KR102226321B1 - System for Providing Smart Traffic Information - Google Patents

Abstract

The smart traffic information providing system is a traffic control mechanism of a smart city, selecting an ambulance from a vehicle object photographed with a video camera and transmitting a priority message to a plurality of vehicles at the corresponding location to stop vehicle movement in advance, thereby enabling efficient traffic driving and congestion. Even in an urban area, emergency vehicles can reduce dangerous situations while enabling fast driving.

Description

Smart Traffic Information System {System for Providing Smart Traffic Information}

The present invention relates to a system for providing traffic information, and in particular, by selecting an ambulance from a vehicle object photographed with a video camera as a traffic control mechanism in a smart city and transmitting a priority message to a plurality of vehicles at the corresponding location, vehicle movement is stopped in advance. It relates to a smart traffic information providing system capable of efficient traffic driving.

Emergency vehicles, such as ambulances, generally have the right to cross an intersection against traffic signals. Emergency vehicles in the past typically rely on whistles, sirens and flashlights to warn other drivers entering an intersection where the emergency vehicle intends to cross the intersection.

However, in large cities, hearing impairment, air conditioning, audio systems, and other distractions often cause drivers of vehicles entering an intersection to be unaware of warnings issued by an incoming emergency vehicle, or are unable to move even if the driver attempts to move to another location. There was a problem in that there was no space to give way to an emergency vehicle.

If the emergency vehicle is a hospital ambulance, the patient's life depends on how quickly the emergency vehicle arrives at the hospital. However, in the case of a large city, the increasingly congested road situation does not improve, and it is a reality that fast driving is becoming more and more difficult as emergency vehicles reduce dangerous situations.

Korean Patent Application Publication No. 10-2017-0002640

In order to solve such a problem, the present invention selects an ambulance from a vehicle object photographed with a video camera as a traffic control mechanism of a smart city and transmits a priority message to a plurality of vehicles at the corresponding location to stop vehicle movement in advance. Its purpose is to provide a smart traffic information providing system that enables traffic driving.

Smart traffic information providing system according to the features of the present invention for achieving the above object,

Generate image data by photographing a vehicle on the road through a video camera unit, image processing the generated image data to generate an edge image representing the outline of the image, and an original image obtained from the video camera unit A connection server for extracting image data related to the vehicle object by mapping it to data; And

When receiving image data related to a vehicle object from the connection server, a classification result is generated by comparing a first image characteristic of each cell of the previously stored ambulance image data with a second image characteristic of each cell of the image data of the vehicle object. And, when a classification result indicating that the first image feature and the second image feature are the same is generated, the vehicle object is determined as an ambulance, and a priority control message including vehicle object information determined as an ambulance is generated to move a plurality of It includes a cloud server that transmits to the terminal.

The connection server includes a video camera unit that photographs a vehicle on the road to generate image data, and a gray scale conversion unit that converts RGB data, which is image data input from the video camera unit, into gray scale image data. , A Gaussian filter unit that removes noise by performing Gaussian blur processing on the gray scale image data received from the gray scale conversion unit, and an edge detection unit that generates an edge image using the canny edge detector from the noise And a mapping unit for extracting image data related to the vehicle object by mapping the edge image generated by the edge detection unit to the original image data obtained from the video camera unit.

According to the above-described configuration, according to the present invention, a priority message is transmitted to a plurality of vehicles to stop vehicle movement in advance, thereby enabling efficient traffic driving and rapid driving while reducing a dangerous situation even in a crowded city center.

1 is a diagram showing the configuration of a smart traffic information providing system according to an embodiment of the present invention.
2 is a block diagram schematically showing an internal configuration of a connection server according to an embodiment of the present invention.
3 is a flowchart illustrating an image processor for extracting a vehicle object from a connection server according to an embodiment of the present invention.
4 is a diagram showing an example of a traffic information service processed in a mobile terminal according to an embodiment of the present invention.

Throughout the specification, when a part "includes" a certain component, it means that other components may be further included rather than excluding other components unless specifically stated to the contrary.

1 is a diagram showing the configuration of a system for providing smart traffic information according to an embodiment of the present invention, FIG. 2 is a block diagram schematically showing the internal configuration of a connection server according to an embodiment of the present invention, and FIG. Is a flowchart showing an image processor for extracting a vehicle object from a connection server according to an embodiment of the present invention, and FIG. 4 is a view showing an example of a traffic information service processed by a mobile terminal according to an embodiment of the present invention.

