KR102177614B1 - Lane generation system for collecting traffic information and method thereof - Google Patents

Abstract

The present invention relates to a lane generation system for collecting traffic information and a method thereof. According to the present invention, the lane generation system for collecting traffic information comprises: a vanishing point extraction unit extracting a vanishing point by using an image of a road acquired from a photographing device; a vehicle detection unit detecting both end points of a lower end of a vehicle for each of vehicles appearing in the image; a lane estimation unit estimating at least one lane region in the image on the basis of a triangular region generated by connecting the vanishing point of the road to the both end points of the lower end of the vehicle; and a control unit applying the estimated lane region within the image. The present invention can estimate each lane region on the basis of a result of overlapping the region generated by connecting the vanishing point of the road within the taken image of the road to the both end points of the lower end of the vehicle to automatically apply the lane region within the image and, based thereon, can effectively provide a traffic volume for each lane.

Description

Lane generation system for collecting traffic information and its method {LANE GENERATION SYSTEM FOR COLLECTING TRAFFIC INFORMATION AND METHOD THEREOF}

The present invention relates to a lane generation system and method for collecting traffic information, and more particularly, to a lane generation system and method for collecting traffic information capable of automatically setting a lane area within an image by analyzing a captured image. About.

Recently, many social problems have arisen due to an increase in traffic volume, an increase in vehicles, and improvement in vehicle performance. Accordingly, various information collection devices have been installed to measure traffic information or monitor illegal behavior of vehicles on various roads.

Among them, the traffic information collection device captures an image of a road using a camera fixed to a facility on the road and provides traffic volume and traffic information of the road based on a result of detecting a vehicle passing through a set point in the image.

However, in order to provide traffic information for each lane by analyzing the image of the road through a single camera, it is necessary to identify or recognize each lane from the image and then set the lane within the image. Traffic information for each lane can be provided by comparing the area of each lane set in the image and the detection area of the vehicle.

However, if the position of the camera is shifted or shaken due to wind or external force, the lane area preset in the image and the lane area shown in the actual image do not coincide with each other on the screen, which may cause false positives during image processing. The possibility exists.

As a method for solving this problem, a method of generating a lane based on a vehicle movement trajectory has been disclosed. However, this method requires precise tracking (tracking) of vehicles in the video and requires at least 50 vehicle trajectories to be collected.Therefore, in lanes or roads with low traffic, the inconvenience of manually setting the vehicle movement trajectory in the control center exist.

The technology behind the present invention is disclosed in Korean Patent Publication No. 2010-0047053 (published on May 7, 2010).

The present invention provides a lane generation system for collecting traffic information capable of quickly estimating each lane area of a road and applying the estimated lane area in an image using the vanishing point of the road detected in the image and both end points of the lower part of the vehicle, and It aims to provide a method.

The present invention provides a vanishing point extracting unit for extracting a vanishing point using an image of a road obtained from a photographing device, a vehicle detecting unit for detecting both end points of the lower end of the vehicle for each vehicle appearing in the image, and the vanishing point of the road Traffic information collection comprising a lane estimating unit for estimating at least one lane area in an image based on a triangular area generated by connecting both end points of the lower end of the vehicle and a control unit for applying the estimated lane area to the image It provides a car generation system for

In addition, the lane estimating unit may estimate at least one lane area in the image based on a result of overlapping the triangular area for each time or vehicle.

In addition, the vanishing point extracting unit may extract the vanishing point using a boundary line of the road detected in the image.

In addition, the vanishing point extractor may detect a boundary line of the road from the image using a stick cell technique.

In addition, when the boundary line of the road is curved, the vanishing point extracting unit may divide the road into a plurality of road sections and extract vanishing points for each road section to obtain multiple vanishing points along the road.

In addition, the vanishing point extractor may individually extract a vanishing point corresponding to a corresponding road section by using a virtual straight line generated based on a boundary line of a corresponding road section for each road section.

In addition, the virtual straight line may be generated by connecting a start point and an end point of the boundary line with a straight line for each boundary line of the corresponding road section.

In addition, the lane estimator generates a triangular area corresponding to the vehicle by using the vanishing point of the road section in which the vehicle is located among the plurality of road sections, overlaps the triangular area for each road section, and acquires it on each road section. The lane area may be estimated from the result of connecting the overlapped areas along the entire road section.

In addition, the lane estimating unit may update the lane region by overlapping the newly acquired triangular region in the image with the lane region previously generated in the previous time, and provide it to the control unit.

