KR102212991B1 - Black ice sensing system and method of thereof - Google Patents

Abstract

The present invention relates to a system and a method for black ice sensing. The system includes: an image collecting unit collecting a vehicle image from at least one camera installed in a preset monitoring area and collecting an image of the back of at least one vehicle passing through the monitoring area; an image determination unit recognizing a vehicle in the image taken by the image collecting unit, extracting traveling information on the recognized vehicle, performing comparison to preset conditions, and determining whether black ice has been generated; and an image transmitting unit transmitting the vehicle image taken by the image collecting unit to a central management server in a case where the image determination unit determines that the monitoring area is dangerous due to black ice generation.

Description

Black ice detection system and its method TECHNICAL FIELD [BLACK ICE SENSING SYSTEM AND METHOD OF THEREOF]

The present invention relates to a black ice detection system and method thereof, and more particularly, to determine whether black ice occurs by analyzing a braking pattern of a vehicle passing through a surveillance area, and to provide guidance for preventing an accident in real time. It relates to a detection system and a method thereof.

In general, information on road surface conditions in bad weather such as water curtains, freezing, and snowfall plays an important role in efficient road management and traffic safety. Existing road surface condition management uses equipment called Road Weather Information System (RWIS), and this RWIS can provide predictive information to road surface and atmospheric condition prediction system using data from the weather observation system. System.

This RWIS provides information for the driver's safe driving by providing measured weather data and road surface condition information in real time, and can provide efficient decision-making information for road management to road operators, but such RWIS is an equipment installation. And because the maintenance cost is very expensive, there is a limit to the spread.

On the other hand, there are known various technologies related to a road surface condition determination device for reducing driving accidents by automatically detecting dangerous conditions such as freezing of the road or snow and notifying the driver in advance.

Korean Patent Publication No. 10-1365634, which is a prior art, uses a temperature sensor, a humidity sensor, an ultrasonic sensor, etc. to detect climate factors and analyze them to automatically provide spraying agents for snow removal and thawing when snow or roads are frozen. It discloses an unmanned spraying system that sprays with a device, and Korean Patent Publication No. 10-1136330 uses a plurality of polarized images to determine the road surface condition such as the presence or absence of moisture on the road surface and the road surface condition. Disclosed is a method of discrimination.

However, such a conventional technology is equipped with a buried sensor or a method of determining by watching a camera installed at each point of a road, and thus there are problems such as relatively high installation cost, an increase in labor costs, and frequent breakdowns.

In addition, there are many research cases and technology development cases at home and abroad for the acquisition and use of real-time road surface information, but the road surface information acquisition technology currently being used in the actual road site is implemented based on laser or laser, and due to its nature, installation and operation It has the disadvantage of being very expensive.

Therefore, there is a need for a study on a black ice detection system and a method for efficiently determining the freezing state of the road using CCTVs that have been used on the road, and for storing, sharing, and processing data for each road section.

The present invention can be used as a main variable of a vehicle's braking ramp to determine whether or not black ice is generated, thereby enabling detection at the initial stage of black ice occurrence, and providing information to managers and drivers immediately upon occurrence of black ice to prevent accidents in advance. It is an object of the present invention to provide a system and method for detecting black ice.

In addition, the present invention determines whether black ice occurs by using a neural network to determine whether black ice occurs by reflecting all of a plurality of input conditions, and builds a database for each condition to derive a judgment result optimized for each condition. It is an object of the present invention to provide a system and method for detecting black ice.

The problems to be solved by the present invention are not limited to the above-mentioned problems, and other problems to be solved by the present invention not mentioned here are to those of ordinary skill in the technical field to which the present invention belongs from the following description. It can be clearly understood.

The black ice detection system according to an embodiment of the present invention collects vehicle images from at least one or more cameras installed in a preset surveillance area, but collects an image from the rear of at least one vehicle passing through the surveillance area, The image collection unit recognizes the vehicle located in the image captured by the image collection unit, extracts the driving information of the recognized vehicle, and compares it with a preset condition to determine whether black ice is generated or not, and the surveillance area is black. When it is determined that it is dangerous due to ice generation, it includes an image transmission unit for transmitting the vehicle image photographed by the image collection unit to the central management server.

