated from the vehicle extracted in steps S7-1 and S7-2 and its speed (step S7-3). That is, the traffic density is calculated by measuring the number of vehicles in the measurement range, the space occupancy is calculated by dividing the sum of the vehicle lengths measured for the vehicles in the measurement range by the measurement range length (direction of travel), and the spatial average speed is calculated as follows: Calculate the average value of the speed of the vehicle existing within the measurement range.

At this time, since the image is picked up in a wide field of view, the depression angle of the CCTV camera 11 becomes shallow, so that the coordinate conversion process for the camera view angle and the separation process of the adjacent vehicle are performed in order to grasp the behavior of each vehicle. It is desirable to perform processing such as correction of the vehicle length and the vehicle movement amount.

Next, the congestion determination means of the traffic control central unit 2
An example of the congestion determination procedure in 22 will be described with reference to the flowchart of FIG. The space sensor 1 is operated in the above-described manner.
The combination of the space average speed and the space occupancy measured by the processing unit 12 is taken in at regular intervals (step S8-1). The combination of the input space average speed and space occupancy is applied to the threshold combination table 21, and the state of the traffic flow at that time is determined depending on which of the three levels is included (step S8).
-2).

By the way, at the change point of the traffic flow, the result of the congestion determination may be repeatedly inverted in a short time. To prevent this, each threshold value is provided with hysteresis. For example, the threshold at the time of switching from level 0 (normal flow) to level 1 (congested flow) is set to level 1 (congested flow).
The threshold value at the time of switching from to the level 0 (normal flow) is set to a low value.

As described above, since the spatial parameters can be collected from an instantaneous image in principle, it is not necessary to accumulate data for 5 minutes or the like, and there is no time averaging process. The situation of the traffic flow can be immediately grasped, and rapid changes in the traffic flow at the time of occurrence of traffic congestion can be quickly and precisely grasped.

In the above description, the configuration in which the processing unit 12 of the space sensor 1 for measuring the space parameter and the traffic control central device 2 for determining traffic congestion are connected by a line has been described. It is also possible to adopt a configuration in which the determination of the traffic jam is executed by one computer.

The computer program for determining traffic congestion as described above can be provided by being preinstalled in a computer, or can be provided on a portable recording medium such as a CD-ROM or MO. Further, it can be provided via a line.

[0036]

As described above, according to the present invention, the spatial occupancy, the spatial average speed, the traffic density, and other spatial parameters of a vehicle are measured from the image at the moment when the road is overlooked, and the spatial occupancy and the spatial average speed are measured. Or the state of the traffic flow is determined based on whether the correlation between the traffic density and the spatial average speed is included in any of a plurality of areas obtained by dividing the correlation distribution area of these parameters according to the plurality of states of the traffic flow. Therefore, an excellent effect of rapidly or precisely catching a rapid change in traffic flow such as occurrence of traffic congestion in a wide detection range is achieved.

[Brief description of the drawings]

FIG. 1 is a block diagram showing a configuration of a traffic control system to which a traffic jam determination system according to the present invention is applied.

FIG. 2 is a diagram illustrating an example of a measurement range of one CCTV camera.

FIG. 3 is an example of a correlation graph between a spatial average speed and a space occupancy.

FIG. 4 is a conceptual diagram of a threshold value combination table.

FIG. 5 is a conceptual diagram in which a correlation distribution area of a space occupancy and a space average speed is divided into a plurality of areas according to a traffic flow state.

FIG. 6 is a correlation graph of spatial average speed-space occupancy, and spatial average speed-traffic density.

FIG. 7 is a flowchart of an example of a spatial parameter measurement procedure.

FIG. 8 is a flowchart illustrating an example of a traffic jam determination procedure.