In addition, the lane estimating unit may update at least one of a shape, an area, a length, and a width of the lane area.

In addition, the vehicle detection unit may set a rectangular box surrounding an outer edge of the object corresponding to the vehicle object detected in the image, and then detect both end points of the lower end of the vehicle within the rectangular box.

In addition, the vehicle detection unit may detect the speed of the vehicle by using a position change of the lower portion of the vehicle detected in correspondence with the vehicle object in the image by time, and provide vehicle detection information including the speed for each vehicle.

Further, the controller may calculate and provide an average speed for each lane based on the speeds of vehicles detected for each lane area in the image.

In addition, the present invention provides a method for generating a lane using a lane generation system, the step of extracting a vanishing point using an image of a road obtained from a photographing device, and both end points at the bottom of the vehicle for each vehicle appearing in the image. Estimating an area of at least one lane in the image based on a triangular area generated by connecting the vanishing point of the road and both end points of the lower part of the vehicle, and applying the estimated lane area to the image It provides a lane generation method for collecting traffic information including the step of.

Further, in the estimating step, an area of at least one lane in the image may be estimated based on a result of overlapping the triangular area for each time or for each vehicle.

In addition, in the extracting, the vanishing point may be extracted using a boundary line of a road detected in the image.

In addition, the method may further include calculating and providing an average speed for each lane based on the speeds of vehicles detected for each lane area in the image.

According to the present invention, the area of each lane can be estimated and automatically applied in the image based on the result of overlapping the area in which the vanishing point of the road and the two end points of the vehicle are connected to each other in the image of the road. By using information, traffic volume for each lane can be effectively provided.

In addition, the present invention can create a lane area only by accumulating and overlapping triangular areas collected by time for vehicles appearing in an image, so that a lane area can be automatically generated with only one vehicle, and even on a road with low traffic It is possible to create a lane area.

In addition, since the vehicle tracking process in the image is not required at all, image analysis and processing time can be reduced, and a lane area can be quickly and easily estimated by simply detecting an object (vehicle).

1 is a diagram illustrating a lane generation system for collecting traffic information according to an embodiment of the present invention.
FIG. 2 is a diagram showing a detailed configuration of the lane generation device shown in FIG. 1.
3A to 3C are diagrams illustrating a process of extracting a vanishing point using a boundary line of a road in an image in an embodiment of the present invention.
4 is a diagram illustrating a principle of detecting both end points of a lower end of a vehicle from an image in an embodiment of the present invention.
5 is a diagram showing a state in which a triangular region is generated using the result of FIG. 4.
6 is a diagram illustrating a lane area generated by accumulating the results of FIG. 5.
7 is a diagram illustrating a state in which a lane area is estimated from an image and applied to an image according to an embodiment of the present invention.
8 is a diagram illustrating a process of extracting multiple vanishing points corresponding to a curved road in an embodiment of the present invention.
9 is an exemplary diagram showing a lane area generated based on the process of FIG. 8.
10 is a diagram for explaining a method of generating a lane using the lane generation system shown in FIG. 1.

Then, 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 can easily implement the present invention.

Then, 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 can easily implement the present invention.

When a part of the specification is said to "include" a certain component, it means that other components may be further included rather than excluding other components unless otherwise stated. In addition, the term "unit" used in the specification 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 be in an addressable storage medium, or may be configured to reproduce one or more processors.

Thus, as an example, "unit" refers to components such as software components, object-oriented software components, class components, and task components, processes, functions, properties, procedures, Includes subroutines, segments of program code, drivers, firmware, microcode, circuitry, data, database, data structures, tables, arrays and variables. The functions provided within the components and "units" may be combined into a smaller number of components and "units" or may be further separated into additional components and "units".

1 is a diagram illustrating a lane generation system for collecting traffic information according to an embodiment of the present invention.

As shown in FIG. 1, a lane generation system for collecting traffic information according to an exemplary embodiment of the present invention includes a photographing device 100 and a lane generation device 200.

The photographing device 100 photographs an image of a road including a plurality of lanes, and provides the photographed image to the lane generating device 200. The photographing apparatus 100 includes a camera, photographs an image using the camera, and provides the photographed image to the lane generation apparatus 200.

The photographing apparatus 100 includes a fixed or rotating (PTZ) type camera, and may be installed and operated on a facility such as a prop on roads such as general roads, expressways, unpaved roads, and intersection points.

