KR102422235B1 - System for intelligently notifying traffic jam in tunnel - Google Patents

Abstract

An intelligent vehicle congestion notification system in the tunnel is provided. According to the embodiments of the present invention, traffic conditions in the tunnel can be intuitively displayed according to the vehicle driving speed based on the LED display panel and the speed sensor provided in the tunnel. As a result, it is possible to prevent a collision accident in the tunnel by raising the alertness of the vehicle driver.

Description

Intelligent vehicle congestion notification system in tunnels {SYSTEM FOR INTELLIGENTLY NOTIFYING TRAFFIC JAM IN TUNNEL}

Embodiments of the present invention are related to a technology that provides more accurate and intuitive information on vehicle congestion in a tunnel and predicts traffic conditions in a tunnel based on artificial intelligence technology.

Korea ranks second in the world in the number of traffic accident deaths per 10,000 vehicles among OECD countries, with a high rate of traffic accident fatalities. In the case of Korea, although an excellent road network has been established, the safety system for it is insufficient, resulting in a high rate of fatalities in traffic accidents.

In particular, in Korea, about 70% of the land is mountainous, and there are also about 2,600 tunnels. The total length of these tunnels is over 2,000 km, and as the number of tunnels increases year by year, the incidence of traffic accidents in the tunnel is also increasing in proportion. In the case of tunnels, unlike general roads, there is no shoulder or very narrow, so the risk of secondary accidents is greater when an accident occurs. Accordingly, in order to prevent frequent traffic accidents in the tunnel, it is necessary to raise the driver's awareness in the tunnel.

Korea Patent Publication No. 10-2294286 (2021.08.20)

The embodiments of the present invention intuitively display the traffic situation in the tunnel according to the vehicle driving speed based on the LED display board and the speed sensor provided in the tunnel, and the traffic situation in the tunnel based on the deep learning model and the occurrence of an accident in the tunnel This is to provide useful information to the user by predicting a point where the probability is greater than or equal to a threshold.

According to an exemplary embodiment, a plurality of LED display panels are attached to the upper portion of the tunnel spaced apart at a set interval along the longitudinal direction of the tunnel; a speed sensor attached to each of the LED display panels to detect a driving speed of a vehicle passing through the tunnel; a control unit controlling the light emission of each LED display panel so that the LED display panel emits light in different colors according to the driving speed of the vehicle detected by the speed sensor; a collecting unit for collecting information on patterns of emission colors from the plurality of LED display panels according to the passage of time and information on a duration time of emission colors from each of the LED display panels; and a prediction unit for predicting the traffic situation in the tunnel after a set time has elapsed from the current time by inputting information on the pattern for the emission color and the duration time for the emission color into a set deep learning model, wherein the LED display panel is a first floodlight corresponding to the traveling speed of the first range, a second floodlight corresponding to a traveling speed of a second range lower than the first range, and a traveling speed of a third range lower than the second range and a corresponding third floodlight, wherein the first floodlight, the second floodlight, and the third floodlight emit light in different colors, and the control unit includes the first floodlight and the second floodlight. Among the lights and the third floodlight, the floodlight corresponding to the driving speed of the vehicle passing the LED display panel is emitted for a set time, and the floodlight not corresponding to the driving speed of the vehicle passing the LED display panel is turned off, and the control unit Under the control of a plurality of LED display panels provided in the tunnel, the emission color is changed in real time, and the control unit displays information on the predicted traffic situation on a plurality of prediction electronic boards installed at a set point in the tunnel. An intelligent vehicle congestion notification system in the tunnel is provided.

The collecting unit collects information on the actual number of accidents and accident occurrence points in the tunnel for each information on the traffic situation displayed on the predicted electric sign board, and the predicting unit collects information on the traffic condition displayed on the predictive electric sign board. The deep learning model learns information about the actual number of accidents and the point of occurrence of accidents in the tunnel, and predicts the point where the probability of occurrence of an accident in the tunnel according to the predicted traffic situation in the tunnel is greater than or equal to a threshold, and the control unit may additionally display information on the possibility of an accident on the prediction electronic board closest to a point where the probability of the accident is greater than or equal to a threshold among the plurality of prediction electronic boards.