[Explanation of symbols]

 Reference Signs List 1 space sensor 11 CCTV camera 12 processing unit 2 traffic control central device 21 threshold combination table 22 traffic congestion judgment means

 ────────────────────────────────────────────────── ─── Continuing on the front page (72) Inventor Kuniyoshi Nakatani 4-3-1 Kutarocho, Chuo-ku, Osaka-shi, Osaka F-term in Hanshin Expressway Public Corporation 5B057 AA16 DA06 DA20 DC02 DC19 5H180 AA01 CC04 CC11 CC24 DD02 DD04 5L096 BA04 CA03 FA34 FA70 GA02 GA08 HA03 9A001 HH23 JJ77 KK37

Claims (8)

[Claims] 1. A traffic congestion determination method for determining the state of traffic flow on a road by using an image obtained by photographing a road from a bird's-eye view, the method including a space occupancy and a space average speed of vehicles existing in a measurement range in the image. The spatial parameters are measured, and the correlation between the measured space occupancy and the spatial average speed is calculated by dividing the correlation distribution area between the spatial occupancy and the spatial average speed by the combination of the threshold for the spatial occupancy and the spatial average speed. A traffic congestion determination method characterized by determining a state of a traffic flow on a road according to which of a plurality of areas according to each of a plurality of states is included. 2. A combination of different thresholds for a case where a traffic congestion occurs when the traffic flow state is switched to a congestion state and a case where traffic congestion is resolved when the traffic flow state is switched to a normal state.
2. The traffic congestion determination method according to claim 1, wherein a correlation distribution region of the space occupancy and the space average speed is divided. 3. The traffic congestion determination method according to claim 1, wherein the traffic density is used for the determination instead of the space occupancy. 4. A traffic congestion determination system that determines a state of a traffic flow on a road by using an image obtained by overhead-viewing a road, including a space occupancy rate and a spatial average speed of a vehicle existing in a measurement range in the image. Means for measuring spatial parameters,
And a terminal device having means for transmitting the measured spatial parameters, and a means for receiving the spatial parameters transmitted from the terminal device, and a correlation distribution area of the space occupancy and the spatial average velocity corresponding to each of a plurality of states of the traffic flow. Means for storing a combination of the threshold values of the space occupancy rate and the space average speed, and a correlation between the measured space occupancy rate and the space average speed, which is divided into a plurality of areas, depending on which of the plurality of areas includes the road. A traffic congestion determination system, comprising: a central device having means for determining a traffic flow state. 5. A combination of different thresholds is stored in a case where a traffic congestion occurs when a traffic flow state is switched to a traffic congestion state and in a case where traffic congestion is switched from a traffic congestion state to a normal traffic state. The traffic congestion determination system according to claim 4. 6. The congestion determination system according to claim 4, wherein the traffic density is used for the determination instead of the space occupancy. 7. A traffic congestion determination device that determines a state of a traffic flow on a road using an image obtained by photographing a road from an overhead view, wherein a space occupancy and a spatial average speed of a vehicle existing in a measurement range in the image are measured. Means for dividing the correlation distribution area of the space occupancy rate and the space average speed into a plurality of areas corresponding to a plurality of states of the traffic flow, and storing a combination of threshold values of the space occupancy rate and the space average speed, Means for judging the state of traffic flow on the road based on which of the plurality of areas the correlation between the measured space occupancy rate and the spatial average speed is included in. 8. A method for calculating the state of traffic flow on a road using a space parameter including a space occupancy and a space average speed of a vehicle existing in a measurement range in the image measured from an image of a bird's-eye view of the road. A computer program which causes a computer to acquire a measured spatial parameter; and a computer which has a computer operable to determine a correlation between a space occupancy and a spatial average speed. Program code means for dividing the distribution area into a plurality of areas corresponding to a plurality of states of the traffic flow, referring to data of a combination of a threshold value of a space occupancy rate and a space average speed; and The correlation of the spatial average velocity is included in any of a plurality of areas divided by the combination of the thresholds. A computer program including program code means for judging a state of traffic flow on a road depending on whether or not the road is in a running state.