KR101055789B1 - Auto transportation information extraction system - Google Patents

Abstract

PURPOSE: An automatic traffic information extracting system is provided to efficiently extract the traffic information of a road by automatically recognizing a traffic camera area and setting a road and a lane area according to the changed area. CONSTITUTION: A traffic situation input unit(210) receives traffic images which are photographed by a traffic camera in real time. A traffic image converter(220) sets traffic images from the traffic situation input unit and converts the set traffic image into a binary images. An image change sensor(230) divides the binary image into five parts or more, stores a specific coordinate at each divided binary image, and compares the specific coordinates stored in a current frame with the specific coordinates stored at a prior frame. A road area determining unit(240) distinguishes a static image from a dynamic image in the traffic image provided from the traffic situation input unit. A traffic information extracting unit(250) extracts and transmits traffic information in real time through the static image and the dynamic image provided from a road area determining unit.

Description

Auto Transportation Information Extraction System

The present invention relates to an automatic traffic information extraction system.

The conventional traffic information extraction system has extracted traffic information through a traffic image detector, CCTV, and the like. At this time, in case of the traffic image detector, there is a problem of manually setting the road area and the traffic information detection area and resetting the camera area.

In addition, in case of image detection using CCTV, the technology is used to manually set the area manually by using the pan-tilt and lens preset functions, and to move to the preset area after monitoring the operator. In the case of the -set function, it is impossible to move accurately to a preset Pan / Tilt / Zoom / Focus value. Due to this problem, the traffic information extraction area is set incorrectly, resulting in a problem that the reliability of the traffic information is deteriorated.

An object of the present invention is to provide an automatic traffic information extraction system capable of automatically recognizing a traffic camera area change and automatically setting roads and lane areas according to the changed area, thereby extracting traffic information on the road more efficiently. have.

In an automatic traffic information extraction system according to an embodiment of the present invention, the traffic situation input unit receives a traffic image captured in real time from a traffic camera disposed on a road, sets a traffic image provided by the traffic situation input unit, and sets the traffic image. A traffic image converter for converting a binary image into a binarized image, and classify at least five binarized images provided by the traffic image converter, store specific coordinates in each of the divided binarized images, and store the specific coordinates stored in the current frame and the previous frame. In the image change detection unit comparing the specific coordinates stored in the image, the road region determination unit that distinguishes the static image from the dynamic image from the traffic image provided from the traffic situation input unit, and the image detection information and the road region determination unit provided by the image change detection unit. Traffic information is extracted in real time through the static and dynamic video provided. Traffic information extracted to be transmitted is configured by comprising: a.

The traffic information extracting unit may include extracting road information through a static image, and extracting traffic information in real time using the extracted road information and Equation (1).

In addition, the image change detector may include differently setting weights for specific coordinates in each of the divided binarized images.

In addition, the image change detection unit may include automatically re-executing the area setting when multiplying a weight set differently for each specific coordinate by a specific coordinate and then adding each value to 5 or more.

In addition, the image change detection unit may include detecting the traffic image by comparing the previous frame and the current frame every second.

The automatic traffic information extraction system according to an embodiment of the present invention automatically recognizes a traffic camera area change and automatically sets roads and lane areas according to the changed area, so that traffic information on the road can be extracted more efficiently. It works.

The automatic traffic information extraction system according to an embodiment of the present invention can automatically set a road area and a traffic information extraction area when a road image is input by a traffic camera, thereby improving the efficiency of operation of conventionally installed CCTVs and traffic cameras. It can be effected.

1 is for explaining an automatic traffic information extraction system according to an embodiment of the present invention.
2 is for explaining the operation of the image change detection unit according to an embodiment of the present invention.
3 is for explaining the operation of the traffic information extraction unit according to an embodiment of the present invention.
4 is a view for explaining a method for extracting automatic traffic information according to an embodiment of the present invention.

Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings. Specific details of other embodiments are included in the detailed description and drawings. Advantages and features of the present invention, and methods for achieving them will be apparent with reference to the embodiments described below in detail in conjunction with the accompanying drawings. Like reference numerals refer to like elements throughout.

1 is for explaining an automatic traffic information extraction system according to an embodiment of the present invention.

1, the automatic traffic information extraction system 200 according to an embodiment of the present invention is a traffic situation input unit 210, traffic image converter 220, image change detection unit 230, road area determination unit 240 ) And the traffic information extraction unit 250.

The traffic situation input unit 210 receives a traffic image captured in real time by the traffic camera 100 disposed on the road. The traffic situation input unit 210 may receive and accurately collect road traffic information such as road traffic volume, vehicle driving speed, road occupancy rate, queue length, and accident situation provided by the traffic camera 100.

The traffic image converting unit 220 sets a traffic image provided by the traffic situation input unit 210 and converts the set traffic image into a binarized image.