In addition, the image determination unit extracts driving information including the current speed and braking information of the vehicle, compares the current speed of the vehicle with a preset reference speed, and when the current speed of the vehicle exceeds the reference speed, the vehicle brakes The information is analyzed, but the braking information of the vehicle is at least one of the number of braking times of the vehicle, the starting position of the braking, and the degree of distribution of the braking ramp as the main variable, and the number of braking per unit time of the vehicle passing through the monitoring area is counted. It is characterized by capturing the image of the vehicle as many times as the number of braking per unit time at the braking start position to extract the distribution image of the braking lamp, and determining whether there is a danger by combining the distribution image of the braking lamp and the weather information of the monitoring area.

In addition, the image judgment unit determines whether there is a danger by receiving at least one of the vehicle speed, the number of braking times per unit time, the braking start position, the distribution image of the braking lamp, and the weather in the monitoring area as a variable, including a multilayer neural network. A database is constructed by reflecting all of the plurality of inputs, determining whether black ice is generated in the surveillance area based on pre-stored learning data, and adding input conditions and determination results to the learning data.

In addition, the image transmission unit transmits the location information of the surveillance area to the predefined road situation room and the predefined navigation service company using the central management server, and further includes a separate communication device, and the surveillance area is black ice from the image judgment unit. When it is determined that it is dangerous due to the generation, it is characterized by displaying a notice of black ice on an electric signboard installed in the nearest location to the surveillance area before passing through the surveillance area.

In addition, the black ice detection method according to an embodiment of the present invention collects vehicle images from at least one or more cameras installed in a preset surveillance area, and collects images from the rear of at least one vehicle passing through the surveillance area. In step, the image determination step and the image determination step of recognizing the vehicle located in the image captured in the image collection step, extracting driving information of the recognized vehicle, and determining whether black ice is generated by comparing it with a preset condition When it is determined that it is dangerous due to the generation of black ice, transmitting the vehicle image captured in the image collection step to the central management server,

The image analysis step includes determining whether a vehicle passing through the surveillance area satisfies a preset minimum speed, determining whether the vehicle's brake lamp operates when the minimum speed is satisfied, and the vehicle's brake lamp If operated, extracting driving information of the vehicle, and determining whether black ice is generated by synthesizing the extracted driving information, weather information of the monitoring area, and pre-stored learning data.

The black ice detection system and method of the present invention can be detected in the initial stage of black ice generation by using it as a main variable of a vehicle's braking lamp to determine whether or not black ice is generated, and information is provided to the manager and driver immediately after the black ice occurs. By providing it, it has the effect of preventing accidents in advance.

In addition, by determining whether black ice occurs using a neural network, it is possible to determine whether black ice occurs by reflecting all of a plurality of input conditions, and to build a database for each condition to derive a judgment result optimized for each condition. Have.

1 is a block diagram of a black ice detection system according to an embodiment of the present invention.
2 is a view for explaining a multilayer neural network of the black ice detection system according to an embodiment of the present invention.
3 is a diagram illustrating an image determination unit of a black ice detection system according to an embodiment of the present invention.
4 is a diagram illustrating a warrant transmission unit of a black ice detection system according to an embodiment of the present invention.
5 is a flowchart of a black ice detection method according to an embodiment of the present invention.

Specific matters, including the problems to be solved, means for solving the problems, and effects of the invention as described above, are included in the following examples and drawings. Advantages and features of the present invention, and a method of achieving them will become apparent with reference to the embodiments described below in detail together with the accompanying drawings.

Hereinafter, the present invention will be described in more detail with reference to the accompanying drawings.

1 is a block diagram of a black ice detection system according to an embodiment of the present invention, FIG. 2 is a diagram for explaining a multilayer neural network of a black ice detection system according to an embodiment of the present invention, and FIG. A view for explaining the image determination unit of the black ice detection system according to an embodiment of the present invention, Figure 4 is a view for explaining the warrant transmission unit of the black ice detection system according to an embodiment of the present invention, Figure 5 is the present invention A flow chart of a black ice detection method according to an embodiment of the present invention.

<Example 1>

Referring to FIG. 1, the black ice detection system 100 may include an image collection unit 110, an image determination unit 120, and an image transmission unit 130.

More specifically, the image collection unit 110 may collect vehicle images from at least one or more cameras in a preset surveillance area, and collect the rear of at least one vehicle passing through the surveillance area as an image.

At this time, the camera may be used by using a CCTV installed on the road.