The intelligent vehicle congestion notification system in the tunnel further includes an image acquisition unit provided at a side of the tunnel to acquire image data of a vehicle passing through the tunnel, wherein the prediction unit includes: The predicted information on the traffic situation is trained on the deep learning model, and information on the pattern of the luminous color and the duration of the luminous color and the image data obtained from the image acquisition unit are used in the tunnel. traffic conditions can be predicted.

The image acquisition unit selects pixel values in a range set based on the driving direction of the vehicle, and calculates a Gaussian function calculated by the following equation based on a pixel value (x) corresponding to the center of the set range After obtaining (G(x)), the noise according to the driving direction may be removed by subtracting the calculated value (G(x)) from the pixel value (x).

[Equation]

(where σ is a predefined standard deviation value)

The image acquisition unit sets a different number of cases for a combination of the set range or the number of pixel values selected from the set range, the kernel size of the Gaussian function, and the standard deviation values. By repeating more than a number of times, the optimal number of pixel values selected from the set range or the set range, the kernel size of the Gaussian function, and the standard deviation value may be selected, respectively.

According to embodiments of the present invention, it is possible to intuitively display the traffic situation in the tunnel according to the vehicle driving speed based on the LED display panel and the speed sensor provided in the tunnel. Accidents can be prevented.

In addition, according to embodiments of the present invention, after a set time from the current time has elapsed based on the pattern for the emission color in the LED display, the duration for the emission color, image data of a vehicle passing through the tunnel, etc., in the tunnel It is possible to provide useful information to vehicle drivers by predicting the traffic situation of

In addition, according to the embodiments of the present invention, information on the actual number of accidents and the point of occurrence of accidents in the tunnel is collected for each information about the traffic situation displayed on the predicted electric sign board, and based on this, the probability of occurrence of an accident in the tunnel is a threshold value It is possible to provide useful information to the vehicle driver by predicting an abnormal point and displaying the relevant information on the prediction electronic board.

1 is a block diagram showing a detailed configuration of an intelligent vehicle congestion notification system in a tunnel according to an embodiment of the present invention;
2 is a view for explaining a process of controlling the light emission of the LED display panel according to the driving speed of the vehicle in the controller according to an embodiment of the present invention;
3 is an example of a light emitting pattern of an LED display panel according to an embodiment of the present invention;
4 is an example of a light emitting pattern of an LED display panel according to another embodiment of the present invention;
5 is a flowchart illustrating an intelligent vehicle congestion notification method in a tunnel according to an embodiment of the present invention;
6 is a block diagram illustrating and describing a computing environment including a computing device suitable for use in example embodiments;

Hereinafter, specific embodiments of the present invention will be described with reference to the drawings. The following detailed description is provided to provide a comprehensive understanding of the methods, apparatus, and/or systems described herein. However, this is merely an example and the present invention is not limited thereto.

In describing the embodiments of the present invention, if it is determined that the detailed description of the known technology related to the present invention may unnecessarily obscure the gist of the present invention, the detailed description thereof will be omitted. And, the terms to be described later are terms defined in consideration of functions in the present invention, which may vary according to intentions or customs of users and operators. Therefore, the definition should be made based on the content throughout this specification. The terminology used in the detailed description is for the purpose of describing embodiments of the present invention only, and should in no way be limiting. Unless explicitly used otherwise, expressions in the singular include the meaning of the plural. In this description, expressions such as “comprising” or “comprising” are intended to indicate certain features, numbers, steps, acts, elements, some or a combination thereof, one or more other than those described. It should not be construed to exclude the presence or possibility of other features, numbers, steps, acts, elements, or any part or combination thereof.

1 is a block diagram illustrating a detailed configuration of an intelligent vehicle congestion notification system 100 in a tunnel according to an embodiment of the present invention. 1, the intelligent vehicle congestion notification system 100 in the tunnel according to an embodiment of the present invention includes an LED display panel 102, a speed sensor 104, a control unit 106, a collection unit 108, It includes an image acquisition unit 110 and a prediction unit 112 .

