KR101728507B1

KR101728507B1 - Device for Measuring Visibility for Fog Guardian Device

Info

Publication number: KR101728507B1
Application number: KR1020160011662A
Authority: KR
Inventors: 이상우
Original assignee: 주식회사 누리플랜
Priority date: 2016-01-29
Filing date: 2016-01-29
Publication date: 2017-04-21

Abstract

The present invention provides a visibility measurement device used for a device to prevent damage by fog which is configured to capture a movement of a vehicle driven on a road into an image, and to be able to precisely measure visibility in real time. The present invention comprises: a communicator (10) configured to transmit an image signal captured from any one of a closed-circuit television (CCTV) camera, a charge-coupled device (CCD) camera, and a complementary metal-oxide-semiconductor (CMOS) camera installed in an area anticipated to be foggy to capture an area; an image processor (20); a controller (30); a memory (40); and a display (50).

Description

Technical Field [0001] The present invention relates to a device for measuring fog,

The present invention relates to a visual distance measuring apparatus for a fog damage preventive apparatus, and more particularly, to a fog damage preventive apparatus capable of constructing a step-by-step countermeasure according to a visual distance when a fog occurs, The present invention relates to a visible range measuring device for a fog damage prevention device capable of accurately measuring a visible range in real time.

Generally, fog is a condensation phenomenon that occurs when the water vapor in the air comes into contact with a cold water surface or the ground. It causes condensation, which contains a large amount of water vapor in the air, cools the air below the dew point temperature, or promotes the formation of fine water droplets And the case where there is a large amount.

In this way, it becomes the most dangerous factor obstructing the driver's view of the motorway in the area where the fog occurs, especially the motorway in the highway. In fact, occurrence of fog causes great obstacle to safe driving, .

As a countermeasure against this, the fog lamp is mounted on the front of the automobile or the accident light is blinked to prevent the accident. However, such a preventive measure can prevent the accident caused by fog on the road, There is no problem.

Meanwhile, in order to solve such problems, a technique of dissipating fog has been developed, and as a publicly known technology, there is a patent publication No. 10-2015-0083981.

The fog dissipating device control system of Patent Publication No. 10-2015-0083981 includes a plurality of fog dissipating devices installed in each of a plurality of areas, for spraying wind generated through a blower and a hot air blower to remove fog; A plurality of photographing devices installed in each of the plurality of areas and photographing an atmospheric condition; And an image of the atmospheric condition photographed in each of the plurality of areas is analyzed to determine whether or not fog occurs in each of the plurality of areas, and when the fog occurs, the plurality of fog diffusers installed in each of the plurality of areas A plurality of control devices for driving the plurality of control devices; And a controller for controlling a plurality of mist dispersing devices installed in any one of the plurality of control devices when a user command for controlling the plurality of mist dispersing devices installed in any one of the plurality of areas is input, And a remote control device for transmitting a control command for controlling the plurality of fog dissipating devices.

With this configuration, the fog dissipating device control system can control a plurality of fog dissipating devices installed in a plurality of areas in a remote place, and therefore, convenience in managing a plurality of fog dissolving devices installed in a plurality of areas And provides an action effect that can be improved.

However, the fog dissipating apparatus control system can not accurately measure fog correction according to the degree of occurrence of fog in real time, so that countermeasures against the visible distance can not be accurately established, thereby reducing reliability.

Meanwhile, a recent embodiment for solving the above problems is known from Japanese Patent Registration No. 10-1032160, which is registered as a dedicated licensee of the present applicant.

According to the road correction measuring system and the method using the camera disclosed in the above-mentioned Japanese Patent No. 10-1032160, a communication unit for receiving a video signal from a camera installed on a road; An image processing unit for extracting a moving region of the vehicle from the received video signal and determining a visualline using the extracted moving region; And a controller for calculating a visibility corresponding to the determined line of sight using the coefficients of the corrective calculation function, the correction function being determined by performing non-linear curve fitting on the image signal, The method comprising: receiving a video signal from a camera installed on a road; Extracting a moving region of the vehicle from the received video signal; Determining a visual line using the extracted moving area; And calculating a visibility corresponding to the determined visible line by using a correction calculation function determined by performing nonlinear curve fitting on the image signal and a coefficient of the correction calculation function.

