KR102330614B1 - Traffic signal control system and method thereof - Google Patents

Abstract

The present invention relates to a traffic signal control system and method therefor. According to the present invention, a traffic information acquisition unit that detects a vehicle from an image of a road to generate traffic information and detects at least one vehicle of interest passing in a lane of interest of the road in the image, and the location of the vehicle of interest A passage prediction information generating unit that calculates the expected time of passing the stop line of the vehicle of interest, and designates at least one vehicle of interest expected to pass the stop line within each time interval of the minimum green time of the traffic light and the green signal extension limit time as a threshold vehicle; and a control unit configured to determine at least one of whether or not to extend the green signal time of the traffic light and the extension time based on the expected time of passing the stop line of the critical vehicle and the signal time information of the traffic light.
According to the present invention, by analyzing the image of the road, traffic information such as the vehicle type, speed, and traffic amount of the vehicle passing on the road can be collected, and traffic in the lane of interest in the image just before the green signal of the traffic light installed on the road ends When a vehicle of interest is detected, by extending the green signal time of the traffic light so that the vehicle of interest passes through the stop line without stopping, it is possible to maximize passage efficiency in a limited road situation.

Description

TRAFFIC SIGNAL CONTROL SYSTEM AND METHOD THEREOF

The present invention relates to a traffic signal control system and method, and more particularly, to a traffic signal control system and method capable of varying the green signal time of a traffic light by analyzing an image of a road.

Recently, the problem of traffic congestion due to a rapidly increasing number of vehicles in limited road conditions has become a social and environmental issue. At the same time, various efforts are being made to solve traffic problems by collecting and analyzing traffic information from various parts of the road.

The traffic information collection system collects real-time vehicle information from a camera or sensor means (eg, a loop sensor, a laser sensor) on the road, and processes the collected information to provide traffic information, related images, and accident information.

Recently, a system for improving traffic flow by appropriately controlling the cycle of traffic signals based on traffic information collected from the road has been developed.

Also, in the transportation field, passenger & car equivalent (PCE) is used to predict traffic demand. Most commonly, small vehicles: 1.0, medium vehicles: 1.5, and large vehicles: 2.0 are applied based on flat roads. This means that the time it takes for one large car to pass on the road is the same as for two small cars to pass on the road.

In the case of medium or large vehicles, it is a factor that obstructs or obstructs the flow of traffic. In particular, if a heavy vehicle such as a dump truck is stopped in front of the stop line, even after the green signal is switched on, the following vehicle It is a major cause of impeding the smooth flow of traffic.

The technology that is the background of the present invention is disclosed in Korean Patent No. 0446461 (published on Aug. 30, 2004).

The present invention analyzes the image of the road to detect the vehicle expected to pass the stop line within the green signal time of the traffic light, and to improve the traffic flow by changing the green signal time of the traffic light so that the detected vehicle can pass the stop line without stopping An object of the present invention is to provide a traffic signal control system and method therefor.

The present invention provides a traffic information acquisition unit that detects a vehicle from an image of a road to generate traffic information and detects at least one vehicle of interest passing in a lane of interest of the road in the image, and based on the location of the vehicle of interest A passage prediction information generating unit that calculates the estimated time of passing the stop line of the vehicle of interest, and designates at least one vehicle of interest expected to pass the stop line within each time interval of the minimum green time of the traffic light and the green signal extension limit time as a threshold vehicle, and It provides a traffic signal control device including a control unit for determining at least one of whether or not to extend the green signal time of the traffic light and the extension time based on the expected time of passing the stop line of the critical vehicle and the signal time information of the traffic light.

In addition, the travel prediction information generating unit designates a vehicle of interest that is expected to pass the last stop line before the end of the minimum green time of the traffic light as the first threshold vehicle, and the end time of the minimum green time and the green signal extension limit time At least one vehicle of interest, which is expected to pass the stop line between the end time points of , may be designated as the second threshold vehicle.

Also, the vehicle of interest may include a heavy vehicle detected based on characteristic information of the vehicle.

In addition, the vehicle of interest may further include at least one vehicle traveling behind the heavy vehicle.

In addition, the vehicle of interest may further include a traffic-blocking vehicle having a speed less than a set speed detected based on the speed of the vehicle.

In addition, the traffic information acquisition unit may generate the traffic information including at least one of a vehicle type, a traffic volume, a queue length, a section speed, and an average speed by using the detection information and characteristic information of each vehicle in the image.

In addition, the traffic information acquisition unit may generate the traffic information using the image and the radar installed on the road.

In addition, the traffic information acquisition unit, when a time before a set time based on the end time of the minimum green time of the traffic light arrives, detects the vehicle of interest from the image and calculates the location and speed of the vehicle of interest to make the passage It can be provided to the prediction information generator.

Also, the travel prediction information generating unit may calculate an estimated time of passing the stop line of the vehicle of interest by using the location of the stop line in the image, the location and the speed of the vehicle of interest.

In addition, the travel prediction information generating unit calculates the estimated time of passing the stop line of the vehicle of interest by using the distance value corresponding to the area in which the vehicle of interest is detected and the speed of the vehicle of interest among a plurality of areas set along the road in the image. can

Also, the speed of the vehicle of interest may be a prescribed speed of a road in the image.

In addition, the travel prediction information generating unit may calculate the estimated time of passing the stop line for each vehicle of interest in the image when a time before a set time arrives based on the end time of the minimum green time of the traffic light.

In addition, the controller may determine whether to extend the green signal time and at least one of an extension time by comparing a current remaining time of the minimum green time of the traffic light and an expected time of passing a stop line of the threshold vehicle.

Also, when the estimated time of passing the stop line of the critical vehicle is greater than the current remaining time, the controller may determine the extension time within an extension limit time range after the minimum green time by using a time difference.

Also, the controller may determine the number of extension sections from among the multi-stage extension sections in which the extension limit time range is divided into a plurality of sections, based on the time difference.

In addition, when the estimated time of passing the stop line of the critical vehicle is greater than the current remaining time, and the time difference is greater than the extension limit time range after the minimum green time, the controller minimizes the extension time corresponding to the time difference. The signal time period of the traffic light may be corrected by applying after the green time period, and after the current signal time period is progressed, it may be restored to a preset original signal time period.