A plurality of LED display panels 102 are attached to the top of the tunnel (not shown) at set intervals along the longitudinal direction of the tunnel. In the present embodiments, a tunnel is a road made by drilling a mountain or a bottom part of the sea, a river, etc., and means a passage through which a vehicle passes. A plurality of LED display panels 102 may be attached at intervals of 100 meters at the top of the tunnel, for example. In this case, each of the LED display panels 102 may include a plurality of floodlights for emitting light of a specific color.

As an example, each of the LED display panels 102 may include one or more of a first floodlight 102-1, a second floodlight 102-2, and a third floodlight 102-3. One LED display panel 102 may include, for example, three first floodlights 102-1, three second floodlights 102-2, and three third floodlights 102-3. can Here, the first floodlight 102-1 emits green light, for example, the second floodlight 102-2 emits yellow light, and the third floodlight 102-3 emits light. A plurality of LED modules may be provided therein to emit red light. As will be described later, the driving of the first floodlight 102-1, the second floodlight 102-2, and the third floodlight 102-3 is controlled under the control of the controller 106 to emit light or can be turned off

The speed sensor 104 is attached to each of the LED display panels 102 to detect the running speed of the vehicle passing through the tunnel. The speed sensor 104 may be attached to the front of the LED display panel 102 to face the vehicle, for example. Each of the speed sensors 104 can detect the driving speed of the vehicle passing through the tunnel, and since the speed sensors 104 are spaced apart at a set interval, the driving speed of the vehicle at each point where the speed sensor 104 is installed is It is measured. If, for a set time period (for example, from 2:0 pm to 2:01 pm), when the driving speed of the vehicle detected by one speed sensor 104 is plural, the speed sensor 104 may The average value of the driving speed may be determined as the driving speed sensed during the set time.

The control unit 106 controls the light emission of each LED display panel 102 so that the LED display panel 102 emits light in different colors according to the driving speed of the vehicle detected by the speed sensor 104 . As described above, each LED display panel 102 may include a plurality of first floodlights 102-1, a plurality of second floodlights 102-2, and a plurality of third floodlights 102-3. Also, each floodlight 102-1, 102-2, 102-3 may be configured to emit light in different colors. The control unit 106 collects information on the traveling speed of the vehicle detected by each speed sensor 104 from each speed sensor 104, and drives each LED display panel 102 according to the collected traveling speed of the vehicle. can be controlled.

Specifically, the control unit 106 controls the vehicle passing the LED display panel 102 for a set time among the first floodlight 102-1, the second floodlight 102-2, and the third floodlight 102-3. The floodlight corresponding to the driving speed may be emitted, and the floodlight not corresponding to the driving speed of the vehicle passing the LED display panel 102 may be turned off.

For example, when the driving speed of the vehicle passing the first LED display panel 102 is 40 to 50 km/h, the control unit 106 may emit the first floodlight 102-1 of the LED display panel 102 to light. In this case, green light is emitted. This indicates that the current traffic situation in the tunnel is smooth, and vehicle drivers can intuitively know this.

In addition, when the driving speed of the vehicle passing the first LED display panel 102 is 20 to 40 km/h, the control unit 106 may emit the second floodlight 102-2 of the LED display panel 102, and this In this case, a yellow light is emitted. This indicates that the current traffic situation in the tunnel is delayed, and vehicle drivers can intuitively know this.

In addition, when the driving speed of the vehicle passing the first LED display panel 102 is 0 to 20 km/h, the control unit 106 may emit the third floodlight 102-3 of the LED display panel 102, and this In this case, a yellow light is emitted. This indicates that the current traffic situation in the tunnel is congested, and vehicle drivers can intuitively know this.

In this way, the emission color of each of the N LED display panels 102 provided in the tunnel may be changed in real time according to the control of the controller 106 . Since each of these N LED display panels 102 emits light corresponding to the driving speed of the vehicle in the corresponding section, vehicle drivers intuitively see the pattern of light emitted from the continuous LED display panel 102 in the tunnel. You can easily check the traffic conditions.