With such a configuration and operation, the visibility can be measured using a camera installed on the road, and visibility can be measured using a visual line and a road model obtained from the moving region of the vehicle In addition, even when rain or snow is detected from the video signal, it is possible to flexibly measure the visibility through image filtering and displays the measured correction result on a wired or wireless device such as a radio or a highway information board (VMS) Inducing action effect.

However, in the system and method for road correction using the camera of the above-mentioned Japanese Patent No. 10-1032160, when an object moves on a part other than the road, for example, it is mistaken as an automobile traveling on the road, It is not possible to accurately calculate the correctness and, at the same time, cause malfunction of the entire system.

Accordingly, it is an object of the present invention to provide a visual distance measuring apparatus for a fog damage prevention apparatus capable of accurately extracting only a pure moving region in which a vehicle actually moves and accurately measuring a visible distance even if an object moves in a region other than a road have.

The object of the present invention is to provide a visual distance measuring device for a fog damage prevention device for constructing countermeasures for a step-by-step countermeasure according to a visual distance when a fog occurs, comprising a CCTV camera, a CCD camera, a CMOS camera A communication unit configured to transmit a video signal photographed from any one of the plurality of video cameras; (N is a natural number) image frames input from the communication unit for a predetermined period of time to obtain difference images through a difference operation between a first image frame and subsequent successive image frames, Extracts a moving region by combining the binarized difference images, removes a region other than the road from the extracted moving region, and then performs horizontal projection and threshold automatic extraction An image processing unit configured to search for an upper portion of the moving region through an algorithm to determine a line of sight; And performing non-linear curve fitting on the video signal using the three correction baselines before operating the visual distance measuring device, so as to calculate the corrective calculation function and the visibility A control unit configured to determine a coefficient of the calculation function and calculate a correction using the correction calculation function and the coefficient of the correction calculation function; A memory unit connected to the control unit and configured to store the calculated corrective calculation function and the corrected corrective function extracted by the control unit and the calculated corrective action; And a display unit connected to the control unit and configured to display the visibility calculated by the control unit.

According to one aspect of the present invention, the image processing unit is configured to perform thresholding on each neighboring image frame pair of each of the image frames (1 to n) photographed in real time by the CCTV and transmitted to the control unit through the communication unit, ), And after holding the entire dress-holding frame again in the dress, it is actually operated to mask areas other than the valid road where the vehicles move.

According to another aspect of the present invention, the image processing unit allocates a high bit to a moving region (ROI) of the vehicle by contrasting each of the dress-held frames, and allocates a high bit to a remaining ROI region (ROI) And relatively low bits are allocated to set the moving region of the actual vehicle.

According to the present invention, even if motion of an object occurs in a region other than a road, the region is regarded as a non-interest area, and masking processing is performed to accurately measure only the pure vehicle moving region to measure a visibility distance or a visibility Of course, this provides a remarkable effect of preventing malfunction of the mist preventing device.

BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a block diagram showing a basic configuration of a visible range measuring apparatus for a fog damage prevention apparatus according to the present invention; FIG.
2 is a view showing a process of measuring a corrective distance of a visual distance measuring apparatus for a fog damage prevention apparatus according to the present invention.
Figure 3 illustrates one embodiment for extracting a moving region of a vehicle from a video signal in Figure 2;
FIG. 4 illustrates one embodiment for calculating the visibility using the corrective function in FIG. 2; FIG.
5 is a view showing one embodiment of a corrective reference line area for receiving a corrective reference line from a camera image in a visual distance measuring apparatus according to the present invention.

Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings. The present invention is an improvement on the registered patent application No. 10-1032160 registered by the present applicant as the exclusive licensee, and the registered patent will be partially cited for clarity of explanation.

1 to 5, a visible range measuring apparatus for a fog damage prevention apparatus according to an embodiment of the present invention basically comprises a communication unit 10, an image processing unit 20, a control unit 30, (40) and a display unit (50).

Here, the communication unit 10 is connected to a means for securing an image from a camera such as a CCTV camera, a CCD camera, a CMOS camera, etc. installed on a highway, a public road, an airport port, etc., and is configured to receive a captured image signal from the camera .

The communication unit 10 may also transmit the information on the calculated correction or the safe running speed information according to the correction to the traffic management center when the control unit 30 calculates the correction. Accordingly, the traffic management center notifies the driver of the occurrence of fog, information on the correction, and the safe running speed information through a wired / wireless device such as a radio and / or a highway information information board (VMS).

The control unit 30 to which the communication unit 10 is connected is connected to an image processing unit 20 configured to extract a moving area of the vehicle using the video signal received from the communication unit 10. [

The image processing unit 20 performs a difference operation between a first image frame and a remaining image frame among n image frames (n is a natural number) input from the photographing unit for a predetermined time, And binarizes the obtained difference images using a preset threshold value, and extracts the moving region by synthesizing the binarized difference images. Then, the image processing unit 20 determines a visual line using the extracted moving area. In order to determine such a visible line, the image processing unit 20 extracts, from the extracted moving area, The visual line can be determined by searching the upper portion of the moving region through a horizontal projection and a threshold automatic extraction algorithm (FIG. 4).

Here, the image processing unit 20 controls the processing capacity of the image signal so that an excessive load is not generated during data processing due to a large-capacity image signal captured and transmitted in real time.

More preferably, the image processing unit 20 is configured to calculate a data processing speed of the video signal so that the video signal has a processing capacity suitable for the data maximum processing speed calculated by the data processing speed calculating program, (For example, 3-5 frames which is the sharpest among 20 frames per second)).

Here, the frame editing module divides the video signal photographed in real time into respective frames (for example, 20 frames per second), extracts branching, bending, curvature and shading lines of the frames with respect to the frames, (E.g., 3 to 5 frames per second) of the clearest frames among the frames that have not been deleted, and a frame editing process Algorithm.

Accordingly, since the number of frames of the video signal is reduced by the frame editing module, the processing capacity of the video signal to be processed is reduced, and overload that may occur when the video signal having a large capacity is transmitted in real time is prevented, Measurement can be made possible.

The controller 30 connected to the communication unit 10 and the image processing unit 20 is configured to calculate a visibility corresponding to the determined line of sight using the corrective calculation function.

The controller 30 is configured to calculate a visibility corresponding to the determined line of sight using the corrective calculation function.

More specifically, the control unit 30 performs non-linear curve fitting on the image signal using three correction baselines before operating the visual distance measuring apparatus according to the present invention And the coefficients of the corrective calculation function and the corrective calculation function. For example, the nonlinear curve fitting is a method of determining a coefficient of a function so as to have a minimum error in a specific function using a given data, wherein the three correction baselines are one horizontal line of the lower end region of the image corresponding to the video signal A second visual reference line sl indicating one horizontal line of the middle area of the image and a third visual reference line sl indicating one horizontal line of the upper part of the image, tl) (see Fig. 5).

The control unit 30 can output the calculated corrective information so that the communication unit 10 transmits the outputted corrective information to the traffic management center. In the traffic management center, And informs the driver through a wired / wireless device such as a radio and / or a highway information sign (VMS).

