KR101032160B1 - System and method for road visibility measurement using camera - Google Patents

Abstract

Road visibility measurement system using a camera according to an embodiment of the present invention includes a communication unit for receiving an image signal from the camera installed on the road and transmits the result; An image processor extracting a moving area of the vehicle from the received image signal and determining a visual line using the extracted moving area; And a control unit that calculates visibility corresponding to the determined line of sight using a visibility calculation function.

Camera, video signal, visibility measurement, moving area, line of sight, fog alarm, road

Description

Road visibility measurement system using camera and its method {SYSTEM AND METHOD FOR ROAD VISIBILITY MEASUREMENT USING CAMERA}

Embodiments of the present invention relate to a system and method for measuring visibility using a camera on a roadway.

The fog alarm system transmits this fact to the traffic management center through the fog sensor installed on the highway when the fog occurs, and the traffic management center informs the driver through wired / wireless devices such as radios and highway information signs (VMS). to be. When the fog alarm system is installed on the road, there is an effect of reducing traffic accidents such as chain collisions that frequently occur when fog occurs.

Fog sensors that can be used in the fog alarm system (fog sensor) is currently a wide variety of sensors have been developed. However, currently commercially available sensors are mainly light-scatter type optical sensors, and these sensors have a high possibility of serious error under non-uniform conditions such as snow, rain, and fog density changes. Large differences can occur. Therefore, there is a need for the development of an inexpensive visibility measurement method that is most similar to human visibility and capable of global visibility measurement.

Visibility measurements using camera images have been studied by many researchers in terms of overcoming the limitations of optical sensors and resembling human vision. However, since visibility measurement using a camera image is difficult to measure at night, a separate auxiliary device or target, such as a region of interest (ROI), must be installed.

However, the installation of a separate ROI or target is expensive to install, and it is difficult to calculate the visibility when the day and night division is ambiguous, such as dawn. Therefore, it is necessary to develop a new visibility measurement system that can measure visibility on the road by using a camera installed on the road without using a target such as ROI.

One embodiment of the present invention provides a visibility measurement system and a method for measuring visibility using a camera installed on a road without using a target such as ROI.

One embodiment of the present invention provides a visibility measurement system and method capable of measuring visibility using a visual line and a road model obtained from a moving area of a vehicle.

An embodiment of the present invention provides a visibility measurement system and method that can be measured flexibly through image filtering even when rain or snow is detected from the image signal.

An embodiment of the present invention provides a visibility measurement system and a method for inducing a driver to drive safely by displaying the measured visibility result on a wired / wireless device such as a radio or a highway information guide board (VMS).

The problem to be solved by the present invention is not limited to the above-mentioned problem, another task (s) not mentioned will be clearly understood by those skilled in the art from the following description.

A visibility measurement system using a camera according to an embodiment of the present invention includes a communication unit for receiving an image signal from a camera installed on a road and transmitting a result; An image processor extracting a moving area of the vehicle from the received image signal and determining a visual line using the extracted moving area; And a control unit that calculates visibility corresponding to the determined line of sight using a visibility calculation function.

A visibility measurement method using a camera according to an embodiment of the present invention comprises the steps of receiving an image signal from a camera installed on the road; Extracting a moving area of the vehicle from the received image signal; Determining a visual line by using the extracted moving area; And calculating a visibility corresponding to the determined line of sight using a visibility calculation function.

Specific details of other embodiments are included in the detailed description and the accompanying drawings.

Advantages and / or features of the present invention and methods for achieving them will become apparent with reference to the embodiments described below in detail in conjunction with the accompanying drawings. However, the present invention is not limited to the embodiments disclosed below, but will be implemented in various different forms, and only the embodiments make the disclosure of the present invention complete, and those skilled in the art to which the present invention pertains. It is provided to fully inform the person having the scope of the invention, which is defined only by the scope of the claims. Like reference numerals refer to like elements throughout.

According to an embodiment of the present invention, the visibility may be measured by using a camera installed on a road without using a target such as an ROI.

According to an embodiment of the present invention, the visibility may be measured using a visual line and a road model obtained from the moving area of the vehicle.

According to an embodiment of the present invention, even when rain or snow is detected from an image signal, visibility can be measured flexibly through image filtering.