The collecting unit 108 collects information on patterns of emission colors from the plurality of LED display panels 102 and duration of emission colors from each of the LED display panels 102 according to the passage of time. For example, if it is assumed that there are 100 LED display panels 102 in the tunnel, the collection unit 108 generates a pattern for each of the 100 LED display panels 102 and 100 LED panels according to the passage of time. Information about the duration of the emission color at 102 may be collected.

1st LED display panel - green, green, yellow, yellow, green,… / Continuous green from 2:00 to 2:10, yellow from 2:10 to 2:20,… etc.

2nd LED display panel - green, yellow, yellow, yellow, green,… / Sustains green from 2:00 to 2:05, yellow lasting from 2:05 to 2:15,… etc.

3rd LED display panel - yellow, yellow, yellow, yellow, green,… / Lasts yellow from 2:20 to 2:20, green from 2:20 to 2:30,… etc.

…

As will be described later, the pattern for the emission color of the plurality of LED display panels 102 according to the passage of time collected by the collection unit 108 and the duration of the emission color from each LED display panel 102 . The information is input to the set deep learning model, and accordingly, it can be used as learning data to predict the traffic situation in the tunnel.

The image acquisition unit 110 is provided on the side of the tunnel to acquire image data of a vehicle passing through the tunnel. As will be described later, the prediction unit 112 includes a pattern for the emission color of the plurality of LED display panels 102 according to the passage of time collected by the collection unit 108 , and the emission color of each LED display panel 102 . It is possible to predict the traffic situation in the tunnel based on the information on the duration of . In this case, the image data of the vehicle acquired by the image acquisition unit 110 is a photographed image continuously acquired along the longitudinal direction of the tunnel, and may include different image patterns according to the driving speed of the vehicle. That is, a pattern included in the image data in a case in which the driving speed of the vehicle is relatively fast and a case in which the driving speed of the vehicle is relatively slow may be different from each other. For example, when the driving speed of the vehicle is relatively fast, the pattern included in the image data may be larger than the original size of the vehicle. If the driving speed of the vehicle is relatively slow, the pattern included in the image data is the original size of the vehicle. may be similar to However, the image data may include noise for elements other than the vehicle image (eg, a guide rail, a pedestrian passage, etc. existing in a tunnel along the driving direction of the vehicle). Accordingly, in the embodiments of the present invention, the image acquisition unit 110 may perform a noise removal process after acquiring image data of a vehicle.

Specifically, the image acquisition unit 110 selects pixel values in a range set based on the driving direction of the vehicle from the acquired image data, and performs the following math on the basis of the pixel value (x) corresponding to the center of the set range After obtaining the calculated value G(x) of the Gaussian function expressed by the equation, the calculated value G(x) may be subtracted from the pixel value x to remove the noise according to the driving direction. Here, x represents a pixel in the image data. In particular, the noise removal is to remove a signal (eg, a signal for a guide rail, a pedestrian passage, etc. existing in a tunnel along the driving direction of the vehicle) that is seen in a horizontal direction, that is, a straight line in the driving direction of the vehicle.

[Equation]

(where σ is a predefined standard deviation value)

In general, the mask matrix of the Gaussian filter has a large value in the center and has a value close to 0 toward the edge. It is the same as in the case of calculating the weighted average by setting the weights smaller as the distance increases. In the case of a general Gaussian filter or average value filter, the pixel value of the target pixel is replaced with the average value of neighboring pixel values of the target pixel or the calculated value of the Gaussian function. There is a problem in that not only the noise according to the direction but also the signal related to the vehicle is removed. Accordingly, in the embodiments of the present invention, only noise according to the driving direction of the vehicle can be selectively removed by subtracting the calculated value of the Gaussian function from the pixel value of the target pixel, rather than applying the Gaussian filter as it is.