The image determination unit 120 recognizes a vehicle located in an image captured by the image collection unit 110, extracts driving information of the recognized vehicle, and determines whether black ice is generated by comparing it with a preset condition. can do.

In addition, when the image determination unit 120 determines that the surveillance area is dangerous due to the black ice generation, the image transmission unit 130 transmits the vehicle image captured by the image collection unit 110 to the central management server. I can.

The central management server may communicate with a general road control room, a highway management office, and the like.

More specifically, the image determination unit 120 extracts driving information including the current speed and braking information of the vehicle, and when the current speed of the vehicle and a preset reference speed are exceeded, the braking information of the vehicle Can be analyzed.

For example, when the reference speed is 30 km/h, braking information of the vehicle is not analyzed for a vehicle driving at a speed of 30 km or less, and braking information of the vehicle for a vehicle traveling exceeding 30 km. Can be analyzed.

That is, by determining a reference speed and determining whether black ice is generated using driving information of a vehicle that satisfies the reference speed, the amount of calculation is reduced and data processing is possible more quickly.

On the other hand, the braking information of the vehicle has at least one of the number of braking times of the vehicle, a braking start position, and a distribution degree of a braking ramp as a main variable, counting the number of braking per unit time of the vehicle passing through the monitoring area, and the vehicle It is possible to determine whether there is a danger by capturing the image of the vehicle at the braking start position of, extracting the distribution image of the braking lamp, and synthesizing the distribution image of the braking lamp and weather information of the monitoring area.

In addition, the image determination unit 120 receives as a variable at least one of the vehicle speed, the number of braking times per unit time, the braking start position, the distribution image of the braking ramp, and the weather information of the monitoring area as a variable to determine whether there is danger. However, by reflecting all of a plurality of inputs including a multilayer neural network, determining whether black ice is generated in the surveillance area based on pre-stored learning data, and adding the input variables and determination results to the learning data. Build a database.

As an example, referring to FIG. 2, the image determination unit 120 may receive 7 types of data, and use a multilayer neural network structure in which one data is output by collecting data input through a multilayer structure. It is possible to determine whether the surveillance area is in danger.

In addition, since it is necessary to collect a sufficient amount of training data in order to more accurately determine whether black ice is generated, the input data generated by the image determination unit 120 and the determination result derived based on the input data are learned. In addition to data, a database may be built, and performance of the black ice detection system 100 may be improved.

In addition, it is possible to implement the black ice detection system 100 optimized for the monitoring area by dividing the constructed database by conditions such as time and weather.

Meanwhile, the image determination unit 120 may use braking information of the vehicle as a main variable to determine whether black ice is generated.

As an example, referring to FIG. 3, the braking lamp distribution image is generated by capturing the rear image of the vehicle photographed by the image collecting unit 110, but if the braking lamp is operated 4 times during the unit time of the vehicle, the The vehicle image while the vehicle is braked 4 times may be captured, and the captured image may be superimposed to generate the brake lamp distribution image.

As shown in Fig. 3(a), when the amount of change per unit time of the distribution image of the braking lamp is small, and the distance the vehicle moves from the braking start position while the braking lamp is operating is within a preset distance, the image judgment The unit 120 determines that the risk of the monitoring area is low by combining the current time zone and weather information, and since the vehicle driving above the preset reference speed has been braked four times during a unit time, the monitoring area is called a congestion section. I can judge.

Meanwhile, as shown in FIG. 3(b), when the amount of change per unit time in the distribution image of the brake lamp is large, and the distance the vehicle moves from the braking start position while the brake lamp is operating exceeds a preset reference distance , The image determination unit 120 synthesizes the current time zone and weather information to determine that a precursor of black ice generation has occurred in the surveillance area, and transmits the vehicle image to the central management server through the image transmission unit 130. Can be transmitted.

In this case, when the vehicle speed exceeds a preset maximum speed, an error may occur in generating the brake lamp distribution image, so a separate algorithm for correcting the brake lamp distribution image based on the vehicle speed is further added. It can also be included.

In more detail, the image transmission unit 130 transmits the location information of the surveillance area to a known road situation room and a known navigation service company using the central management server, and further includes a separate communication device. When the image determination unit 120 determines that the surveillance area is dangerous due to the generation of black ice, before passing through the surveillance area, a black ice warning phrase may be displayed on an electric sign installed at a position closest to the surveillance area.