Also, the controller may determine whether to extend the green signal time by additionally considering surrounding environment information.

In addition, the apparatus may further include an environment information generator configured to collect the surrounding environment information including at least one of whether a pedestrian is waiting, whether a pedestrian button is operated, queue information of a left-turning lane, and information about a road ahead.

In addition, it may further include a signal time information receiving unit for receiving the signal time information including the green signal time operating in the traffic light.

In addition, the signal time information receiving unit may receive the signal time information from any one of a signal controller, an external control center, and a physical sensor sensing electrical characteristics applied to the traffic light.

In addition, the control unit detects a speed violation of the vehicle using the traffic information, detects a signal violation of the vehicle based on the traffic information and the signal time information, and collects the speed violation or enforcement information of the signal violation vehicle It can be created and provided as an operating device.

In addition, the traffic information obtaining unit is configured to set the lane of interest for a straight lane, or a lane causing congestion based on the traffic information, a lane in which the straight vehicle travel ratio is greater than or equal to a certain value, or a left-turning lane in which the saturation ratio is greater than or equal to a certain value. to set the interest lane.

In addition, the present invention provides a photographing device for photographing an image of a road, detecting a vehicle from the photographed image of the road to generate traffic information and enforcement information of the vehicle, and at least one interest passing in a lane of interest of the road in the image Calculate the estimated time of passing the stop line of the vehicle of interest based on the location of the vehicle, and designate at least one vehicle of interest expected to pass the stop line within each time interval of the minimum green time of the traffic light and the green signal extension limit time as a threshold vehicle, , a traffic signal control device for determining at least one of whether to extend the green signal time of the traffic light and an extension time based on the expected time of passing the stop line of the critical vehicle and the signal time information of the traffic light, and the vehicle from the traffic signal control device It provides a traffic signal control system including an operating device that receives and manages the detection information of the traffic information and the enforcement information.

In addition, the present invention includes the steps of generating traffic information by detecting a vehicle from an image of a road and detecting at least one vehicle of interest passing in a lane of interest of the road in the image, and the interest based on the location of the vehicle of interest Calculating the expected time of passing the stop line of the vehicle, and designating at least one vehicle of interest expected to pass the stop line within each time interval of the minimum green time of the traffic light and the green signal extension limit time as a critical vehicle, and the stop line of the critical vehicle Provided is a traffic signal control method comprising determining at least one of whether or not to extend the green signal time of the traffic light and the extension time based on the expected transit time and the signal time information of the traffic light.

Also, the vehicle of interest may include a heavy vehicle detected based on characteristic information of the vehicle.

Also, in the determining, at least one of whether to extend the green signal time and an extension time may be determined by comparing the current remaining time of the minimum green time of the traffic light with the calculated estimated time to pass the stop line.

According to the present invention, by analyzing an image of a road, traffic information such as vehicle type, speed, and amount of traffic of a vehicle passing on the road can be collected, and immediately before the green signal of a traffic light installed on the road ends, in the lane of interest in the image When a passing vehicle of interest is detected, traffic efficiency can be maximized in a limited road condition by extending the green signal time of the traffic light so that the vehicle of interest passes through the stop line without stopping.

In particular, when the green signal of the traffic light ends, the heavy vehicles on the road can pass through without stopping as much as possible, thereby minimizing the situation in which a low-speed vehicle stops at the end of the stop line, as well as a green signal again in the next signal cycle. It is possible to improve overall traffic flow by preventing unnecessary traffic delays such as departure delays and queues caused by slow acceleration of heavy vehicles when switching to

1 is a diagram illustrating a traffic signal control system according to an embodiment of the present invention.
FIG. 2 is a view showing a detailed configuration of the traffic signal control device shown in FIG. 1 .
3 is a diagram exemplarily showing a state in which a vehicle is detected for each lane in an image in an embodiment of the present invention.
4 is a diagram exemplarily illustrating a state in which a vehicle of interest is detected by analyzing an image according to an embodiment of the present invention.
5 is a view for explaining signal period information applied to a traffic light according to an embodiment of the present invention.
6 is a view for explaining a principle of estimating the location of a vehicle in an image by using a plurality of regions set in the image according to an embodiment of the present invention.
7 is a view illustrating a state in which a front blockage is captured from an image of a road in the opposite direction in an embodiment of the present invention.
FIG. 8 is a view for explaining a traffic signal control method using the traffic signal control system shown in FIG. 1 .

Hereinafter, embodiments of the present invention will be described in detail with reference to the accompanying drawings so that those of ordinary skill in the art to which the present invention pertains can easily implement them.

When a part "includes" a certain element throughout the specification, this means that other elements may be further included, rather than excluding other elements, unless otherwise stated. Also, as used herein, the term “unit” refers to a hardware component such as software, FPGA, or ASIC, and “unit” performs certain roles. However, 'part' is not limited to software or hardware. The 'unit' may be configured to reside on an addressable storage medium or may be configured to refresh one or more processors.

Thus, by way of example, “part” includes components such as software components, object-oriented software components, class components, and task components, processes, functions, properties, procedures, subroutines, segments of program code, drivers, firmware, microcode, circuitry, data, databases, data structures, tables, arrays and variables. The functionality provided within components and “parts” may be combined into a smaller number of components and “parts” or further divided into additional components and “parts”.

1 is a diagram illustrating a traffic signal control system according to an embodiment of the present invention.

As shown in FIG. 1 , the traffic signal control system according to an embodiment of the present invention includes a photographing device 100 , a traffic signal control device 200 , and an operating device 300 .

The photographing apparatus 100 captures an image of a road including a plurality of lanes, and provides the photographed image to the traffic signal control apparatus 200 . The photographing apparatus 100 includes a fixed (Fixed), rotation (PTZ) type camera, and may be operated by being mounted on a facility such as a holding on a road such as a general road, a highway, an unpaved road, an intersection point, or the like.

The photographing apparatus 100 may be connected to the traffic signal control apparatus 200 through a wired/wireless communication method to provide an image. Here, the wired/wireless communication method includes various known communication methods capable of transmitting and receiving data.