The image acquisition unit 110 selects pixel values within a range set based on the driving direction of the vehicle (eg, selects pixel values within 2 m within a specific section along the driving direction), and corresponds to the center of the set range. After obtaining the calculated value G(x) of the Gaussian function expressed by the above equation on the basis of the pixel value x, the calculated value G(x) is subtracted from the pixel value x Thus, noise according to the driving direction can be removed.

That is, the image acquisition unit 110 may perform an operation of x - G(x) to remove noise according to the driving direction of the vehicle. In this case, the signal related to the vehicle is preserved as much as possible, and only the linear signal (noise) that is different in the horizontal direction, that is, the driving direction of the vehicle, can be selectively removed. In addition, the image acquisition unit 110 sets the number of different cases for combinations of the set range or the number of pixel values selected from the set range, the kernel size of the Gaussian function, and the standard deviation values. case) is repeated a set number of times or more to select the optimal number of pixel values selected from the set range or the set range, the kernel size of the Gaussian function, and the standard deviation value, respectively. The image acquisition unit 110 may select an optimal parameter by repeatedly performing the number of cases for all combinations between parameters through, for example, a Python iteration statement.

The prediction unit 112 determines the pattern for the emission color from the plurality of LED display panels 102 according to the passage of time collected by the collection unit 108 and the duration time for the emission color from each LED display panel 102 . It is possible to predict the traffic situation in the tunnel after a set time has elapsed from the current time (for example, 30 minutes have passed from the current time) based on the information. The prediction unit 112 is a deep learning model in which information on a pattern of a light emission color from a plurality of LED display panels 102 according to the passage of time and information on a duration time for a light emission color from each LED display panel 102 are set. can be entered in Here, the deep learning model may be, for example, a Convolutional Neural Network (CNN), a Recurrent Neural Network (RNN), or the like. The deep learning model includes information about the pattern of the emission color in the plurality of LED display panels 102 according to the passage of time and information on the duration of the emission color in each LED display panel 102 and the external server ( For example, by learning the traffic condition information in the tunnel collected from a vehicle navigation server, a traffic control server, etc.), the pattern for the light emission color in the LED display panel 102 and the light emission in each LED display panel 102 . Depending on the duration of the color, it is possible to understand how the traffic conditions in the tunnel change. Thereafter, by learning the traffic condition information in the tunnel from the collecting unit 108 , information on the pattern of the emission color in the LED display panel 102 and the duration of the emission color in each LED display panel 102 . When is newly collected, the prediction unit 112 may predict the traffic situation in the tunnel by inputting it into the deep learning model. The control unit 106 may display information on the predicted traffic condition on a plurality of prediction electronic boards installed at a set point in the tunnel. A plurality of the prediction electronic boards may be installed, for example, spaced apart from each other at intervals of 100 meters in a tunnel.

In addition, the collection unit 108 may additionally collect information on the actual number of accidents and accident occurrence points in the tunnel for each information on the traffic conditions displayed on the predicted electric sign board. The collection unit 108 may additionally collect information on the actual number of accidents and accident locations in the tunnel for each information on the traffic conditions displayed on the predicted electric sign board from an external server such as, for example, a traffic control server. . For example, when information on traffic conditions between sections A to B in the tunnel is congested at a specific time period (eg, 2 pm to 3 pm), the collection unit 108 may It is possible to additionally collect information about the actual number of accidents and the accident point (eg, a specific point in the A to B section) in the section A to B. Accordingly, the prediction unit 112 may teach the deep learning model the information on the actual number of accidents and the accident point in the tunnel for each information about the traffic situation displayed on the prediction electric sign board. In this case, the deep learning model may learn information about how many actual accidents occur and at which point in the tunnel an accident occurs according to the traffic situation predicted by the prediction unit 112 . Thereafter, the prediction unit 112 may predict the traffic situation in the tunnel and at the same time predict the possibility of an accident occurring in the tunnel according to the predicted traffic situation and a point where the accident is likely to occur. That is, the prediction unit 112 may predict a point where the probability of an accident occurring in the tunnel according to the predicted traffic situation in the tunnel is equal to or greater than a threshold. In this case, the control unit 106 may additionally display information on the possibility of an accident on the prediction electronic board closest to a point where the probability of the accident is greater than or equal to a threshold among the plurality of prediction electronic boards.