According to an exemplary embodiment of the present invention, the measured driving result may be displayed on a wired / wireless device such as a radio or a highway information guide board (VMS) to induce the driver to drive safely.

A visibility measurement system using a camera according to an embodiment of the present invention can be divided into an initialization step and an operation step to measure visibility. That is, in the initializing step, the operator constructs a road model using basic data such as the installation height of the CCTV camera, the inclination of the road, the angle of the camera, and the measurement data. Just do it. The operation step is a step of measuring visibility using a visual line obtained from the moving area of the vehicle and the road model constructed in the initialization step.

Hereinafter, with reference to the accompanying drawings will be described embodiments of the present invention;

1 is a block diagram illustrating the configuration of a visibility measurement system using a camera according to an embodiment of the present invention.

Referring to FIG. 1, a visibility measurement system 100 using a camera according to an exemplary embodiment of the present invention may include a communication unit 110, an image processing unit 120, a control unit 130, a memory unit 140, and a display unit 150. ) May be included.

The communication unit 110 receives an image signal from a camera installed on the road. Here, the road may be interpreted to encompass all roads in which vehicles operate, such as highways and general roads. In addition, the camera may be interpreted to encompass all devices for capturing an image, such as a CCTV camera, a CCD camera, a CMOS camera, and the like.

When the visibility is calculated by the controller 130, the communication unit 110 may transmit information about the calculated visibility or safety driving speed information according to the correction to the traffic management center. Accordingly, the traffic management center notifies the driver of the fact that the fog is generated, information on the visibility, safety driving speed information, and the like through a wired / wireless device such as a radio and / or a highway information guide board (VMS).

The image processor 120 extracts a moving area of the vehicle from the received image signal.

To this end, the image processing unit 120 performs a difference image through a difference operation between the first image frame and the remaining image frames among n (n is a natural number) image frames input from the camera for a predetermined time. image). The image processor 120 may extract the moving area by binarizing the obtained difference images using a preset threshold and synthesizing the binarized difference images.

The image processor 120 determines a visual line using the extracted moving area.

To this end, the image processor 120 removes an area other than the road from the extracted moving area, and then determines the line of sight by searching for an upper portion of the moving area through a horizontal projection and a threshold automatic extraction algorithm. Can be.

The controller 130 calculates visibility corresponding to the determined line of sight using a visibility calculation function.

To this end, the control unit 130 calculates the correction by performing non-linear curve fitting on the image signal using three correction reference lines before operating the visibility measurement system 100. A coefficient of the function and the visibility calculation function can be determined.

For reference, the nonlinear curve fitting is a method of determining a coefficient of a function to have a minimum error in a specific function by using some given data.

Here, the three visibility reference lines may include a first visibility reference line representing a horizontal line of one lower end region of an image corresponding to the image signal, a second visibility reference line representing a horizontal line of one middle region of the image, and It points to the 3rd visibility baseline which means the horizontal line of one of the upper area | regions of an image (refer FIG. 8).

In addition, the controller 130 may output the calculated visibility information. Accordingly, the communication unit 110 transmits the output visibility information to the traffic management center, and the traffic management center transmits safety driving speed information through a wired / wireless device such as the radio and / or a highway information guide board (VMS). Inform the driver.

Meanwhile, the controller 130 may detect weather information from the image signal using a predetermined weather information detection algorithm, and determine whether snow or rain falls on the road using the detected weather information. As a result of the determination, when snow or rain falls on the road, the image processor 120 may extract the moving area after filtering snow or rain from the image signal.

Here, the weather information detection algorithm may detect weather information using an image obtained through the camera, the controller 130, for example, a weather information detection algorithm based on the RGB average value of the image of a sunny day, etc. The weather information may be detected using. Since the weather information detection algorithm corresponds to a known technology, a detailed description thereof will be omitted.

The memory unit 140 stores the visibility and the temporary image data calculated by the controller 130. In addition, the memory unit 140 stores the visibility calculation function determined by the controller 130 and the coefficients of the visibility calculation function.

The display unit 150 displays the visibility calculated by the controller 130.

2 is a diagram illustrating an example of extracting a moving region of a vehicle from an image signal according to an embodiment of the present invention.

In general, an object moving on a road in an environment such as a highway makes it difficult to calculate image contrast. On the contrary, the moving region obtained from the image is different from the case where the fog occurs and the case where there is no fog, and the moving region of the object also varies according to the degree of fog.