The traffic signal control device 200 receives an image from the photographing device 100, analyzes the received image, generates traffic information on the road, and considers the location of the vehicle of interest passing in the lane of interest in the image. It is possible to determine whether to extend the green signal time and at least one of the extension time. Here, the traffic light 500 is a device for displaying signals related to progress, stop, change of direction, and attention in the form of a light with respect to road traffic, but may also be implemented as a device for displaying signals in the form of characters and symbols.

The signal controller 400 may be provided in the enclosure around the road site or included in the server of the control center to control the signal cycle and switching of the traffic light 500 . The traffic signal control device 200 may be provided in the enclosure together with the signal controller 400 to communicate with the signal controller 400 , but may also be implemented in a form including the signal controller 400 , and the signal controller 400 . It can be installed separately by being connected to communication.

The traffic signal control device 200 may provide road traffic information (traffic information), vehicle information, surrounding environment information, green signal time control information, display information, and the like to the operation device 300 .

The operation device 300 is connected to the traffic signal control device 200 through a wired/wireless communication network, collects information from the traffic signal control device 200 installed at each point of the road, and synthesizes them to store various information for each point. , manage and analyze.

In addition, the operating device 300 provides real-time traffic information including vehicle detection information, characteristic information, traffic information, and enforcement information for each point on the road, signal control information, and various statistical information to various types of information devices (for example, , manager display, web service, road signage, etc.) can be provided to managers, drivers, and public transportation users.

FIG. 2 is a view showing a detailed configuration of the traffic signal control device shown in FIG. 1 .

As shown in FIG. 2 , the traffic signal control apparatus 200 according to an embodiment of the present invention includes a travel information acquisition unit 210 , a signal time information reception unit 220 , a travel prediction information generation unit 230 , and environment information generation. It includes a unit 240 , and a control unit 250 .

First, the traffic information acquisition unit 210 detects a vehicle from an image of a road to generate traffic information, sets or extracts a lane of interest among lanes of a road in the image, and detects at least one vehicle of interest existing in the lane of interest. do.

When the road form consists of a plurality of lanes, the traffic information acquisition unit 210 sets or extracts a lane of interest for a lane causing congestion based on the traffic information. For example, the traffic information acquisition unit 210 sets or extracts a lane of interest for a straight lane excluding right turns and left turns, or a lane in which the straight vehicle travel ratio is higher than a certain value or a left turn queue is saturated to the straight lane. A lane of interest can be set for an affected left-turn lane, etc. In addition, when a traffic pattern for a specific time period, such as commuting time, is changed, a lane of interest for each time period may be differently set or extracted.

The traffic information acquisition unit 210 may generate traffic information including vehicle type, traffic volume, queue length, section speed, average speed, etc. by using the detection information and characteristic information of each vehicle in the image, and each lane and direction of travel It is possible to generate and provide traffic information for each (straight, left/right turn).

The length of the queue is determined according to the number of vehicles and the vehicle type (vehicle size) forming the queue on the lane. Accordingly, in an embodiment of the present invention, the length of the queue may be a concept including the number and type of vehicles waiting on the corresponding lane.

The traffic information acquisition unit 210 may analyze the image to classify the types of vehicles such as passenger cars, RVs, heavy trucks, and large buses, and convert the number of vehicles stopped at the stop line and the types of vehicles into lengths of the queue. length can be calculated.

The traffic information acquisition unit 210 is capable of detecting and generating traffic information of vehicles traveling in multiple lanes (eg, 4 to 8 lanes) only with the image of one camera, so it is different from the existing loop sensor-based traffic information collection method. Otherwise, it has a great advantage that it does not require any maintenance of the sensor due to damage to the road surface and road control for this purpose.

3 is a diagram exemplarily showing a state in which a vehicle is detected for each lane in an image in an embodiment of the present invention.

3 shows a state in which various types of vehicles (cars, vans, trucks, buses, etc.) are detected for each lane in the image, and rectangular boxes of different colors are set for each detected vehicle for each classified vehicle type.

The traffic information acquisition unit 210 may detect all vehicle-related objects in the image from the image, and analyze characteristic information of the vehicle (eg, vehicle model, color, size, vehicle number, etc.) in response to the detected vehicle object. have. Also, the travel information acquisition unit 210 may perform real-time tracking by allocating an identifier to each vehicle, and may calculate the vehicle's driving speed by using the location for each time.

In general, the travel efficiency of vehicles on the road is related to the passenger car conversion factor (PCE) and varies depending on the type of vehicle.

Table 1 lists the vehicles waiting in each lane in order using identification codes, assuming a situation in which vehicles are stopped and waiting based on the stop line of the road at the intersection. From Table 1, the distribution of vehicle types for each lane can be confirmed.

'A' by car [Stop line] 321111 'B' by car [Stop line] 111123

In Table 1, identification codes 1, 2, and 3 refer to small, medium, and large vehicles, respectively. Therefore, in the case of 'A' by car, large (3), medium (2), small (1), ... , small (1) vehicles are on standby in that order, and in the case of lane 'B', small (1), small (1), small (1), … , and the large (3) vehicle indicates the situation in which the vehicle is on standby. At this time, in a situation where the traffic light 500 is switched back to a green signal and the vehicles start again, the sum of the time each vehicle leaves the stop line and passes through the intersection is calculated using the passenger car conversion coefficient set for each vehicle type. It has a smaller value at 'B' by car than A'.

In other words, lane 'A' has lower traffic efficiency than lane 'B', because the departure delay and low acceleration of the large vehicle located in front of lane 'A' causes the following vehicles to be blocked by the large vehicle at the front end. Because it is difficult to drive at that speed.

As such, when a large vehicle is stopped at the tip of the stop line, it interferes with acceleration or overtaking of subsequent vehicles when the green signal is switched, and causes traffic obstruction and traffic delay.

In an embodiment of the present invention, when a heavy vehicle including a large vehicle is found in the lane of interest of the road in the image when the green signal of the traffic light is imminent, the green signal time of the current time period is extended, so that the large Minimize vehicle stops, thereby minimizing traffic delays in the next signal cycle and increasing the overall traffic efficiency of the road.

The traffic information acquisition unit 210 may detect a vehicle from an image to generate detection information, and extract characteristic information from the image of the corresponding vehicle.