For example, referring to FIG. 4, the image transmission unit 130 may call the manager of the road injury room 200 by transmitting the location information of the surveillance area and the vehicle image using the central management server. .

In addition, by transmitting whether or not black ice is generated in the surveillance area to the predetermined navigation service company 300, information on the black ice generation area may be reflected in real time in a navigation service provided by the navigation service company.

In addition, as a separate communication device, a long-distance wireless communication system including at least one of a LoRa communication system and an Internet of Things (IoT) communication system is applied, and an LED billboard installed in the nearest position to the monitoring area before passing through the monitoring area. A signal can be transmitted to display the black ice caution phrase on 300.

That is, the image transmission unit 130 transmits whether black ice is generated to the road control room 200, so that the manager of the road control room 200 immediately checks whether black ice is generated in the surveillance area, and Actions can be taken to ensure that the black ice removal operation is done quickly.

In addition, when the vehicle image is checked in the road control room 200 and then moved to the surveillance area, the input variable and the determination result corresponding to the vehicle image are designated as a set and added to the learning data, It is possible to establish a guideline for determining the dispatch of the road control room 200.

In addition, by using the navigation service company 300 and the LED billboard 400, the black ice generation information of the monitoring area can be quickly provided to the driver, thereby inducing the driver to drive safely.

<Example 2>

The black ice detection method according to an embodiment of the present invention includes an image collection step of collecting vehicle images from at least one camera installed in a preset surveillance area, and collecting rear images of at least one vehicle passing through the surveillance area, In the image determination step of recognizing a vehicle located in the image captured in the image collection step, extracting driving information of the recognized vehicle, and determining whether black ice is generated by comparing it with a preset condition, and the image determination step, the When it is determined that the surveillance area is dangerous due to the generation of black ice, transmitting the vehicle image captured in the image collection step to a central management server.

In more detail, the image determining step includes determining whether a vehicle passing through the surveillance area satisfies a preset minimum speed, and if the minimum speed is satisfied, whether a braking lamp of the vehicle operates. Determining, when the vehicle's braking lamp is operated, extracting driving information of the vehicle, and determining whether black ice is generated by combining the extracted driving information, weather information of the monitoring area, and pre-stored learning data It may include the step of.

As an example, referring to FIG. 5, the black ice detection method collects an image of a vehicle passing through the surveillance area (S110), extracts the driving speed of the vehicle from the image of the vehicle, and sets the vehicle to a preset minimum. It is determined whether the speed is exceeded (S121), and when the vehicle exceeds a preset minimum speed, it is determined whether the brake lamp of the vehicle is operated (S122), and when the brake lamp is operated, the driving information of the vehicle is It can be extracted (S123).

In this case, the driving information of the vehicle may include at least one of whether the braking ramp is operated, the braking ramp start position, the number of braking ramps, and the braking ramp distribution image, including the driving speed of the vehicle. .

It is determined whether black ice is generated in the surveillance area by comparing the driving information of the vehicle with a preset condition (S124), and when it is determined that black ice is generated in the surveillance area, the vehicle image is transmitted to the central management server. (S131) You can. In addition, before passing through the surveillance area, a black ice warning phrase of the surveillance area may be displayed on an LED billboard installed at a position closest to the surveillance area (S132).

In addition, using the image transmitted to the central management server and the location information of the surveillance area, the road control room helps to determine whether to go to the site for the black ice removal operation in the surveillance area, and the navigation service company monitors the surveillance area. Information can be provided to quickly guide the driver to information about the area's black ice generation area.

Meanwhile, in the step of determining whether the vehicle satisfies a preset minimum speed (S121), in the step of determining whether the vehicle is traveling below the minimum speed and whether the black ice is generated (S124), the black ice is not generated. When it is determined, by displaying the congestion guidance phrase of the surveillance area on the LED billboard (S133), the driver may guide the driver to drive slowly.

According to the effects of the present invention as described above, by determining whether black ice is generated by using it as a main variable of the braking lamp of the vehicle, it is possible to detect black ice in the early stages of occurrence, and information is provided to the manager and driver immediately after the black ice occurs Thus, a black ice detection system and method capable of preventing an accident in advance may be provided.

In addition, the present invention determines whether black ice occurs by using a neural network to determine whether black ice occurs by reflecting all of a plurality of input conditions, and builds a database for each condition to derive a judgment result optimized for each condition. A capable black ice detection system and method may be provided.