Hereinafter, an example of extracting a moving area of the vehicle based on the basic idea will be described in detail.

That is, in the visibility measurement system using a camera according to an embodiment of the present invention, n (natural number) input for a predetermined time from a CCTV camera installed on a road, as shown in FIG. 2, to extract a moving area of a vehicle. Of the two image frames 210 and 215, a difference operation 215 between the first image frame 210 and the remaining image frames 220 is used.

Difference images between the input first image frame 210 and the remaining image frame 220 are binarized using a given threshold, and the binarized difference images 230 are sequentially synthesized 235. The synthesized resultant image may be defined as the moving area 240.

3 is a diagram illustrating an example of determining a visible line using a moving area according to an embodiment of the present invention.

A method of determining the visible line using the moving area 240 that is the resultant image of FIG. 2 is as follows. That is, as shown in FIG. 3, the visible line 310 removes the off-road area from the moving area 240, and then, through the horizontal projection and the threshold automatic extraction algorithm, It can be easily detected by finding the top part.

4 is a diagram illustrating an example of a road model applied to an embodiment of the present invention.

The road model refers to the determination of parameters that can be used for visibility measurement using separation of road and off-road areas, visibility baseline, installation height of CCTV camera, road inclination and angle of CCTV camera.

In general, it is almost impossible to accurately calculate the distance from a 2D image output from a CCTV camera to a specific target. However, visibility measurements on roads do not require the calculation of accurate distances up to several centimeters, taking into account the use for fog alarms in areas of fog. Therefore, in this embodiment, the visibility is measured using the road model as shown in FIG.

A CCD camera is provided at a certain height h on the surface of the road, and the angle θ and the height C _{h of the} CCD at which the camera is installed can be known in advance. Then, the maximum value φ _max of the camera view angle in the road model of FIG. 4 may be calculated by using the focal length f of the camera as shown in Equation 1 below, and at the bottom of the image The distances to the middle, and the top may be calculated as shown in Equations 2, 3 and 4, respectively.

5 is a diagram illustrating an example of calculating visibility using a visibility calculation function according to an embodiment of the present invention.

Referring to FIG. 5, when the distances d _b, d _{m, and} d _u to each position on the image 510, that is, the bottom, the middle, and the top, are calculated, a distance d _v to a specific target may be obtained. Usually, since the distance between d _b and d _u increases exponentially, in this embodiment, by performing non-linear curve fitting using d _b, d _{m, and} d _u , the correction calculation function 500 is performed. After determining, simply by finding a visual line, the correction for the given height of the visible line may be calculated using the determined correction function 500.

However, since the road model assumes that the road is flat and the top end of the image is the end of the road, it can be used only as a reference for defining the first visibility reference line, the second visibility reference line, and the third visibility reference line. In this embodiment, d _b is regarded as the first visibility reference line _, d _m is the second visibility reference line _{, and} d _u is the third visibility reference line, and each of the visibility reference lines d _b, d _{m, and} d _u is the road model and the actual measurement. The operator can decide once.

6 and 7 are flowcharts illustrating a visibility measurement method using a camera according to an embodiment of the present invention. Here, the visibility measurement method may be performed by the visibility measurement system 100 of FIG. 1.

6 and 7, in step 610, the visibility measurement system determines whether a visibility reference line is set.

As a result of the determination, when the visibility reference line is set (YES direction 610), in step 620, the visibility measurement system receives an image signal from a camera installed on the road.

Next, in step 630, the visibility measurement system determines whether the unit time has elapsed.

As a result of the determination, when the unit time has elapsed (YES direction 630), in step 640, the visibility measurement system detects weather information through the image signal. On the other hand, if the unit time has not elapsed (“No” direction 630), the visibility measurement system performs the step 620 until the unit time elapses.

Next, in step 650, the visibility measurement system determines whether snow or rain falls on the road using the sensed weather information.

As a result of the determination, when snow or rain comes (YES direction 650), in step 660, the visibility measurement system filters snow or rain on the image signal. On the other hand, if it is determined that no snow or rain is present (No direction 650), the visibility measurement system skips the filtering process of step 660.

Next, in step 670, the visibility measurement system extracts a moving area of the vehicle from the image signal.