Here, the detection information of the vehicle may include an identifier (ID) assigned to the vehicle, a detection time, a detection position in an image, a speed, lane information, an image, and the like. In addition, the characteristic information of the vehicle may include at least one or more of color, type (vehicle), model, size, license plate information, etc. of the vehicle obtained by analyzing the image of the vehicle.

Here, of course, the traffic information acquisition unit 210 may additionally use an auxiliary device such as a radar installed on the road for accuracy of vehicle speed and location, or mix information detected by the radar with traffic information analyzed as an image. do.

The traffic information acquisition unit 210 detects at least one vehicle of interest passing in a lane of interest of a road in the image based on at least one of detection information and characteristic information of the vehicle in the image, and predicts the passage of the detected vehicle of interest information. It may be provided to the information generating unit 230 .

In addition, the traffic information acquisition unit 210 detects the vehicle of interest from the image and detects the vehicle of interest from the image when a time before the set time (eg, 5 seconds) arrives based on the end time of the minimum green time of the traffic light 500 . Position and velocity can be calculated. For example, if the green signal time of the vehicle on the road is 40 seconds, the vehicle of interest in the image is detected from the time when the green signal time has elapsed 35 seconds, and the location and speed of the vehicle of interest are calculated and provided. The load can be minimized, and it can be implemented with a minimum of hardware.

Here, the vehicle of interest may include all types of vehicles passing in the lane of interest or heavy vehicles detected based on vehicle characteristic information. In this case, the heavy vehicle may include a medium/large (dump) truck, a medium/large bus, and a trailer vehicle, and may be limited to 4 to 5 or more types of trucks and trailers according to road traffic conditions.

Of course, the heavy vehicle may be determined in the order of large (truck, bus), medium (truck, bus), and small (truck) on the road. That is, a heavy vehicle is a relative concept, and a road with mainly small and medium-sized vehicles may be a medium-sized vehicle, and a road with a mixture of small, mold, and large vehicles may be medium and large vehicles.

In addition, the vehicle of interest may further include a group of traffic impeding vehicles (low-speed driving vehicles) having a speed lower than a set speed detected based on the speed of the vehicle. The driving speed of a vehicle is generally affected by the driver's age, driving habits, experience, skill level, and the like. Considering this, the low-speed driving vehicle may also be classified as a vehicle of interest because there is a possibility of causing a departure delay even in the next signal cycle.

This embodiment of the present invention detects a low-speed vehicle among vehicles traveling in the green signal-linked driving direction when the green signal of the traffic light 500 is imminent and estimates whether the low-speed vehicle can pass the stop line, and displays The extension time of the green signal in progress can be varied.

In addition, the vehicle of interest may include a vehicle group that is clustered in the same area transversely to the plurality of lanes, ie, different lanes from the stop line at the same and similar position, and obstructs traffic across the plurality of lanes. Hereinafter, for convenience of explanation, determining a heavy vehicle such as a dump truck as a vehicle of interest is representatively exemplified.

4 is a diagram exemplarily illustrating a state in which a vehicle of interest is detected by analyzing an image according to an embodiment of the present invention.

4 (a) shows the results of detecting each of the waiting vehicles when the traffic light 500 is a red signal. Here, the traffic information acquisition unit 210 may calculate the length of the queue by analyzing the type and number of vehicles waiting for each lane, and may analyze and provide the distribution of vehicle types for each lane.

4B shows an image in which the traffic light 500 is converted from a red signal to a green signal and a vehicle passes. At this time, the traffic information acquisition unit 210 may detect the traffic amount of the vehicle for each lane as well as the traffic amount of the vehicle for each direction (straight, left, right turn) of the vehicle, and pay attention to the straight-line lane among each lane. It can be set as a lane.

4 (c) corresponds to the image obtained before the end of the green signal time of the traffic light 500, and the heavy vehicle corresponding to the dump truck in the image is detected as the vehicle of interest 10 (yellow box) is shown. 4( d ) is an image obtained when the green signal time is imminent (eg, 5 seconds before the end), and shows a state in which the vehicle of interest 10 is closer to the stop line.

In this embodiment, one heavy vehicle is detected as the vehicle of interest 10 (yellow box), but a plurality of heavy vehicles may be detected as the vehicle of interest. It is also possible to detect as a vehicle of interest.

Here, the travel information obtaining unit 210 may calculate or estimate the location and speed of the vehicle of interest detected in the image and provide it to the travel prediction information generating unit 230 . Thereafter, the travel prediction information generating unit 230 may predict an expected required time required for the vehicle of interest to reach a stop line by using the location and speed of the vehicle of interest. This will be described in detail later.

On the other hand, the signal time information receiving unit 220 receives the signal time information (time table) including the green signal time operating in the traffic light 500, and the received signal time information to the travel information obtaining unit 210 and the travel prediction information It may be provided to the generator 230 and the controller 250 .

Here, the signal time information receiving unit 220 receives the signal period and the current information of the traffic light 500, the signal period uses a unit of seconds (Second). For example, if the signal period of the intersection is "180 seconds", the signal time for each direction is allocated within 180 seconds, and when time is allocated, the road type, traffic volume, road condition, etc. are considered in a complex way.

5 is a view for explaining signal period information applied to a traffic light according to an embodiment of the present invention.

As shown in FIG. 5 , a signal time (presentation) including green, yellow, and red signal sections is allocated for each road direction of the intersection. Here, the green signal time includes the minimum green time and the extension limit time, and must not exceed the maximum green time, which is the sum of the minimum green time and the extension limit time. Therefore, the extension time must also be determined within the extension limit time range.

The extension limit time includes N extension sections divided by a set time interval (eg, 1 second, 3 seconds, 5 seconds) as shown in FIG. 5 . Accordingly, the extension time of the green signal time is adjustable through the number of extension sections. In the case of FIG. 5 , it is described that N=3, but the present invention is not necessarily limited thereto. Here, as N increases, the time interval decreases, but may be further subdivided into multi-stage sections.

Such a signal time information receiving unit 220 may receive signal time information by communicating with the signal controller 400, and communicate with a server of a control center that integrally controls a plurality of signal controllers 400 to receive signal time information. may be In addition, the signal time information receiving unit 220 senses the electrical signal characteristics over time through a physical sensor (signal detection device) installed on a terminal or wire applied to the signal light 500 from the signal controller 400. Time information may also be obtained.