The photographing device 100 may be connected to the lane generating device 200 through a wired or wireless communication method to provide an image. Here, the wired and wireless communication methods include various known communication methods capable of transmitting and receiving data.

The lane generation device 200 detects the vanishing point of the road and both ends of the vehicle in the image from the image, and continuously accumulates the triangular regions connecting the vanishing points of the road and both ends of the vehicle, and estimates each lane area in the image from the result of overlapping. And, the estimated lane area can be applied to the image.

In addition, the lane generating device 200 detects a vehicle for each lane area applied in the image, and based on the detected information, traffic information by lane, driving direction, and time (e.g., vehicle speed, average speed, congestion map) , Vehicle type, traffic volume, vehicle occupancy, etc.) can be obtained and provided to the operating device 300.

In addition, the lane generating device 200 may detect a speed violation based on vehicle detection information for each vehicle detected in each lane area in the image and provide vehicle enforcement information to the operating device 300.

The operating device 300 is connected to the lane generating device 200 installed at each point of the road through a wired/wireless communication network, and collects traffic information and vehicle control information provided from each lane generating device 200, The collected traffic information and vehicle control information can be stored, managed and analyzed for each point.

In addition, the operating device 300 provides traffic information and vehicle enforcement information for each point on the road to managers and drivers through various types of information devices (eg, manager display, web service, road signboard, etc.). I can.

FIG. 2 is a diagram showing a detailed configuration of the lane generation device shown in FIG. 1.

As shown in FIG. 2, the lane generation apparatus 200 according to an embodiment of the present invention includes a vanishing point extraction unit 210, a vehicle detection unit 220, a lane estimation unit 230, and a control unit 240, A storage unit 250 may be further included.

First, the vanishing point extractor 210 extracts a vanishing point using an image of a road obtained from the photographing apparatus 100, and obtains a virtual vanishing point from a boundary line of the road based on a perspective method.

The vanishing point extraction unit 210 detects all road-related boundaries (boundaries) in the image using an image processing algorithm (e.g., stick cell technique), and extends the detected boundary lines to extract virtual vanishing points from the intersection points. I can. Here, the vanishing point of the road may be extracted in or outside the image according to the size and location of the region of image (ROI) or the installation height, angle of view, and magnification of the photographing apparatus 100.

3A to 3C are diagrams illustrating a process of extracting a vanishing point using a boundary line of a road in an image in an embodiment of the present invention.

3A shows a road boundary surface is detected by applying a stick cell technique to an image of a road, and FIG. 3B shows a detection area corresponding to the detected boundary surface.

In this way, the vanishing point extracting unit 210 may detect all boundary lines (boundaries) related to the road in the detection region based on the stick cell technique. In addition, the vanishing point extracting unit 210 may obtain a virtual vanishing point from a point crossing each other when the detected boundary lines are respectively extended, as shown in FIG. 3C.

Although the embodiment of the present invention illustrates that the stick cell technique is used, the present invention is not necessarily limited thereto, and any technique capable of detecting the boundary of a road from an image other than the stick cell technique can be applied irrespective of the technique.

Next, the vehicle detection unit 220 detects both end points of the lower end of the vehicle for each vehicle appearing in the image. Here, the vehicle detection unit 220 detects all objects related to the vehicle in the image. Then, the smallest rectangular box surrounding the vehicle exterior is allocated to the detected vehicle object, and both end points of the lower end of the vehicle are detected in the rectangular box.

4 is a diagram for explaining a principle of detecting both end points of a lower end of a vehicle from an image in an embodiment of the present invention.

As shown in FIG. 4, the vehicle detection unit 220 sets a rectangular box having a minimum size surrounding an outer edge of the object corresponding to each vehicle object, and detects both end points of the lower end of the vehicle within the rectangular box.

The lower portion of the vehicle visible in the square box substantially contacts the road surface or corresponds to the portion closest to the road surface. The vehicle detection unit 220 may detect the outer portions of both tires in contact with the road surface as both end points of the lower end of the vehicle, thereby creating a more reliable lane area.

Of course, in addition to this, the vehicle detection unit 220 may detect the speed of the vehicle by using a position change at the bottom of the vehicle detected in response to the vehicle object in the image by time. For example, by using the detection position at the bottom of the vehicle in two image frames photographed at a set time interval and the time interval of the image frame, the vehicle speed is obtained.