The image change detector 230 classifies at least five binarized images provided by the traffic image converter 220, stores specific coordinates in each of the divided binarized images, and stores the specific coordinates stored in the current frame and the previous frame. Compare specific stored coordinates. The image change detector 230 may include differently setting weights for specific coordinates in each of the divided binarized images. Detailed description thereof will be described later with reference to FIG. 2.

The road area determination unit 240 distinguishes the static image from the dynamic image from the traffic image provided by the traffic state input unit 210. The road area determination unit 240 extracts a background image that is a static image by using a difference between a previous frame and a current frame, extracts a vehicle image that is a dynamic image, and accumulates a moving vehicle flow so that a plurality of vehicles are fixed. The function of determining the area moving in the direction as the road area may be performed.

The traffic information extraction unit 250 extracts and transmits the traffic information in real time through the image detection information provided by the image change detection unit 230 and the static image and the dynamic image provided by the road area determination unit 240. . The traffic information extractor 250 may include extracting road information through a static image, and extracting traffic information in real time using the extracted road information and Equation 1. In this case, the extracted traffic information may be transmitted to the central control room 300.

&Quot; (1) "

Actual lane length =

{Vert_Lane_Length (pixel) / (High_Hori_Car_Rect (pixel) + Low_Hori_Car_Rect (pixel))) × vehicle length × 2.

Here, Equation 1 and a detailed description thereof will be described later with reference to FIG. 3.

2 is for explaining the operation of the image change detection unit according to an embodiment of the present invention, Figure 3 is for explaining the operation of the traffic information extraction unit according to an embodiment of the present invention.

2, the image change detection unit 230 according to an embodiment of the present invention shows that the binarized image is divided into five or more regions.

The image change detector 230 classifies at least five binarized images provided by the traffic image converter, stores specific coordinates in each of the divided binarized images, and stores specific coordinates stored in the current frame and specific coordinates stored in the previous frame. To compare.

That is, the image change detector 230 divides the binarized image into A, B, C, D, and E, respectively, and sets and stores specific coordinates in each of A, B, C, D, and E regions. .

In this way, the specific coordinates of the A, B, C, D, and E regions of the current frame are compared with the specific coordinates of the A, B, C, D, and E regions of the previous frame. It is detecting change.

In this case, the image change detector 230 may detect the traffic image by comparing the previous frame and the current frame every second. Accordingly, traffic information can be extracted in real time.

In addition, the image change detector 230 may include setting different weights for specific coordinates in each of the divided binarized images. That is, the weight of a specific point far from the traffic camera 100 among the areas A, B, C, D, and E is high. For example, if the binarized image is divided into areas A, B, C, D, and E, the area A is the farthest from the traffic camera, and the area E is the closest to the traffic camera. Is set to the highest value, and the weight is set to the lowest value in the E area.

Accordingly, Equation 2 below may be formed.

<Equation 2>

Auto configuration judgment value = {(a × 1.0) + (b × 0.8) + (c × 0.6) + (d × 0.4) + (e × 0.2)}

Automatic configuration judgment value> 5.

In this case, a is an object pixel shift value of region A, b is an object pixel shift value of region B, c is an object pixel shift value of region C, d is an object pixel shift value of region D, and e is an E region. The object's pixel shift value.

Accordingly, the image change detection unit 230 may include automatically re-executing the area setting when multiplying a weight set differently for each specific coordinate by a specific coordinate and then adding the respective values to 5 or more. Therefore, the automatic traffic information extraction system 200 according to an embodiment of the present invention can automatically recognize the traffic camera area, and efficiently extract traffic information from the changed area.

Referring to FIG. 3, it is shown that the actual lane length is calculated during the operation of the traffic information extracting unit 250 according to an embodiment of the present invention.

The traffic information extraction unit 250 extracts and transmits the traffic information in real time through the image detection information provided by the image change detection unit 230 and the static image and the dynamic image provided by the road area determination unit 240. . The traffic information extractor 250 may include extracting road information through a static image, and extracting traffic information in real time using the extracted road information and Equation 1.

&Quot; (1) &quot;

Actual lane length =

{Vert_Lane_Length (pixel) / (High_Hori_Car_Rect (pixel) + Low_Hori_Car_Rect (pixel))) × vehicle length × 2,

At this time, Vert_Lane_Length (pixel) = the number of pixels from the first point to the second point, and the vehicle traveling from the first point to the second point may increase constantly. High_Hori_Car_Rect (pixel) = number of horizontal component pixels of the first point, which is the length of the vehicle located at the first point. Low_Hori_Car_Rect (pixel) = number of horizontal component pixels at the second point, which is the road of the vehicle located at the second point. In addition, the vehicle length is based on a passenger car, and the base value can be calculated on the basis of being substantially 4.9.