The smart traffic information providing system 100 according to an embodiment of the present invention includes one or more mobile terminals 110, a connection server 120, and a cloud server 130.

At least one mobile terminal 110 may include a mobile phone, a smart phone, a laptop computer, a digital broadcasting terminal, a personal digital assistant (PDA), a portable multimedia player (PMP), a navigator, and the like. have.

Each mobile terminal 110 may download and store an application providing a traffic information service.

Each mobile terminal 110 is mounted or provided in a vehicle to perform network communication with the cloud server 130 and the connection server 120, and can transmit and receive necessary control messages, transmission messages, and response messages.

The connection server 120 is located at one side of a traffic light and a road to photograph a vehicle.

The cloud server 130 transmits and receives related message types, message protocols, message senders, and receivers through the IoT protocol, MQTT (Message Queuing Telemetry Transport), when data is transmitted and received between the mobile terminal 110 and the connection server 120.

The cloud server 130 is, for example, a device including a wired or wireless communication module, and may be a device that accesses a cloud environment and performs a predetermined task.

There is no limitation on the type of the cloud server 130, for example, a general desktop PC or portable, providing a cloud environment through wireless Internet or mobile Internet, and communicating with a cellular phone, a PCS phone ( PCS phone: Includes mobile terminals such as Personal Communications Services phone), Synchronous/Asynchronous IMT-2000 (International Mobile Telecommunication-2000), and in addition, Palm Personal Computer (PDA), Personal Digital Assistant (PDA), Smart It may be a smart phone, a wireless application protocol phone (WAP phone), a mobile play-station device, or a laptop computer, a tablet computer, a navigation system, or the like.

The network connecting the cloud server 130 may be a wired or wireless Internet, or may be a core network integrated with a wired public network, a wireless mobile communication network, or a portable Internet, and the TCP/IP protocol and various services existing at the upper layer thereof. That is, it may mean an open computer network structure that provides HTTP (Hyper Text Transfer Protocol), Telnet, FTP (File Transfer Protocol), DNS (Domain Name System), SMTP (Simple Mail Transfer Protocol), and the like.

The environment of the cloud server 130 may refer to a set of physical servers for providing a plurality of cloud services, and may refer to an environment in which a plurality of virtual machines implemented on a physical server are provided.

The connection server 120 according to an embodiment of the present invention includes a video camera unit 121, a gray scale conversion unit 122, a Gaussian filter unit 123, a noise removing unit 124, and a binary filter unit ( 125), an edge detection unit 126, a mapping unit 127, and a server control unit 128.

The video camera unit 121 generates image data by photographing a vehicle on the road.

The gray scale conversion unit 122 converts RGB data, which is image data input from the video camera unit 121, into gray scale image data. That is, the gray scale conversion unit 122 calculates an average value of RGB data and converts it into gray scale image data. Here, the gray scale image data is expressed in 0 to 255 gray scales.

The Gaussian filter unit 123 removes noise from gray scale image data received from the gray scale conversion unit 122.

The Gaussian filter unit 123 is configured with a size of 5×5, which is a basic mask size of a bilateral filter. Gaussian blur processing receives 5 pixels corresponding to a vertical line at every clock, completes the mask over a total of 5 clocks, and receives 5 vertical pixels at each subsequent clock after an initial delay of 5 clocks. Process the mask.

The noise removing unit 124 determines whether there is random noise by filtering the average value of pixel values for each line and the difference between the average values for each line in the image data divided into n areas in a specific direction, and determines whether there is random noise. A corrected pixel value obtained by correcting the pixel value is calculated to generate a corrected image from which random noise has been removed.

The binary filter unit 125 generates a binary image using a binary filter in the corrected image received from the noise removing unit 124 as vehicle area 1 and non-vehicle area 0.

The edge detection unit 126 uses the canny edge detector to determine the luminance component data of the binary image input from the binary filter unit 125 at a point where the contrast change is greater than that of other points, and traces the point to a line formed by tracking the point. An edge image is created by extracting the points that belong to it.

In more detail, the edge detection unit 126 removes noise by applying a Gaussian filter to a luminance image in a masking method, and differentially calculates the horizontal and vertical directions of the luminance component data to calculate the horizontal and vertical brightness values from the luminance component data. Calculate the degree of this change.