In addition, the controller 30 may detect weather information from the video signal using a predetermined weather information detection algorithm, and determine whether the road is snowy or rainy by using the detected weather information. As a result of the determination, if the road is snowing or raining, the image processing unit 20 may extract the moving region after filtering the snow or rain in the image signal. Here, the weather information detection algorithm may detect weather information using an image obtained through the camera. The controller 30 may use the weather information detection algorithm based on the average RGB value of a clear day, Weather information can be detected. Since the weather information detection algorithm is well known to those skilled in the art, a detailed description thereof will be omitted for clarity of explanation.

The memory unit 40 connected to the control unit 30 is configured to store the corrective and temporary image data calculated by the control unit 30.

The memory unit 40 may store the corrective calculation function determined by the controller 30 and the coefficient of the corrective calculation function.

The display unit 50 is connected to the control unit 30 and displays the calculated correction.

In particular, according to one of the main features of the present invention, the image processing unit 20 preferably has a masking function for acquiring or extracting only a pure road on which a vehicle moves, that is, a pure vehicle moving region.

More specifically, when a video signal photographed by a photographing means such as a CCD camera is transmitted to the control unit 30 through the communication unit 10, each pair of video frames adjacent to each video frame 1 to n (N-1) dress holding frames, and masking areas other than the effective road where the vehicles actually move.

As a result, the image frame is dressed and masked by the image processing unit 20, so that even if an object moving or moving from a portion other than the pure vehicle moving region (pure road) is moved or moved, It can be confirmed and set correctly.

According to another aspect of the present invention, the image processing unit 20 allocates a high bit to a specific region of interest, that is, an effective road on which the vehicle actually travels, in preparation for all of the extracted difference images df1 to dfn And allocates a relatively low bit to the remaining non-interest areas. It is preferable that the sections of the ROI and the ROI are executed before the masking process or simultaneously with the masking process.

In other words, when all of the vehicle images df1 to dfn of the image processing unit 20 are contrasted, the effective road on which the vehicle actually travels increases in area depending on the movement of the traveling vehicle, Since the mountains, rivers, and fields surrounding the roads other than the roads are relatively small in area to be changed by the moving object, And a weighting module for allowing different weights to be assigned.

Here, the weighting module sequentially compares the frames of the video signal, extracts only the change region by irradiating the entire region of the frame, preprocesses and binarizes the change region, removes the hash from each frame, So that each frame is divided into a group having a large change area and a group having a little change area. At this time, a high bit is assigned to a redundant area of a group having a large change area, Only a frame having a capacity equal to or larger than a predetermined size can be selected and subjected to a difference operation so that the extraction of the mobile coincidence can be performed quickly so that an overall data processing speed Can be improved.

For example, if a riverside road is located along a river with a width similar to the width of the road, a lot of mist occurs more frequently on the road than on the river. In such a situation, if the river other than the road is recognized as an effective road, . Thus, when the road and the road periphery are divided into the ROI and the ROI by the weighting module as described above and only the weights are assigned thereto, only the pure vehicle movement region can be extracted, It is possible to prevent the measurement of the correcting distance due to a moving object such as a cruise ship or the like.

Accordingly, only the pure road area where the vehicle moves is extracted to the moving area of the vehicle, so that even if there is an object moving in the area other than the moving area, the measurement of the visual distance measuring device is prevented, It is possible to prevent a malfunction.

The process of measuring the visible range by the visible range measuring device for the fog damage prevention device constructed as described above is as follows.

2, the first image frame photographed by the CCD camera A provided at a predetermined position where the road or the road can be viewed is transmitted to the control unit 30 through the communication unit 10 . The original image frame thus transmitted is initialized to a road model by the operator B or engineer (C). The initialization of such a load model determines variables that can be utilized for corrective measurement, for example, by separating the road area and the off-road area, setting the reference line, installing the CCD camera, tilting the road, and angle of the CCD camera . Although it is difficult to accurately calculate the distance from the 2D image outputted from the CCTV camera to the specific target without error, the load model can be determined in a manner known in, for example, Japanese Patent No. 10-1032160 (D).