Also, as described above, the image acquisition unit 110 may be provided at the side of the tunnel to acquire image data of a vehicle passing through the tunnel. The prediction unit 112 may train the deep learning model with the image data acquired over time and the predicted traffic situation information, in this case, between the pattern included in the image data and the traffic situation in the tunnel. Correlations can be learned automatically. Thereafter, the prediction unit 112 may predict the traffic situation in the tunnel based on the image data newly acquired by the image acquisition unit 110 . In addition, the prediction unit 112 includes at least two or more of the pattern of the emission color collected by the collection unit 108 , information on the duration of the emission color, and the image data acquired by the image acquisition unit 110 . Based on the information, it is possible to predict the traffic situation in the tunnel. The prediction unit 112 can predict the traffic situation in the tunnel by using the information collected by the collection unit 108 and the information acquired by the image acquisition unit 110 in various ways. The accuracy of traffic situation prediction can be improved.

2 is a view for explaining a process of controlling the light emission of the LED display panel 102 according to the driving speed of the vehicle in the controller 106 according to an embodiment of the present invention.

As described above, the LED display panel 102 may include a plurality of first floodlights 102-1, a plurality of second floodlights 102-2, and a plurality of third floodlights 102-3. In addition, each floodlight 102-1, 102-2, 102-3 may be configured to emit light in different colors. The controller 106 may control the light emission of each LED display panel 102 so that the LED display panel 102 emits light in different colors according to the driving speed of the vehicle detected by the speed sensor 104 . The control unit 106 controls the driving speed of the vehicle passing the LED display panel 102 for a set time among the first floodlight 102-1, the second floodlight 102-2, and the third floodlight 102-3. The corresponding floodlight may be emitted, and the floodlight not corresponding to the driving speed of the vehicle passing the LED display panel 102 may be turned off. At this time, the controller 106 controls each floodlight 102 through the SMPS 114 connected to the first floodlight 102-1, the second floodlight 102-2, and the third floodlight 102-3, respectively. ), that is, whether or not to emit light can be controlled.

For example, when the driving speed of the vehicle passing the first LED display panel 102 is 40 to 50 km/h (Case 1), the control unit 106 controls the first floodlight 102-1 of the LED display panel 102 ). can emit light, and in this case, green light is emitted. This indicates that the current traffic situation in the tunnel is smooth, and vehicle drivers can intuitively know this.

In addition, when the driving speed of the vehicle passing the first LED display panel 102 is 20 to 40 km/h (Case 2), the control unit 106 emits the second floodlight 102-2 of the LED display panel 102. In this case, yellow light is emitted. This indicates that the current traffic situation in the tunnel is delayed, and vehicle drivers can intuitively know this.

In addition, when the driving speed of the vehicle passing the first LED display panel 102 is 0 to 20 km/h (Case 3), the control unit 106 emits the third floodlight 102-3 of the LED display panel 102 . In this case, yellow light is emitted. This indicates that the current traffic situation in the tunnel is congested, and vehicle drivers can intuitively know this.

3 and 4 are examples of light emission patterns of the LED display panel 102 according to an embodiment of the present invention.

Referring to FIG. 3 , when the traffic situation in the tunnel is smooth, it can be seen that each LED display panel 102 in the tunnel emits green light.

Referring to FIG. 4 , when the traffic situation in the tunnel is a delay or congestion situation, it can be confirmed that each of the LED display panels 102 in the tunnel is lit in yellow or red. Vehicle drivers can intuitively know the current traffic situation of the tunnel by checking the continuous light emission pattern of the LED display panel 102 as described above.

5 is a flowchart illustrating an intelligent vehicle congestion notification method in a tunnel according to an embodiment of the present invention. In the illustrated flowchart, the method is described by dividing the method into a plurality of steps, but at least some of the steps are performed in a different order, are performed together in combination with other steps, are omitted, are performed in sub-steps, or are not shown. One or more steps may be added and performed.

In step S102, the speed sensor 104 detects the traveling speed of the vehicle passing through the tunnel.