The edge detection unit 126 is a part in the form of a line, and the part with a large change in brightness value corresponds to the edge, and the points corresponding to the edge have directionality, so after analyzing the strength and direction of the edge, a preset maximum value The weak edge is removed, and the edge is detected by changing it to black and white based on a specific value. The edge detection unit 126 converts the detected edge into a contour, which is a structure in the form of a point, and fills the inside using the converted contour to generate an edge image. Here, the outline refers to a line along the boundary line of the object formed in the form of a closed orbit between edges.

The mapping unit 127 maps the edge image generated by the edge detection unit 126 to the original image data obtained from the video camera unit 121 to extract image data related to the vehicle object.

The server control unit 128 receives a control message requesting vehicle object information from the client communication unit 133 of the cloud server 130, and generates a response message including image data related to the vehicle object extracted from the mapping unit 127 Then, it is transmitted to the client communication unit 133.

The cloud server 130 includes an artificial neural network processing unit 131, a control unit 132 and a client communication unit 133.

The client communication unit 133 receives a control message requesting ambulance object information from the mobile terminal 110, generates a response message including ambulance object information processed by the artificial neural network processing unit 131, and transmits it to the mobile terminal 110. do.

The artificial neural network processing unit 131 stores image features of a plurality of cells obtained by dividing ambulance image data.

The artificial neural network processing unit 131 combines image features for each of a plurality of cells each extracted from the ambulance image data, extracts all image features of the ambulance, and stores them in a database (not shown).

The artificial neural network processing unit 131 learns feature data of ambulance image data by applying a deep learning technique, and uses it as training data.

A neural network is a recognition model implemented in software or hardware that mimics the computational power of a biological system using a large number of artificial neurons connected by connecting lines.

In a neural network, artificial neurons that simplify the function of biological neurons are used, and the artificial neurons can be interconnected through a connection line having a connection weight. The connection weight is a specific value of the connection line and can also be expressed as the connection strength. Neural networks can perform human cognitive processes or learning processes through artificial neurons. Artificial neurons may also be referred to as nodes.

The neural network may include an input layer, a hidden layer, and an output layer. The input layer may receive an input for performing learning and transmit it to the hidden layer, and the output layer may generate an output of the neural network based on signals received from nodes of the hidden layer.

The hidden layer is located between the input layer and the output layer, and can change the learning data transmitted through the input layer into a value that is easy to predict. Nodes included in the input layer and the hidden layer are connected to each other through a connection line having a connection weight, and nodes included in the hidden layer and the output layer may be connected to each other through a connection line having a connection weight.

The input layer, the hidden layer, and the output layer may include a plurality of nodes. The neural network may include a plurality of hidden layers. A neural network including a plurality of hidden layers is called a deep neural network, and learning a deep neural network is called deep learning.

The artificial neural network processing unit 131 may train a neural network through supervised learning. Supervised learning is a method of inputting training data and output data corresponding thereto to a neural network, and updating connection weights of connection lines so that output data corresponding to the training data is output.

For example, the artificial neural network processing unit 131 may update connection weights between artificial neurons through a delta rule and backpropagation learning.

In error backpropagation learning, after estimating the error by forward computation for the given training data, it starts at the output layer and advances in the direction of the hidden layer and the input layer, propagating the estimated error, and reducing the error. This is a method of updating the connection weight. The neural network processing proceeds in the direction of the input layer -> hidden layer -> output layer, but in error backpropagation learning, the update direction of the connection weight may proceed in the direction of the output layer -> hidden layer -> input layer.

The artificial neural network processing unit 131 is composed of an input layer, a hidden layer, and an output layer to input vehicle object image data to the input layer, and to output the classification result of whether the input vehicle object image data matches the ambulance to the output layer. You can learn.

When the vehicle object image data matches the ambulance image, the artificial neural network processor 131 may detect a feature value of the vehicle object image data and train the neural network using the feature value. In other words, the artificial neural network processing unit 131 may detect feature vectors from vehicle object image data and train the neural network using the detected feature vectors.

When receiving a response message including image data related to a vehicle object from the client communication unit 133, the control unit 132 extracts image data related to the vehicle object from the response message and transmits it to the artificial neural network processing unit 131.

The artificial neural network processing unit 131 generates a classification result by comparing the first image characteristic of each cell of the previously stored ambulance image data with the second image characteristic of each cell of the image data of the vehicle object, and transmits the classification result to the control unit 132.

When receiving a classification result indicating that the first image feature and the second image feature are the same from the artificial neural network processing unit 131, the controller 132 may determine the vehicle object as an ambulance.