When the image E photographed from the CCD camera A is transmitted to the image processing unit 20 through the communication unit 10, the image processing unit 20 collates or contrasts the respective image images, (G) the final movement region is determined (H) by extracting the movement region using the generated image.

3, when n image frames f1 to fn are input to the image processing unit 20 through the communication unit 10 after being photographed for a predetermined time from the CCD camera A, The second image frame f2 and the third image frame f3, the third image frame f3 and the fourth image frame f4 of the first image frame f1 and the second image frame f2 of the first image frame f1, , .......... Finally, thresholding the n-1th image frame fn-1 and the nth image frame fn successively to generate n-1 difference images df1- dfn).

When the difference images df1 to dfn are extracted as described above, all of the difference images df1 to dfn are compared with each other to allocate a high bit to a specific region of interest, that is, an effective road actually traveled by the vehicle, And allocates relatively low bits to set an actual vehicle movement area. At the same time, a portion other than the region of interest, i.e., the pure motion region of the vehicle, is masked to finally extract only the pure vehicle motion region.

When the pure vehicle moving region is determined as described above, the visible line is detected using the vehicle moving region image frame (I), and the detected line of sight is determined. That is, an upper part of the moving image frame is detected through the horizontal projection and the threshold automatic extraction algorithm on the moving area image frame to determine the line of sight (J).

When the line of sight is determined as described above, a plurality of correction baselines are compared with or synthesized with a set load model, nonlinear curve joining is performed to determine a corrective calculation function, and where the line of sight is located is extracted, It is possible to calculate the visibility (K).

As a result, according to the main feature of the present invention, in the process of comparing and combining the difference images obtained from the plurality of image frames photographed by the photographing unit, the distinction processing and the masking process of the region of interest and the region of non-interest are added, If there is a moving object in an area other than the pure moving area, it is possible to measure the accurate viewing distance by removing it.

Hereinafter, a visible range measuring method using the visible range measuring apparatus for a fog damage prevention apparatus will be described in detail.

First, the macroscopic distance measuring apparatus for a fog damage prevention apparatus according to the present invention determines whether a corrective reference line is set in the controller 30.

When the corrective reference line is set, an image frame photographed from the CCD camera installed on the road is received through the communication unit 10. [ However, if the correction reference line is not set, the control unit 30 receives the image signal through the communication unit 10 and then outputs the first correction reference line, the second correction reference line, and the third correction reference line in the control unit 30 and the image processing unit 20, After setting the correction baseline, nonlinear curve fitting is performed using each correction baseline, and the coefficients of the correction calculation function and the correction calculation function are determined and stored as a correction reference line.

Then, the control unit 30 determines whether the image frame has been received for a predetermined time or unit time.

When the unit time is reached or elapsed, the weather information is detected through the received image signal. If the unit time does not elapse, the image frame is repeatedly received until the unit time elapses.

On the other hand, when the unit time has elapsed, the control unit 30 determines whether snow or rain is on the road by using weather information.

The eye or rain is filtered from the received video signal when an eye or a rain is present, and a moving region of the vehicle is extracted from the received video signal if no eye or rain is detected. More specifically, when n image frames (f1 to fn) captured by the CCD camera for a predetermined time and input to the image processing unit 20 through the communication unit 10 are input, the first image frame f1 and the first image frame Next, n-1 difference images df1 to dfn are extracted by successively dressing and holding the final image frame fn through repeated image frames. Next, all of the extracted difference images df1 to dfn are compared and synthesized, and then a high bit is assigned to a specific region of interest and a relatively low bit is assigned to the remaining regions of interest to divide and set an actual region of motion . Next, the portion excluding the moving region of the vehicle is masked to extract the final moving region of the vehicle.

When the final vehicle movement area is determined, the upper part of the movement area is searched to determine the line of sight using the horizontal projection and the threshold automatic extraction algorithm.