In this way, the vehicle detection unit 220 may generate vehicle detection information including a vehicle speed for each vehicle, and transmit the generated vehicle detection information to the control unit 240. Here, the vehicle detection information may include an image in which the vehicle is photographed, a detection speed, a detection point, and a detection time.

In addition, the lane estimating unit 230 estimates at least one lane area from the image based on a result of overlapping a triangular area generated by connecting the estimated vanishing point of the road and both end points of the detected vehicle bottom with each other. .

For example, when triangular areas acquired over time for a single vehicle are overlapped with each other, a single large lane area extending toward the vanishing point may be obtained. In addition, even for each vehicle scattered at various locations on a single lane, when all triangular regions acquired over time are overlapped, a single large lane area may be obtained.

5 is a diagram showing a state in which a triangular region is generated using the result of FIG. 4, and FIG. 6 is a diagram illustrating a lane region generated by accumulating the result of FIG. 5.

As shown in FIG. 5, the lane estimating unit 230 connects the vanishing point of the road and both end points of the lower end of the vehicle corresponding to each vehicle detected in the image to obtain a triangular area corresponding to the vehicle.

In addition, as shown in FIG. 6, the lane estimating unit 230 may estimate each lane region in the image based on the result of superimposing the triangular region generated corresponding to each vehicle in the image by time and by vehicle. .

Here, in response to a vehicle invading a lane or changing lanes, a case in which a part of a triangular area is very close to the lane or invaded the lane may occur. In order to remove noise due to such lane invasion and lane change, the lane estimating unit 230 excludes a pixel portion having an overlapping frequency (number) of less than a threshold among the pixels constituting the overlapping area. Accordingly, the lane estimating unit 230 may generate a lane region from which noise has been removed, and through this, clearly distinguish between lanes and estimate each lane region more reliably.

The control unit 240 controls the data flow of the vanishing point extraction unit 210, the vehicle detection unit 220, the lane estimation unit 230, and the storage unit 250, and can exchange data with each unit 210, 220, 230, 250. .

The control unit 240 may apply the lane area estimated by the lane estimator 230 into the image.

7 is a diagram illustrating a state in which an estimated lane area is applied in an image according to an embodiment of the present invention.

Figure 7 (a) shows a process of estimating a lane area by generating a triangular region from an image of an actual road by using the vanishing point and vehicle detection information, and Figure 7 (b) shows a process of estimating the lane area in the image. Show appearance. In this way, the control unit 240 applies each lane area estimated by the lane estimating unit 230 into the image of the photographing apparatus.

Here, the lane estimating unit 230 may update the lane region by overlapping the newly acquired triangular region in the image with the lane region previously estimated at a previous time. The lane estimating unit 230 may update at least one of a shape, an area, a length, and a width of the lane area and provide the update to the control unit 240.

Then, the controller 240 may correct and reset the lane area in the image based on the updated lane information.

In addition, the controller 240 may detect a vehicle within each lane area applied in the image and provide traffic information for each lane, driving direction, or time to the operating device 300. At this time, the traffic information may include vehicle speed, average speed, congestion degree, vehicle type, traffic volume, vehicle occupancy, and the like.

In addition, the control unit 240 can detect a speed violation based on vehicle detection information for each vehicle provided from the vehicle detection unit 220 and provide vehicle enforcement information to the operating device 300 as well as detect each lane area. By accumulating the speeds of each of the vehicles, the average speed of the road and the average speed of each lane are calculated in real time and provided to the operating device 300 so that traffic conditions such as congestion and smoothness can be checked.

The storage unit 250 may store setting information (position, size, shape in the image) of the lane area in the image, and may store traffic volume and vehicle occupancy information obtained by time in each lane area.

Meanwhile, according to an exemplary embodiment of the present invention, when the boundary line of the road is curved, the lane area may be estimated using multiple vanishing points extracted for each section of the road along the road direction.

To this end, when the boundary line of the road is curved, the vanishing point extractor 210 divides the road into a plurality of road sections and extracts vanishing points for each road section to obtain multiple vanishing points.

Specifically, the vanishing point extracting unit 210 individually extracts a vanishing point corresponding to a corresponding road section by using a virtual straight line generated based on the boundary line of the road section for each road section. Here, the virtual straight line may be generated by connecting a start point and an end point of the boundary line with a straight line for each boundary line of a corresponding road section, as shown in FIG. 8 to be described later.

8 is a diagram showing a process of extracting multiple vanishing points corresponding to a curved road in an embodiment of the present invention, and FIG. 9 shows a lane area created based on the process of FIG. 8.