To this end, the visibility measurement system first performs a difference image through a difference operation between the first image frame and the remaining image frames among n (n is a natural number) image frames inputted from the camera for a predetermined time. ) To obtain them. Subsequently, the visibility measurement system binarizes the acquired difference images using a preset threshold, and then synthesizes the binarized difference images to extract the moving area.

Next, in step 680 the visibility measurement system determines a visual line using the extracted moving area.

To this end, the visibility measurement system removes an area other than the road from the extracted moving area, and then determines the line of sight by searching the upper portion of the moving area through a horizontal projection and a threshold automatic extraction algorithm.

Next, at step 690, the visibility measurement system calculates visibility corresponding to the determined line of sight using the visibility calculation function. In this case, the visibility calculation function is determined in advance in step 610 that the reference line is already set. Thus, the visibility measurement system can calculate the visibility corresponding to the visible line using a predetermined visibility calculation function.

Next, at step 695 the visibility measurement system displays and stores the measured visibility.

On the other hand, if it is determined in step 610 whether the correcting reference line is set, and when the correcting reference line is not set (No direction in 601), the corrective measurement system performs steps 710 to 7 of FIG. Perform 770.

That is, in step 710, the visibility measurement system receives an image signal from a camera installed on a road.

Next, the visibility measurement system receives the first visibility reference line in step 720, receives the second visibility reference line in step 730, and then receives the third visibility reference line in step 740. .

Referring to FIG. 8, the visibility measurement system receives a horizontal line of one of the first visibility reference line regions 810, which is a lower region of the image 800, as the first visibility reference line, and the image 800. A horizontal line of one of the second corrective reference line regions 820, which is an intermediate region of the input line, is received as the second corrective reference line, and a horizontal line of one of the third corrective reference line regions 830, which is an upper region of the image 800, is received. 3 Input to corrective baseline.

For reference, FIG. 8 illustrates an example of a visibility reference line region for receiving a correction reference line in an image according to an embodiment of the present invention.

Next, in step 750, the visibility measurement system performs a non-linear curve fitting using the input first to third visibility reference lines, and then in step 760, the visibility calculation function. And determine a coefficient of the visibility calculation function.

Next, in step 770, the determined visibility calculation function and the coefficient are stored.

If steps 710 to 770 are performed once before operating the visibility measurement system, they do not have to be performed thereafter. In other words, the visibility measurement system continues to use the initially determined visibility calculation function and coefficients when measuring visibility.

As described above, according to the exemplary embodiment of the present invention, the visibility may be measured by using a camera installed on the road without using a target such as an ROI. In addition, according to an exemplary embodiment of the present invention, visibility may be measured using a visual line and a road model obtained from a moving area of the vehicle.

Furthermore, according to an embodiment of the present invention, even when rain or snow is detected from the image signal, the visibility can be measured flexibly through image filtering, and the measured results of the correction can be obtained from a wired / wireless device such as a radio or a highway information guide board ( VMS) can be used to induce the driver to drive safely.

Embodiments of the present invention include computer readable media including program instructions for performing various computer implemented operations. The computer readable medium may include program instructions, local data files, local data structures, or the like, alone or in combination. The media may be those specially designed and constructed for the purposes of the present invention, or they may be of the kind well-known and available to those having skill in the computer software arts. Examples of computer-readable recording media include magnetic media such as hard disks, floppy disks, and magnetic tape, optical recording media such as CD-ROMs, DVDs, magnetic-optical media such as floppy disks, and ROM, RAM, flash memory, and the like. Hardware devices specifically configured to store and execute the same program instructions are included. Examples of program instructions include not only machine code generated by a compiler, but also high-level language code that can be executed by a computer using an interpreter or the like.

While specific embodiments of the present invention have been described so far, various modifications are possible without departing from the scope of the present invention. Therefore, the scope of the present invention should not be limited to the described embodiments, but should be determined not only by the claims below, but also by the equivalents of the claims.

As described above, the present invention has been described by way of limited embodiments and drawings, but the present invention is not limited to the above-described embodiments, which can be variously modified and modified by those skilled in the art to which the present invention pertains. Modifications are possible. Accordingly, the spirit of the present invention should be understood only by the claims set forth below, and all equivalent or equivalent modifications thereof will belong to the scope of the present invention.

1 is a block diagram illustrating the configuration of a visibility measurement system using a camera according to an embodiment of the present invention.