Next, the control unit 30 measures the visibility corresponding to the line of sight using the corrective calculation function.

Finally, the measured correction is displayed on the display unit 50 and stored in the memory unit 40.

The visual distance measuring apparatus constructed and executed as described above is not limited but can be applied to a mist preventing apparatus to accurately measure a visible distance in a mist generating region and to warn and propagate the visible distance to prevent damage due to mist.

In addition, in the process of comparing and combining the difference images obtained from the plurality of image frames photographed by the photographing means, even if there is a moving object in an area other than the pure moving area of the vehicle, Measurement and malfunction are prevented, and the mist preventing device is operated only when the mist is generated, so that the reliability and the productability can be improved.

Claims

A visual distance measuring device for a fog damage prevention device for establishing a stepwise countermeasure according to a visible distance when a fog occurs,
A communication unit 10 configured to transmit a video signal photographed from any one of a CCTV camera, a CCD camera, and a CMOS camera installed in a fog occurrence expected area and photographing the corresponding area;
Difference images are obtained through a difference operation between an original image frame and subsequent consecutive image frames among n (n is a natural number) image frames input from the communication unit 10 for a predetermined period of time A binarization unit that binarizes the obtained difference images using a preset threshold value, extracts a moving region by combining the binarized difference images, removes a region other than the road from the extracted moving region, An image processing unit (20) configured to search a top portion of the moving region through a threshold value automatic extraction algorithm to determine a line of sight;
And performing non-linear curve fitting on the image signal using three correction baselines before the visual distance measuring apparatus is operated, A control unit (30) configured to determine a corrective calculation function and a coefficient of the corrective calculation function, and calculate a correction using the corrective calculation function and the corrective calculation function;
A memory unit (40) connected to the control unit (30) and configured to store the calculated corrective calculation function and the corrective correction function extracted by the control unit (30) and the calculated corrective system; And
And a display unit (50) connected to the control unit (30) and configured to display a visibility calculated by the control unit (30)
The image processing unit (20)
(For example, the most clear of the 20 frames per second), in order that the image signal has the processing capacity suitable for the data maximum processing speed calculated by the data processing speed calculating program 3 to 5 frames), and the frame editing module includes:
Wherein the frame editing module comprises:
A video signal captured in real time is divided into respective frames (for example, 20 frames per second), and branching, bending, curving, and shading lines of the shape are extracted from the frames, and the branching, bending, And a frame editing algorithm for deleting a frame from which the hatched line is extracted and selectively saving some frames (e.g., 3-5 frames per second) of the clearest frames among the frames that have not been deleted,
The image processing unit (20)
Each neighboring pair of image frames of each video frame 1 to n transmitted to the control unit 30 through the communication unit 10 is thresholded and each of the dress- After the holding, is actually operated to mask an area other than the valid road where the vehicles move,
A high bit is assigned to a moving region (a region of interest) of the vehicle in comparison with each dress-held frame, and a relatively low bit is allocated to the remaining region of the moving region (non-interest region) And is configured to compartmentalize and set,
An actual vehicle movement region is set in such a manner that a high bit is assigned to a specific region of interest and a relatively low bit is allocated to remaining regions of interest other than the extracted difference images df1 to dfn ,
And a weighting module for dividing the portion where the image is largely changed and the portion where the image is largely changed into a region of interest and a region of non-interest, to which different weights are given, in contrast to all of the difference images df1 to dfn,
The weighting module,
The video signal frames are sequentially compared and the entire region of the frame is examined to extract only the change region, and the pre-processing and binarization are performed on the change region, and then the neuron is removed from each frame and the changed image is stored through the region analysis In this case, a high bit is assigned to a redundant area of a group having a large change area, and a low bit is assigned to a redundant area of a group having a small change area. , And extracting only the frames having a capacity equal to or larger than a predetermined size during the extraction of the moving region so as to carry out a difference operation processing for the fog damage prevention apparatus.

delete