In addition, the black ice detection method according to an embodiment of the present invention may be recorded in a computer-readable medium including program instructions for performing various computer-implemented operations. The computer-readable medium may include program instructions, data files, data structures, etc. alone or in combination. The medium may be a program instruction specially designed and configured for the present invention, or may be known and usable to those skilled in computer software. Examples of computer-readable recording media include magnetic media such as hard disks, floppy disks, and magnetic tapes, optical media such as CD-ROMs and DVDs, and magnetic media such as floptical disks. -A hardware device specially configured to store and execute program instructions such as magneto-optical media, and ROM, RAM, flash memory, and the like. Examples of program instructions include not only machine language codes such as those produced by a compiler but also high-level language codes that can be executed by a computer using an interpreter or the like.

As described above, although an embodiment of the present invention has been described by a limited embodiment and the drawings, an embodiment of the present invention is not limited to the above-described embodiment, which is a common knowledge in the field to which the present invention belongs. Anyone who has it can make various modifications and variations from these substrates. Therefore, one embodiment of the present invention should be grasped only by the claims described below, and all equivalent or equivalent modifications thereof will be said to belong to the scope of the inventive concept.

110: image collection unit
120: image judgment unit
130: image transmission unit

Claims (5)

An image collection unit for collecting vehicle images from at least one or more cameras installed in a predetermined surveillance area, and collecting rear images of at least one vehicle passing through the surveillance area;
An image determination unit that recognizes a vehicle located in the image captured by the image collection unit, extracts driving information of the recognized vehicle, and determines whether black ice is generated by comparing it with a preset condition; And
An image transmission unit for transmitting the vehicle image captured by the image collection unit to a central management server when the image determination unit determines that the surveillance area is dangerous due to black ice generation;
Black ice detection system comprising a. The method of claim 1,
The image determination unit,
Extracting driving information including the current speed and braking information of the vehicle, comparing the current speed of the vehicle with a preset reference speed, and when the current speed of the vehicle exceeds the reference speed, braking information of the vehicle But analyze
The braking information of the vehicle has at least one of the number of braking times of the vehicle, a braking start position, and a distribution degree of a braking ramp as a main variable,
Counting the number of braking times per unit time of the vehicle passing through the monitoring area, capturing the image of the vehicle as much as the number of braking times per unit time at the braking start position of the vehicle to extract the distribution image of the braking lamp, and the distribution of the braking lamp Black ice detection system, characterized in that to determine whether there is a danger by synthesizing the image and weather information of the surveillance area. The method of claim 2,
The image determination unit,
At least one of the vehicle speed, the number of braking times per unit time, the braking start position, the distribution image of the braking ramp, and the weather in the monitoring area are input as a variable to determine whether there is danger,
A database is built by reflecting all of a plurality of inputs including a multilayer neural network, determining whether black ice is generated in the surveillance area based on pre-stored learning data, and adding the input variables and determination results to the learning data. Black ice detection system, characterized in that. The method of claim 1,
The image transmission unit,
Using the central management server, the location information of the surveillance area is transmitted to a known road situation room and a known navigation service company,
In the event that the video judgment unit further includes a separate communication device and determines that the surveillance area is dangerous due to the generation of black ice, pay attention to black ice on the billboard installed in the nearest location to the surveillance area before passing through the surveillance area. Black ice detection system, characterized in that to display a phrase. An image collection step of collecting vehicle images from at least one or more cameras installed in a preset surveillance area, and collecting rear images of at least one vehicle passing through the surveillance area;
An image determination step of recognizing a vehicle located in the image captured in the image collecting step, extracting driving information of the recognized vehicle, and determining whether black ice is generated by comparing it with a preset condition; And
When it is determined in the image determination step that the surveillance area is dangerous due to the black ice generation, transmitting the vehicle image captured in the image collection step to a central management server; and
The image judgment step,
Determining whether a vehicle passing through the surveillance area satisfies a preset minimum speed;
Determining whether or not the brake lamp of the vehicle operates when the minimum speed is satisfied;
Extracting driving information of the vehicle when the brake lamp of the vehicle is operated; And
And determining whether to generate black ice by synthesizing the extracted driving information, weather information of the monitoring area, and pre-stored learning data.

Images