In step S104 , the controller 106 controls the light emission of each LED display panel 102 so that the LED display panel 102 emits light in different colors according to the driving speed of the vehicle detected by the speed sensor 104 .

In step S106 , the collection unit 108 collects information on a pattern for a light emission color from a plurality of LED display panels 102 according to time and a duration time for a light emission color from each LED display panel 102 . do.

In step S108, the prediction unit 112 inputs information on the pattern for the emission color and information on the duration of the emission color to the set deep learning model.

In step S110, the prediction unit 112 predicts the traffic situation in the tunnel after a set time has elapsed from the current time by using the deep learning model.

In addition, although not shown in the drawings, as described above, the collection unit 108 may collect information on the actual number of accidents and accident locations in the tunnel for each information on traffic conditions displayed on the predicted electric sign board. In this case, the prediction unit 112 trains the deep learning model with information on the actual number of accidents and the point of occurrence of accidents in the tunnel for each information about the traffic conditions displayed on the prediction electronic board, and predicts the number of accidents in the tunnel. It is possible to predict a point where the probability of an accident occurring within the tunnel according to traffic conditions is greater than or equal to a threshold. Thereafter, the control unit 110 may additionally display information on the possibility of an accident on the prediction electronic board closest to a point where the probability of the accident is greater than or equal to a threshold among the plurality of prediction electronic boards.

Also, the image acquisition unit 110 may acquire image data of a vehicle passing through the tunnel. In this case, the prediction unit 112 trains the deep learning model with the image data acquired over time and the predicted traffic situation information, and the pattern for the luminous color and the duration for the luminous color. It is possible to predict the traffic situation in the tunnel based on the information about , and the image data acquired by the image acquisition unit.

6 is a block diagram illustrating and describing a computing environment including a computing device suitable for use in example embodiments. In the illustrated embodiment, each component may have different functions and capabilities other than those described below, and may include additional components other than those described below.

The illustrated computing environment 10 includes a computing device 12 . In one embodiment, the computing device 12 may be one or more components included in the intelligent vehicle congestion notification system 100 in a tunnel, or the intelligent vehicle congestion notification system 100 in a tunnel.

Computing device 12 includes at least one processor 14 , computer readable storage medium 16 , and communication bus 18 . The processor 14 may cause the computing device 12 to operate in accordance with the exemplary embodiments discussed above. For example, the processor 14 may execute one or more programs stored in the computer-readable storage medium 16 . The one or more programs may include one or more computer-executable instructions that, when executed by the processor 14, configure the computing device 12 to perform operations in accordance with the exemplary embodiment. can be

Computer-readable storage medium 16 is configured to store computer-executable instructions or program code, program data, and/or other suitable form of information. The program 20 stored in the computer readable storage medium 16 includes a set of instructions executable by the processor 14 . In one embodiment, computer-readable storage medium 16 includes memory (volatile memory, such as random access memory, non-volatile memory, or a suitable combination thereof), one or more magnetic disk storage devices, optical disk storage devices, flash It may be memory devices, other forms of storage medium that can be accessed by computing device 12 and store desired information, or a suitable combination thereof.

Communication bus 18 interconnects various other components of computing device 12 , including processor 14 and computer readable storage medium 16 .

Computing device 12 may also include one or more input/output interfaces 22 and one or more network communication interfaces 26 that provide interfaces for one or more input/output devices 24 . The input/output interface 22 and the network communication interface 26 are coupled to the communication bus 18 . Input/output device 24 may be coupled to other components of computing device 12 via input/output interface 22 . Exemplary input/output device 24 may include a pointing device (such as a mouse or trackpad), a keyboard, a touch input device (such as a touchpad or touchscreen), a voice or sound input device, various types of sensor devices, and/or imaging devices. input devices and/or output devices such as display devices, printers, speakers and/or network cards. The exemplary input/output device 24 may be included in the computing device 12 as a component constituting the computing device 12 , and may be connected to the computing device 12 as a separate device distinct from the computing device 12 . may be

Although the present invention has been described in detail through representative embodiments above, those of ordinary skill in the art to which the present invention pertains can make various modifications to the above-described embodiments without departing from the scope of the present invention. will understand Therefore, the scope of the present invention should not be limited to the described embodiments, and should be defined by the claims described below as well as the claims and equivalents.