As such, the traffic information extraction unit 250 calculates the length of the extraction region of the image detection information calculated by the image change detection unit 230 and the arrival time from the predetermined first point to the predetermined second point to drive the road. It is possible to efficiently calculate the traffic volume and occupancy rate by calculating the traffic speed of the vehicle and tracking by vehicle.

As described so far, the automatic traffic information extraction system 200 according to an embodiment of the present invention automatically recognizes a traffic camera area change, and automatically sets roads and lane areas according to the changed area, thereby providing traffic on the road. Information can be extracted more efficiently.

4 is a view for explaining a method for extracting automatic traffic information according to an embodiment of the present invention.

Referring to Figure 4, it shows a method for extracting automatic traffic information according to an embodiment of the present invention.

First, a traffic image photographed in real time of a traffic situation occurring on a road may be provided, and the traffic image may be analyzed by comparing the current frame with the previous frame every second. That is, it is determined whether the traffic image sent from the traffic camera on the road is the initial image (S100). In this case, when the traffic image is the initial image, the road area is first extracted (S130).

Subsequently, when a change is detected in the analyzed traffic image, the method may include determining a road area that is a static image of the traffic image by tracking a vehicle. That is, after the image change is detected (S110), the vehicle traveling on the road is tracked (S120). Then, the road area is detected through the traffic image (S130).

Thereafter, after determining the road area, converting the traffic image into a binary image and calculating lanes and lane lengths from the converted binary image to extract traffic information. That is, the lane is detected in the detected road area (S140), the lane length is detected through the detected lane (S150), and the traffic information is extracted by calculating the detected lane and the lane length (S160).

In this case, the method may include dividing the binarized image into at least one region, setting specific coordinates to each of the divided binarized images, and setting weights differently for specific coordinates, and assigning different weights to specific coordinates. After multiplying by adding the respective values, if the sum is 5 or more, it may include the step of automatically redoing the area setting. In addition, since the detailed description of the method for obtaining the lane length has been fully described with reference to FIGS. 1 to 3, a detailed description thereof will be omitted.

Thereafter, the image change is detected by comparing the current frame and the previous frame every second in real time (S170).

As described above, through the method of extracting the automatic traffic information according to an embodiment of the present invention, when the road image is input by the traffic camera, the road area and the traffic information extraction area may be automatically set, and through this, the conventionally installed CCTV And the efficiency of the operation of traffic cameras can be improved.

While the present invention has been particularly shown and described with reference to exemplary embodiments thereof, it is to be understood that the invention is not limited to the disclosed exemplary embodiments. Rather, those skilled in the art will appreciate that many modifications and variations of the present invention are possible without departing from the spirit and scope of the appended claims. And all such modifications and changes as fall within the scope of the present invention are therefore to be regarded as being within the scope of the present invention.

100: traffic camera 200: automatic traffic information extraction system
210: traffic situation input unit 220: traffic image conversion unit
230: image change detection unit 240: road area determination unit
250: traffic information extraction unit 300: central control room

Claims (4)

Traffic situation input unit for receiving a traffic image taken in real time from a traffic camera disposed on the road;
A traffic image converter configured to set the traffic image provided by the traffic situation input unit and convert the set traffic image into a binarized image;
The binarization image provided by the traffic image converter is classified into at least five or more, storing specific coordinates in each of the divided binarized images, and comparing the specific coordinates stored in the current frame with the specific coordinates stored in the previous frame. An image change detection unit;
A road area determination unit for distinguishing a static image from a dynamic image from the traffic image provided by the traffic situation input unit; And
A traffic information extraction unit for extracting and transmitting traffic information in real time through the image detection information provided by the image change detection unit and the static image and the dynamic image provided by the road area determination unit;
Automatic traffic information extraction system comprising a. The method according to claim 1,
The traffic information extracting unit extracts road information through the static image, and extracts the traffic information in real time using the extracted road information and Equation (1).
<Equation 1>
Actual lane length is
{Vert_Lane_Length (pixel) / (High_Hori_Car_Rect (pixel) + Low_Hori_Car_Rect (pixel))) × vehicle length × 2
At this time, Vert_Lane_Length (pixel) is the number of pixels from the first point to the second point,
High_Hori_Car_Rect (pixel) is the number of horizontal component pixels at the first point,
Low_Hori_Car_Rect (pixel) is the number of horizontal component pixels at the second point,
If the vehicle length is based on a passenger car, the basis value is 4.9. The method according to claim 1 or 2,
The image change detection unit is an automatic traffic information extraction system for setting different weights for each specific coordinate in each of the divided binarized image. The method of claim 3,
And the image change detection unit multiplies the weights set differently for each specific coordinate by the specific coordinates, and automatically redo the area setting when the sum of the respective values is 5 or more.

Images