In the case of (a) of FIG. 8, the road is divided into five sections (A to E), and the vanishing point is extracted corresponding to each divided section. Here, for convenience of explanation, it is assumed that vanishing points 1, 2, and 3 are respectively extracted corresponding to sections B, C, and D. As described above, by extending the virtual straight line obtained by connecting the starting point and the end point of the boundary line in a straight line in each road section, vanishing points intersecting each other can be obtained.

This method can be applied to all boundaries of roads detected in the image. That is, a large number of boundary lines are detected in a two-way multi-lane road image as shown in (a) of FIG. 8, and a greater number of vanishing points are obtained by applying the above-described principle to each of these boundary lines. Of course, in some cases, some vanishing points may be shared between both roads.

Here, the lane estimating unit 230 generates a triangular area corresponding to the vehicle by using the extracted vanishing point corresponding to the road section where the vehicle is located among the plurality of road sections. Here, the lane estimating unit 230 may select only an area on a corresponding road section from among the triangular areas as an effective area.

FIG. 8B shows that in each road section, a triangular area corresponding to a vehicle located on a corresponding section is generated by using a vanishing point corresponding to the section. Here, the lane estimating unit 230 may select a rhombic region located within a corresponding road section among the triangular regions as an effective region.

Then, the lane estimating unit 230 overlaps the triangular regions for each road section and then estimates the lane region from a result of connecting the overlapped regions secured on each road section to each other along the entire road section. For example, when the effective areas for the triangular areas collected in each road section are accumulated and overlapped with each other, and the overlapping areas acquired for each road section are connected to one along the road direction, a curved road area for a curved road is created. can do.

9 is an exemplary view showing a result of combining each lane area estimated from a curved road into an image. Here, of course, as data is accumulated over time, the lane area may be derived in a smoother curved shape.

10 is a diagram for explaining a method of generating a lane using the lane generation system shown in FIG. 1.

As shown in FIG. 10, the vanishing point extracting unit 210 of the lane generating apparatus 100 according to an exemplary embodiment of the present invention extracts a vanishing point using an image of a road obtained from the photographing device 100 (S1010). The vanishing point extractor 210 may extract the vanishing point of the road by using each boundary line of the road detected in the image.

Next, the vehicle detection unit 220 detects both end points at the bottom of the vehicle for each vehicle appearing in the image (S1020). In this case, a box having a minimum size covering the object may be set for all vehicle objects detected in the current frame, and both end points of the lower end of the vehicle may be extracted within the set box.

Here, of course, the vehicle detection unit 220 detects the speed of each vehicle based on the time position of the lower portion of the vehicle detected corresponding to each vehicle in the image, and transmits vehicle detection information including the detected speed to the control unit 240 I can.

Thereafter, the lane estimating unit 230 generates a triangular area by connecting the vanishing point of the road and both end points of the lower part of the vehicle, and then accumulates and overlaps the triangular areas to estimate the lane area in the image (S1030 and S1040). Then, the controller 240 applies the estimated lane area into the image of the photographing apparatus 100 (S1050).

The controller 240 detects a vehicle for each lane area applied in the image and provides traffic information of the road (S1060). In this case, the controller 240 may calculate and provide an average speed for each lane based on the speeds of vehicles detected for each lane area in the image.

Of course, the control unit 240 detects a speed violation based on vehicle detection information for each vehicle received from the vehicle detection unit 250, or the average speed of the road and the average of each lane based on the speed of each vehicle detected for each lane area. The speed can be calculated and provided.

Meanwhile, the lane estimating unit 230 may update at least one of the shape, area, length, and width of the lane region by overlapping the newly acquired triangular region in the image with the lane region previously generated in the previous time. Thereafter, the controller 240 may receive the updated information from the lane estimating unit 230 and set the update within the image.

According to the present invention as described above, the area of each lane on the road can be estimated and automatically set in the image based on the result of overlapping the area in which the vanishing point of the road and the two end points of the vehicle are connected to each other in the image of the road, Traffic volume for each lane can be effectively provided by using lane information applied in the image.

In addition, lane areas can be automatically created by accumulating and overlapping triangular areas collected over time for vehicles appearing in the image, so it is possible to create a lane area with only one vehicle. Can be automatically generated.

In addition, since there is no need for a process of detecting a feature of a vehicle in an image and tracking (tracking) the vehicle based on the feature, the time required for image analysis and processing can be greatly reduced.