The travel prediction information generating unit 230 calculates the estimated time required for the vehicle of interest to pass through the stop line of the road (hereinafter, the estimated time of passing the stop line) based on the location of the vehicle of interest in the image, and controls the calculated information. forward to (250). Then, the controller 250 determines at least one of whether or not to extend the green signal time of the traffic light 500 and the extension time based on the calculated estimated time of passing the stop line.

Specifically, the travel prediction information generating unit 230 calculates the estimated time of passing the stop line by using the location of the stop line in the image, the location and speed of the vehicle of interest. Here, the estimated time of passing the stop line means an estimated time required for the vehicle of interest to reach the stop line. That is, the time required for the vehicle of interest to pass through the stop line of the road is calculated based on the distance between the stop line and the vehicle of interest in the image and the speed of the vehicle.

For example, if a vehicle traveling at a speed of 50 Km/h is at a position 50 m from the stop line, the time required for the vehicle to cross the stop line is calculated as 3.7 seconds.

In this case, if it is extended for about 4 seconds or more, the vehicle can pass through the stop line without stopping. time extension is required. Of course, if the vehicle was at a location of 100 m, a time extension of two sections (10 seconds in total) including the first and second sections is required.

As another example, it is assumed that the minimum green time is 50 seconds, the extension limit time is 6 seconds, and a situation in which interest "A" and "B" are detected in the image. If the car of interest "A" and "B" are driving 50m and 70m behind the stop line, respectively, and the driving speed is about 50 km/h, the car of interest "A" is about 3.7 seconds and "B" is about 3.7 seconds until passing the stop line It takes about 5.2 seconds. At this time, the extension time for both vehicles A and B of interest to pass the stop line without stopping is 6 seconds, which is within the extension limit time, so the extension time is determined to be 6 seconds.

In an embodiment of the present invention, it is preferable to apply the actual calculated speed of the corresponding vehicle to the vehicle speed, but if necessary, the prescribed speed of the corresponding road and the average speed of the vehicle may be applied. For example, in an environment in which accurate speed detection through image analysis is difficult, the speed limit of the corresponding road or the real-time average travel speed calculated by the travel information acquisition unit 210 may be used as the vehicle speed value.

In addition, depending on the installation height of the camera for acquiring the image of the road, the slope of the road, weather and night lighting environment, etc., the position and speed of the vehicle may inevitably occur. In this case, if the resolution of the image is increased, the speed error is somewhat reduced, but this increases the hardware load and computational complexity required for image transmission and analysis.

In order to rationally solve this problem, the travel prediction information generating unit 230 may set a plurality of regions of interest in the corresponding road image and calculate the estimated time of passing the stop line using them.

6 is a view for explaining a principle of estimating the location of a vehicle in an image by using a plurality of regions set in the image according to an embodiment of the present invention.

The travel prediction information generating unit 230 uses the distance value corresponding to the area in which the vehicle of interest is detected and the speed of the vehicle of interest among a plurality of areas S1, S2, and S3 set along the road in the image as shown in FIG. 6 , An estimated time required for the vehicle of interest to pass through the stop line may be calculated.

For example, it is assumed that a plurality of regions of interest (S1, S2, S3, S4, S5, S6, S7, ??) are set at intervals of 10 m from the stop line. If the driving speed of the vehicle of interest in the image is 50 km/h and the vehicle of interest is detected within the S6 area, the time required to the stop line is calculated as about 3.7 to 4.4 seconds because the vehicle is located in the 50 to 60 m area, so it is extended The time is finally determined to be 5 seconds.

Of course, even in this case, the speed limit or the average speed of the road can be used as the speed value of the vehicle.

In addition, the travel prediction information generating unit 230 does not perform the above-described operation for the entire minimum green time of the traffic light 500 so as to minimize the hardware load, and the minimum green time is a time before the set time based on the end time. When (at least 5 seconds before the end of the green time) arrives, the estimated time of passing the stop line can be calculated for each vehicle of interest in the image.

In addition, the traffic prediction information generation unit 230 is based on the expected time of passing the stop line of the vehicle of interest and the signal time information of the traffic light 500, the stop line within each time section of the minimum green time of the traffic light 500 and the green signal extension limit time At least one vehicle of interest expected to pass is designated as a critical vehicle, and detection information and characteristic information of the designated critical vehicle are transmitted to the controller 250 .

In detail, the travel prediction information generating unit 230 designates the vehicle of interest that is expected to pass the last stop line before the end of the minimum green time of the traffic light 500 as the first threshold vehicle, and the end of the minimum green time and At least one vehicle of interest expected to pass the stop line between the end of the green signal extension limit time is designated as the second critical vehicle, and the detection information and characteristic information of the designated first and second critical vehicles are transmitted to the controller 250 . transmit

The environmental information generating unit 240 collects and generates state information around the road on which the traffic light 500 controlled by the signal controller 400 is installed, which is affected by the passage due to the extension time of the green signal time of the vehicle of interest. Real-time generation or collection of surrounding environment information (eg, cross-road situation, left-turn response, pedestrian presence, and front-blocking information).

Specifically, the environment information generating unit 240 collects surrounding environment information from peripheral devices in the field, such as a plurality of photographing devices installed in each direction of the intersection and a pedestrian manipulation button corresponding to the current road, and provides it to the control unit 250 . . In this case, the surrounding environment information may include whether a pedestrian is waiting, whether a pedestrian button is operated (manipulated), queue information for a left-turning lane, information about blocking a road ahead, and the like.

If a pedestrian is detected near the crosswalk on the road in the video, there is a pedestrian waiting signal by the pedestrian manipulation button, or the left turn queue on the road in the video occurs longer than the set length, or if the video of the road crossing the intersection is already If a road blockage is found, not extending the green signal time can actually help traffic flow.

As such, if the green signal time is extended in a situation in which the front of the road is blocked, left-turn queues, pedestrian waiting, etc. are detected, it may interfere with pedestrian traffic or traffic flow, or interfere with and impede traffic flow due to additional traffic delay. There are concerns.

7 is a view illustrating a state in which a front blockage is captured from an image of a road in the opposite direction in an embodiment of the present invention.