2 is a diagram illustrating an example of extracting a moving region of a vehicle from an image signal according to an embodiment of the present invention.

3 is a diagram illustrating an example of determining a visible line using a moving area according to an embodiment of the present invention.

4 is a diagram illustrating an example of a road model applied to an embodiment of the present invention.

5 is a diagram illustrating an example of calculating visibility using a visibility calculation function according to an embodiment of the present invention.

6 and 7 are flowcharts illustrating a visibility measurement method using a camera according to an embodiment of the present invention.

8 is a diagram illustrating an example of a visibility reference line region for receiving a correction reference line in an image according to an embodiment of the present invention.

<Explanation of symbols for the main parts of the drawings>

110: communication unit 120: image processing unit

130: control unit 140: memory unit

150: display unit

Claims (13)

A communication unit receiving an image signal from a camera installed on a road; An image processor extracting a moving area of the vehicle from the received image signal and determining a visual line using the extracted moving area; And A control unit that calculates visibility corresponding to the determined visible line using a correction calculation function determined by performing non-linear curve fitting on the image signal, and coefficients of the correction calculation function. Visibility measurement system using a camera comprising a. The method of claim 1, The image processor Among the n image frames input for the predetermined time from the camera, n is a natural number, and obtains difference images through a difference operation between the first image frame and the remaining image frames, and sets a preset threshold value. Binarizing the acquired difference images, and synthesizing the binarized difference images to extract the moving area. The method of claim 1, The image processor The visibility measurement system using a camera, characterized in that for determining the line of sight by searching the upper portion of the moving area through a horizontal projection and a threshold automatic extraction algorithm for the moving area. The method of claim 1, The nonlinear curve fit is And a coefficient of the visibility calculation function and the visibility calculation function is determined for the video signal using three or more visibility reference lines. The method of claim 1, A memory unit for storing coefficients of the visibility calculation function and the visibility calculation function More, The control unit Before operating the visibility measurement system, a coefficient of measurement of the visibility calculation function and the visibility calculation function is determined and stored in the memory unit. The method of claim 4, wherein The visibility baseline is A first visibility reference line representing a horizontal line of one of the lower regions of the image corresponding to the image signal, a second visibility reference line representing a horizontal line of one of the middle regions of the image, and a horizontal line of one upper region of the image A visibility measurement system using a camera, characterized in that it comprises a third visibility reference line. The method of claim 1, The control unit Detecting weather information from the video signal using a predetermined weather information detection algorithm, and determining whether snow or rain falls on the road using the detected weather information, The image processor And, when the snow or rain falls on the road as a result of the determination, filtering the image corresponding to the snow or rain from the image signal, and then filtering and extracting the moving area. The method of claim 1, The communication unit Determining the safety driving speed information according to the calculated visibility, and transmitting at least one of the calculated visibility or the determined safety driving speed information to a traffic management center. Receiving an image signal from a camera installed on a road; Extracting a moving area of the vehicle from the received image signal; Determining a visual line by using the extracted moving area; And Calculating a visibility corresponding to the determined visible line by using a correction calculation function determined by performing a nonlinear curve fitting on the image signal and a coefficient of the correction calculation function Visibility measurement method using a camera comprising a. 10. The method of claim 9, Extracting the moving area Acquiring difference images from a difference operation between an initial image frame and a remaining image frame among n image frames inputted from the camera for a predetermined time; Binarizing the obtained difference images using a preset threshold; And Extracting the moving area by synthesizing the binarized difference images Visibility measurement method using a camera comprising a. 10. The method of claim 9, Determining the line of sight Determining the line of sight by searching an upper portion of the moving area through a horizontal projection and a threshold automatic extraction algorithm for the moving area. Visibility measurement method using a camera comprising a. 10. The method of claim 9, The nonlinear curve fit is And a coefficient of the visibility calculation function and the visibility calculation function is determined with respect to the image signal using three or more visibility reference lines. 10. The method of claim 9, Detecting weather information from the video signal using a predetermined weather information detection algorithm between receiving the video signal and extracting the moving area; Determining whether snow or rain falls on the road using the detected weather information; And If it is determined that snow or rain falls on the road, removing and filtering an image corresponding to snow or rain from the video signal Visibility measurement method using a camera, characterized in that it further comprises.

Images