In particular, the method of automatically generating a lane through vehicle tracking is based on a uniform input image and precise tracking, but in the case of the present invention, it is possible to automatically generate a lane by only detecting an object without tracking the vehicle, as well as the interruption of the input image. Even if it occurs, it provides the advantage of being able to create a car.

The present invention has been described with reference to the embodiments shown in the drawings, but these are only exemplary, and those of ordinary skill in the art will appreciate that various modifications and equivalent other embodiments are possible therefrom. Therefore, the true technical protection scope of the present invention should be determined by the technical spirit of the appended claims.

100: photographing device 200: car generating device
210: vanishing point extraction unit 220: vehicle detection unit
230: lane estimation unit 240: control unit
250: storage unit 300: operating device

Claims (17)

A vanishing point extracting unit for extracting a vanishing point using the image of the road obtained from the photographing device;
A vehicle detection unit for detecting both end points of a lower end of the vehicle for each vehicle appearing in the image;
A lane estimating unit for estimating at least one lane area in an image based on a triangular area generated by connecting the vanishing point of the road and both end points of the lower part of the vehicle; And
A control unit that applies the estimated lane area to the image
Lane generation system for collecting traffic information comprising a. The method according to claim 1,
The car estimation unit,
A lane generation system for collecting traffic information for estimating an area of at least one lane in the image based on a result of overlapping the triangular regions for each time or for each vehicle. The method according to claim 1,
The vanishing point extraction unit,
A lane generation system for collecting traffic information to extract the vanishing point by using the boundary line of the road detected from the image. The method of claim 3,
The vanishing point extraction unit,
A lane generation system for collecting traffic information for detecting a boundary line of the road in the image using a stick cell technique. The method according to claim 1,
The vanishing point extraction unit,
When the boundary line of the road is curved, the road is divided into a plurality of road sections, and vanishing points are extracted for each road section to obtain multiple vanishing points along the road. The method of claim 5,
The vanishing point extraction unit,
A lane generation system for collecting traffic information individually extracting a vanishing point corresponding to a corresponding road section by using a virtual straight line generated based on a boundary line of the corresponding road section for each road section. The method of claim 6,
The virtual straight line,
A lane generation system for collecting traffic information generated by connecting a start point and an end point of the boundary line with a straight line for each boundary line of the corresponding road section. The method of claim 5,
The car estimation unit,
A triangular area corresponding to the vehicle is generated by using the vanishing point of the road section where the vehicle is located among the plurality of road sections,
A lane generation system for collecting traffic information for estimating the lane area from a result of overlapping the triangular regions for each road section and connecting the overlapped regions acquired on each road section to each other along the entire road section. The method according to claim 1,
The car estimation unit,
A lane generation system for collecting traffic information, updating the lane region by overlapping the newly acquired triangular region in the image with the lane region previously generated in the previous time and providing it to the control unit. The method of claim 9,
The car estimation unit,
A lane generation system for collecting traffic information to update at least one of a shape, area, length, and width of the lane area. The method according to claim 1,
The detection unit,
A lane generation system for collecting traffic information for setting a rectangular box surrounding an outer edge of the object in response to the detected vehicle object in the image, and then detecting both end points of the lower end of the vehicle within the rectangular box. The method according to claim 1,
The vehicle detection unit,
A lane generation system for collecting traffic information for detecting a vehicle speed by using the position change of the vehicle lower part detected in response to the vehicle object in the image by time and providing vehicle detection information including the speed for each vehicle. The method according to claim 1,
The control unit,
A lane generation system for collecting traffic information for calculating and providing an average speed for each lane based on the speeds of vehicles detected for each lane area in the image. In the lane generation method using the lane generation system,
Extracting a vanishing point using the image of the road obtained from the photographing device;
Detecting both end points under the vehicle for each vehicle appearing in the image;
Estimating an area of at least one lane in the image based on a triangular area generated by connecting the vanishing point of the road and both end points of the lower part of the vehicle; And
Applying the estimated lane area to the image
A lane generation method for collecting traffic information comprising a. The method of claim 14,
The estimating step,
A lane generation method for collecting traffic information, estimating a region of at least one lane in the image based on a result of overlapping the triangular regions for each time or vehicle. The method of claim 14,
The extracting step,
A lane generation method for extracting the vanishing point using a boundary line of a road detected in the image. The method of claim 14,
And calculating and providing the average speed for each lane based on the speeds of vehicles detected for each lane area in the image.

Images