The image of FIG. 7 exemplifies an image of a photographing device installed on a road in the opposite direction to the corresponding road to be analyzed, and it is confirmed whether the road is blocked through the area at the right end of the image that is visible beyond the center line can be

That is, the state of being blocked in front of the current road is easily confirmed through the right area of the image of the photographing apparatus installed on the road opposite to the current road among photographing apparatuses installed in each road direction on the intersection.

If such a front-blocking occurs, not extending the green signal time has the effect of preventing tail-biting at the intersection in advance, thereby facilitating the flow of traffic.

As such, the environmental information generating unit 240 performs a function of generating or collecting and providing environmental information necessary to finally determine whether to apply the time extension to the traffic light 500 .

Of course, the surrounding environment information may further include a crosswalk pedestrian queue, whether a pedestrian button is operated, and not only the traffic condition of the crossroads, but also whether the main road/sub-road, and traffic information and traffic conditions of the intersection immediately before/after.

In addition, the environment information generation unit 240 may receive surrounding environment information from a peripheral device (photographing device, radar, etc.) that detects a pedestrian's queue, or it is possible to directly generate surrounding environment information by analyzing data provided by the surrounding device And, if there is a pedestrian operation button, it is possible to receive the operation information of the button and include it in the environmental information.

The control unit 250 controls the data flow of each of the travel information obtaining unit 210, the signal time information receiving unit 220, the travel prediction information generating unit 230, and the environment information generating unit 240, and each unit (210, 220, 230, 240) data can be exchanged with

The control unit 250 receives the estimated stop line passage time calculated in response to the threshold vehicle from the travel prediction information generation unit 230, and based on the estimated stop line passage time of the critical vehicle and the signal time information of the traffic light 500, the traffic light ( 500) determines whether to extend the green signal time and at least one of the extension time.

Here, the controller 250 determines whether to extend the green signal time by comparing the current remaining time of the minimum green time of the traffic light 500 and the calculated estimated stop line passing time.

If the estimated time of passing the stop line is greater than the current remaining time, the controller 250 determines that the green signal time needs to be extended, and determines the extension time using the time difference. In this case, the control unit 250 determines the number of extension sections from among the multi-stage extension sections in which the extension limit time range is divided into N sections, and the number of extension sections can be selected according to the time difference.

For example, if the extension limit time after the minimum green time is 6 seconds and includes 6 extension sections with an interval of 1 second, if the time difference between the current remaining time of the minimum green time and the calculated stop line passing estimate is 2.5 seconds, , the control unit 250 determines the extension time to be 3 seconds, and may select three initial periods from among the six extension periods. Of course, if the time difference was 7 seconds, the extension time may be determined to be 6 seconds. The controller 250 may transmit the determined extension time information to the traffic light 500 to extend the green signal time of the traffic light 500 .

Here, in consideration of the surrounding environment information received from the environment information generating unit 240 , the control unit 250 finally determines whether to actually apply the corresponding extended time to the traffic light 500 , and the extended time according to the determination result may or may not be applied to the traffic light 500 .

That is, the control unit 250 determines whether or not there is an interference factor in the vicinity from the surrounding environment information received from the environment information generating unit 240. If there is no interference factor, the control unit 250 applies the extension time to the traffic light 500 to give a green signal. The time is extended, but the extension time is not applied if there is an interference factor.

In this way, the controller 250 may finally determine whether to extend the extension in consideration of the surrounding environment information, generate time control information including the extension time, and provide it to the traffic light 500 .

The control unit 250 may include a setting function such as communication with a signal controller 400, a signal control server or a signal detection device (not shown), communication with peripheral devices for collecting environmental information, and a prescribed speed of the corresponding road. have.

In addition, when the traffic information acquisition unit 210 is continuously configured at the intersection of the road network, the control unit 250 receives the traffic information from the traffic information acquisition unit of the intersection immediately before or immediately after, and uses the information on the environment information of the current intersection. can be used as In addition, information on the occurrence of accidents at the intersection immediately ahead enables the driver to select a detour route, thereby reducing traffic congestion.

In addition, the control unit 250 detects a speed violation of the vehicle using the travel information obtained through the travel information obtaining unit 210, and the traffic light 500 received through the travel information and the signal time information receiving unit 220. By detecting a signal violation of a vehicle based on the signal time information of

In addition, the control unit 250 may receive the detection information and characteristic information of the threshold vehicle, that is, the first threshold vehicle and the second threshold vehicle from the travel prediction information generation unit 230 , and provide it to the operation device 300 .

In an embodiment of the present invention, the travel information acquisition unit 210 and the travel prediction information generation unit 230 include hardware and algorithms for performing image analysis, and each component is implemented by cloud, edge computing, or a mixture thereof. possible, and is not limited to a local configuration, a center configuration, and the like.

Meanwhile, although the embodiment of the present invention exemplifies determining the extension time of the green signal time within the extension limit time, the present invention is not necessarily limited thereto.

For example, both the freeway intersection and the interlocking intersection may allow time extension of the green signal time beyond the extension limit time range.

However, when the green signal time is extended within the extension limit time, since the time extension is performed only within the signal time period as in FIG. 5, the signal time period itself does not change.

However, when the green signal time is extended beyond the extension limit time range, the signal time period itself is changed. In this case, in the case of an interlocking intersection, since there is a risk of signal entanglement between intersections interworking with each other in the future, it is necessary to return to the original signal period again after the current signal time period has progressed.

Therefore, the controller 250 applies the extension time corresponding to the time difference after the minimum green time if the expected time of passing the stop line is greater than the current remaining time, and the time difference is greater than the extension limit time range after the minimum green time. Accordingly, the signal time period of the traffic light 500 may be corrected, and after the modified current signal time period has progressed, it may be restored to the preset original signal time period. This method can be mainly applied to interlocking intersections rather than independent intersections.

FIG. 8 is a view for explaining a traffic signal control method using the traffic signal control system shown in FIG. 1 .

First, the traffic information acquisition unit 210 of the traffic signal control device 200 receives an image from the photographing device 100 and analyzes the image to generate traffic information (S810). Here, the traffic information acquisition unit 210 may generate traffic information including vehicle type, traffic volume, queue length, section speed, and average speed by using detection information and characteristic information of the vehicle in the image.