100: intelligent vehicle congestion notification system in the tunnel
102: LED display panel
104: speed sensor
106: control unit
108: collection unit
110: image acquisition unit
112: prediction unit
114: SMPS (Switching Mode Power Supply)

Claims (5)

a plurality of LED display panels attached to the upper part of the tunnel spaced apart at intervals set along the longitudinal direction of the tunnel;
a speed sensor attached to each of the LED display panels to detect a traveling speed of a vehicle passing through the tunnel;
a control unit controlling the light emission of each LED display panel so that the LED display panel emits light in different colors according to the driving speed of the vehicle detected by the speed sensor;
a collecting unit for collecting information on patterns of emission colors from the plurality of LED display panels according to the passage of time and information on durations of emission colors from each of the LED display panels; and
A prediction unit for predicting the traffic situation in the tunnel after a set time has elapsed from the current time by inputting information about the pattern for the luminescent color and the duration time for the luminous color into a set deep learning model,
The LED display panel includes a first floodlight corresponding to a traveling speed of a first range, a second floodlight corresponding to a traveling speed of a second range lower than the first range, and a driving speed of a third range lower than the second range. Includes a third floodlight corresponding to the speed,
The first floodlight, the second floodlight, and the third floodlight emit light in different colors,
The control unit emits a floodlight corresponding to a traveling speed of a vehicle passing the LED display panel for a set time among the first floodlight, the second floodlight, and the third floodlight, and the vehicle passing the LED display panel The floodlights that do not correspond to the driving speed are turned off, and the light emitting colors of the plurality of LED display panels provided in the tunnel are changed in real time according to the control of the controller,
The control unit, the intelligent vehicle congestion notification system in the tunnel, for displaying the predicted information on the traffic situation on a plurality of prediction electronic boards installed at a set point in the tunnel.
The method according to claim 1,
The collection unit collects information on the actual number of accidents and accident occurrence points in the tunnel for each information on the traffic situation displayed on the predicted electric sign board,
The prediction unit learns the information on the actual number of accidents and the accident point in the tunnel for each information about the traffic situation displayed on the predicted electric signboard to the deep learning model, and according to the predicted traffic situation in the tunnel Predict a point in the tunnel where the probability of an accident is higher than a threshold,
The control unit, the intelligent vehicle congestion notification system in the tunnel, to additionally display information on the possibility of an accident on a prediction electronic board closest to a point where the probability of the accident is greater than or equal to a threshold among the plurality of prediction electronic boards.
3. The method according to claim 2,
Further comprising an image acquisition unit provided on the side of the tunnel to acquire image data of a vehicle passing through the tunnel,
The prediction unit trains the deep learning model to learn the image data acquired over time and information on the predicted traffic situation, and information on the pattern for the luminous color and the duration for the luminous color and the An intelligent vehicle congestion notification system in the tunnel that predicts the traffic situation in the tunnel based on the image data acquired by the image acquisition unit.
4. The method of claim 3,
The image acquisition unit selects pixel values in a range set based on the driving direction of the vehicle, and a calculated value of a Gaussian function expressed by the following equation based on a pixel value (x) corresponding to the center of the set range After obtaining (G(x)), the intelligent vehicle congestion notification system in the tunnel subtracts the calculated value (G(x)) from the pixel value (x) to remove noise according to the driving direction.

[Equation]

(where σ is a predefined standard deviation value)
5. The method according to claim 4,
The image acquisition unit sets a different number of cases for a combination of the set range or the number of pixel values selected from the set range, the kernel size of the Gaussian function, and the standard deviation values. An intelligent vehicle congestion notification system in a tunnel that selects the optimal number of pixel values selected from the set range or the set range, the kernel size of the Gaussian function, and the standard deviation value by repeatedly performing the plurality of times or more.

Images