When the client communication unit 133 receives a request message for requesting ambulance object information from the mobile terminal 110, the control unit 132 analyzes the location information of the mobile terminal in the request message of the mobile terminal, A priority control message including vehicle object information is generated and transmitted to one or more mobile terminals 110.

Here, the priority control message may include vehicle object detected in the C direction, detection time, date information, cloud server name, and traffic stop time information so as to obtain the moving time of the ambulance.

As an example, it can be represented as shown in [Table 1] below.

The cloud server 130 may generate a control message including vehicle object information determined as an ambulance through the client communication unit 133 and transmit it to the plurality of mobile terminals 110.

It is assumed that each mobile terminal 110 is activated by downloading an application providing a traffic information service.

4, in the application of each mobile terminal 110, upon receiving a priority control message from the cloud server 130, the traffic section to be stopped by the traffic information service and the vehicle object, which is an ambulance, are moved. The section is controlled to be output on the display through a map map with different colors.

In the above, embodiments of the present invention are not implemented only through an apparatus and/or a method, but may be implemented through a program for realizing a function corresponding to the configuration of the embodiment of the present invention, a recording medium in which the program is recorded, or the like. And, this implementation can be easily implemented by an expert in the technical field to which the present invention belongs from the description of the above-described embodiment.

Although the embodiments of the present invention have been described in detail above, the scope of the present invention is not limited thereto, and various modifications and improvements by those skilled in the art using the basic concept of the present invention defined in the following claims are also provided. It belongs to the scope of rights.

100: traffic information provision system
110: mobile terminal
120: connection server
130: cloud server

Claims (6)

Generate image data by photographing a vehicle on the road through a video camera unit, image processing the generated image data to generate an edge image representing the outline of the image, and the generated edge image as an original image obtained from the video camera unit A connection server for extracting image data related to the vehicle object by mapping it to data; And
When the image data related to the vehicle object received from the connection server matches the ambulance image, the image data related to the vehicle object is used as ambulance image data, and each image characteristic of a plurality of cells obtained by dividing the ambulance image data is stored. And, the feature data of the ambulance image data is learned by applying a deep learning technique, and used as training data, the first image feature of each cell of the ambulance image data stored in advance and the first image feature of each cell of the image data of the vehicle object. 2 An artificial neural network processor that compares image features to generate a classification result, and when a classification result indicating that the first image feature and the second image feature are the same is generated, determines the vehicle object as an ambulance and determines the vehicle object as an ambulance It includes a cloud server having a control unit for generating a priority control message including information and transmitting it to a plurality of mobile terminals,
The artificial neural network processing unit stores image features of a plurality of cells obtained by dividing the ambulance image data, learns the feature data of the ambulance image data by applying a deep learning technique, and uses it as learning data,
When the cloud server receives a request message for requesting ambulance object information from the mobile terminal, it analyzes the location information of the mobile terminal in the request message and generates a priority control message including vehicle object information determined as an ambulance from the corresponding location information. And transmits it to the plurality of mobile terminals,
The priority control message includes vehicle object detected in a specific direction to obtain the moving time of the ambulance, detection time, date information, cloud server name, traffic stop time information,
In the plurality of mobile terminals, an application that provides a traffic information service is executed, and when a priority control message is received from the cloud server, the traffic section to be stopped and the section in which the vehicle object, which is an ambulance is moving, are changed in color to map Control to output on the display through the map,
The connection server includes a video camera unit that generates image data by photographing a vehicle on the road, and a gray scale conversion unit that converts RGB data, which is image data input from the video camera unit, into gray scale image data. , A Gaussian filter unit that removes noise by performing a Gaussian blur process on the gray scale image data received from the gray scale conversion unit, and an average value of pixel values for each line in the image data divided into n regions in a specific direction and the line A noise removal unit configured to filter the difference between the average values to determine whether there is random noise, calculate a corrected pixel value obtained by correcting the pixel value of the random noise generation portion to generate a corrected image from which the random noise is removed, and the noise suppressor In the correction image received from the rejection, a binary filter unit that generates a binary image using a binary filter as vehicle area 1 and non-vehicle area 0, and the luminance component data of the binary image input from the binary filter unit are canny edge detectors. Using ?O, a point having a greater change in contrast than other points, and extracting points belonging to a line formed by tracking the point to generate an edge image, and an edge image generated from the edge detector are converted into the video. The smart traffic information providing system, further comprising a mapping unit for extracting image data related to the vehicle object by mapping the original image data obtained from the camera unit. delete delete delete delete delete

Images