Then, the traffic information acquisition unit 210 analyzes the image to set a lane of interest among the lanes of the road in the image, and detects at least one vehicle of interest passing in the lane of interest ( S820 ). Here, the vehicle of interest targets a low-speed driving vehicle such as a heavy vehicle, but may also target all types of vehicles.

The traffic information acquisition unit 210 starts to detect the vehicle of interest in the image from the end of the green signal time before the set time (at least 5 seconds before the end of the green time), thereby minimizing the burden on hardware.

The travel prediction information generating unit 230 calculates the estimated time required for the vehicle of interest to pass through the stop line of the road based on the location of the vehicle of interest detected in the image, and provides it to the controller 250 (S830) .

In this case, the travel prediction information generating unit 230 may calculate the estimated time to pass the stop line by using the relative position (distance) and speed of the vehicle of interest with respect to the stop line in the image, but the vehicle of interest among a plurality of areas of interest set in the image is It is also possible to calculate the estimated time of passing the stop line by using the distance value matched to the located area and the speed of the vehicle of interest.

On the other hand, the travel prediction information generation unit 230 is based on the expected time of passing the stop line of the vehicle of interest and the signal time information of the traffic light 500, the stop line within each time section of the minimum green time of the traffic light 500 and the green signal extension limit time At least one vehicle of interest expected to pass is designated as a critical vehicle, and detection information and characteristic information of the designated critical vehicle are transmitted to the controller 250 .

In detail, the travel prediction information generating unit 230 designates the vehicle of interest that is expected to pass the last stop line before the end of the minimum green time of the traffic light 500 as the first threshold vehicle, and the end of the minimum green time and At least one vehicle of interest expected to pass the stop line between the end of the green signal extension limit time is designated as the second critical vehicle, and the detection information and characteristic information of the designated first and second critical vehicles are transmitted to the controller 250 . transmit

The control unit 250 compares the estimated time of passing the stop line calculated corresponding to the threshold vehicle and the remaining time of the current green signal time, and determines whether to extend the green signal time and the extension time (S840). In this case, the extension time of the green signal time may be determined within the extension limit time range.

Thereafter, the control unit 250 checks the surrounding environment information collected by the environment information generating unit 240 (S850). Here, the surrounding environment information may include whether a pedestrian is waiting, whether a pedestrian button is operated, queue information for a left-turning lane, information about blocking a road ahead, and the like.

The control unit 250 determines whether there is an interference factor in the vicinity by checking the surrounding environment information (S860). For example, if at least one of a road blockage, a left turn queue, and a queue of more than a certain number of pedestrians occurs, it is determined that there is an interference factor nearby.

If there is no interference factor, the controller 250 extends the green signal time by directly applying the extension time determined in step S840 to the traffic light 500 (S870), but if there is an interference factor, the extension signal is not applied ( S880).

Meanwhile, the control unit 250 receives the detection information and characteristic information of the threshold vehicle, that is, the first threshold vehicle and the second threshold vehicle from the travel prediction information generation unit 230 , and transmits it to the operation device 300 . The operating device 300 displays the information received from the control unit 250 through an information device such as a manager display, so that the manager can intuitively monitor the detection information and characteristic information of the critical vehicle in real time through the display screen. have.

According to the present invention as described above, when a vehicle of interest including a heavy vehicle is detected in the image when the end of the green signal is imminent, the green signal time of the traffic light 500 is extended and controlled so that the vehicle passes the stop line without stopping. Traffic efficiency can be maximized in limited road conditions.

In particular, it allows heavy vehicles on the road to pass without stopping as much as possible around the end of the green signal, resulting in delays in departure and queues of subsequent vehicles due to slow acceleration of heavy vehicles when switching to a green signal in the next signal cycle. It is possible to minimize unnecessary traffic delays, such as etc., as well as improve overall traffic flow on the road.

Although the present invention has been described with reference to the embodiment shown in the drawings, which is only exemplary, those skilled in the art will understand that various modifications and equivalent other embodiments are possible therefrom. Accordingly, the true technical protection scope of the present invention should be determined by the technical spirit of the appended claims.

100: photographing device 200: traffic signal control device
210: traffic information obtaining unit 220: signal time information receiving unit
230: travel prediction information generation unit 240: environmental information generation unit
250: control unit 300: operating device
400: signal controller 500: traffic light

Claims (26)

a traffic information acquisition unit generating traffic information by detecting a vehicle from an image of a road and detecting at least one vehicle of interest passing in a lane of interest of the road in the image;
The estimated time of passing the stop line of the vehicle of interest is calculated based on the location of the vehicle of interest, and at least one vehicle of interest expected to pass the stop line within each time interval of the minimum green time of the traffic light and the green signal extension limit time is set as the threshold vehicle. a designated travel prediction information generation unit;
an environment information generator for collecting surrounding environment information including at least one of whether a pedestrian is waiting, whether a pedestrian button is operated, queue information for a left-turn lane, and information about a road ahead; and
At least one of whether or not to extend the green signal time of the traffic light and the extension time is determined based on the expected time of passing the stop line of the critical vehicle and the signal time information of the traffic light, but when there is no interference factor in the surrounding environment information a control unit that applies the extension of the green signal time of the traffic light to a threshold vehicle, and determines not to extend the green signal time when the interference factor exists;
The control unit is
If a pedestrian is detected around the crosswalk on the road in the image, there is a pedestrian waiting signal by the pedestrian button, or the left turn queue of the road in the image occurs longer than the set length, or the front of the road is blocked, the green signal Traffic signal control device to ensure that the time is not extended. The method according to claim 1,
The travel prediction information generating unit,
The vehicle of interest that is expected to pass the last stop line before the end of the minimum green time of the traffic light is designated as the first threshold vehicle, and the stop line is expected to pass between the end of the minimum green time and the end of the green signal extension limit time A traffic signal control device for designating at least one vehicle of interest to be a second threshold vehicle. The method according to claim 1,
The vehicle of interest is
A traffic signal control apparatus including a heavy vehicle detected based on the characteristic information of the vehicle. 4. The method according to claim 3,
The vehicle of interest is
The traffic signal control device further comprising at least one vehicle traveling behind the heavy vehicle. The method according to claim 1,
The vehicle of interest is
Traffic signal control device further comprising a vehicle that is less than a set speed detected based on the speed of the vehicle. The method according to claim 1,
The travel information acquisition unit,
A traffic signal control device for generating the traffic information including at least one of a vehicle model, a traffic volume, a queue length, a section speed, and an average speed by using the detection information and characteristic information of each vehicle in the image. The method according to claim 1,
The travel information acquisition unit,
A traffic signal control device for generating the traffic information using the image and the radar installed on the road. The method according to claim 1,
The travel information acquisition unit,
When a time point before a set time arrives based on the end time of the minimum green time of the traffic light, the vehicle of interest is detected from the image, the location and speed of the vehicle of interest are calculated, and the traffic signal provided to the travel prediction information generator controller. The method according to claim 1,
The travel prediction information generating unit,
A traffic signal control apparatus for calculating an estimated time of passing the stop line of the vehicle of interest by using the position of the stop line in the image, the position and speed of the vehicle of interest. The method according to claim 1,
The travel prediction information generating unit,
A traffic signal control apparatus for calculating an expected time of passing a stop line of the vehicle of interest by using a distance value corresponding to a region in which the vehicle of interest is detected and a speed of the vehicle of interest among a plurality of areas set along the road in the image. 11. The method of claim 10,
The speed of the vehicle of interest is a traffic signal control device that is a prescribed speed of the road in the image. The method according to claim 1,
The travel prediction information generating unit,
When a time point before a set time arrives based on the end time of the minimum green time of the traffic light, the traffic signal control apparatus calculates the estimated time of passing the stop line for each vehicle of interest in the image. The method according to claim 1,
The control unit is
A traffic signal control device for determining whether to extend the green signal time and at least one of an extension time by comparing the current remaining time of the minimum green time of the traffic light and an expected time of passing the stop line of the threshold vehicle. 14. The method of claim 13,
The control unit is
When the estimated time of passing the stop line of the critical vehicle is greater than the current remaining time, the apparatus for controlling a traffic signal determines the extension time within an extension limit time range after the minimum green time by using a time difference. 15. The method of claim 14,
The control unit is
A traffic signal control apparatus for determining the number of extension sections among multi-stage extension sections in which the extension limit time range is divided into a plurality of sections based on the time difference. 14. The method of claim 13,
The control unit is
When the estimated time of passing the stop line of the critical vehicle is greater than the current remaining time, and the time difference is greater than the extension limit time range after the minimum green time, an extension time corresponding to the time difference is applied after the minimum green time. A traffic signal control apparatus for correcting the signal time period of the traffic light to return to a preset original signal time period after the current signal time period is progressed. delete delete The method according to claim 1,
The traffic signal control apparatus further comprising a signal time information receiver for receiving signal time information including the green signal time operating at the traffic light. 20. The method of claim 19,
The signal time information receiving unit,
A traffic signal control device for receiving the signal time information from any one of a signal controller, an external control center, and a physical sensor sensing electrical characteristics applied to the traffic light. The method according to claim 1,
The control unit is
Detects a vehicle speed violation using the traffic information, detects a vehicle signal violation based on the traffic information and the signal time information, and generates and provides enforcement information on the speed violation or signal violation vehicle to the operating device traffic signal control device. The method according to claim 1,
The travel information acquisition unit,
Traffic that sets the lane of interest for a straight-through lane, or sets the lane of interest for a congestion-causing lane, a lane in which the straight-through vehicle travel ratio is above a certain value, or a left-turn lane in which the saturation ratio is above a predetermined value based on the traffic information signal control unit. a photographing device for photographing an image of a road;
Detecting a vehicle from the photographed image of the road generates traffic information and enforcement information of the vehicle, and calculating an estimated time of passing the stop line of the vehicle of interest based on the location of at least one vehicle of interest passing in the lane of interest of the road in the image and designate at least one vehicle of interest expected to pass the stop line within each time interval of the minimum green time of the traffic light and the green signal extension limit time as a threshold vehicle, the estimated time of passing the stop line of the critical vehicle and the signal time information of the traffic light a traffic signal control device for determining whether to extend the green signal time of the traffic light and at least one of an extension time based on ; and
and an operating device for receiving and managing vehicle detection information, the traffic information, and the enforcement information from the traffic signal control device,
The traffic signal control device,
If there is no interference factor in the surrounding environment information including at least one of pedestrian waiting, pedestrian button operation, left turn lane queue information, and road blocking information, the green signal time extension of the traffic light is applied to the critical vehicle and determine not to extend the green signal time when the interference factor is present,
If a pedestrian is detected around the crosswalk on the road in the image, there is a pedestrian waiting signal by the pedestrian button, or the left turn queue of the road in the image occurs longer than the set length, or the front of the road is blocked, the green signal Traffic signal control system to avoid extending the time. generating traffic information by detecting a vehicle from an image of a road and detecting at least one vehicle of interest passing in a lane of interest of the road in the image;
The estimated time of passing the stop line of the vehicle of interest is calculated based on the location of the vehicle of interest, and at least one vehicle of interest expected to pass the stop line within each time interval of the minimum green time of the traffic light and the green signal extension limit time is set as the threshold vehicle. specifying;
determining at least one of whether to extend the green signal time of the traffic light and the extension time based on the expected time of passing the stop line of the critical vehicle and the signal time information of the traffic light;
collecting surrounding environment information including at least one of whether a pedestrian is waiting, whether a pedestrian button is operated, queue information for a left-turn lane, and information about a road ahead;
determining whether an interference factor exists from the surrounding environment information; and
applying the green signal time extension of the traffic light to the threshold vehicle when the interference factor does not exist, and determining not to extend the green signal time when the interference factor exists,
Determining not to extend the green signal time comprises:
If a pedestrian is detected around the crosswalk on the road in the image, there is a pedestrian waiting signal by the pedestrian button, the left turn queue of the road in the image occurs longer than the set length, or the front of the road is blocked, the green signal A method of controlling traffic signals so as not to prolong the time. 25. The method of claim 24,
The vehicle of interest is
A traffic signal control method including a heavy vehicle detected based on the characteristic information of the vehicle. 25. The method of claim 24,
The step of determining whether to extend the green signal time of the traffic light and at least one of the extension time includes:
A traffic signal control method for determining whether to extend the green signal time and at least one of an extension time by comparing the current remaining time of the minimum green time of the traffic light with the calculated estimated stop line passing time.

Images