KR20180068247A - Parking control method for vehicle - Google Patents

Abstract

The present invention relates to a parking management method, and more particularly, to a parking management method of a vehicle based on a technique of real-time comparison of images photographed by a camera. According to an embodiment of the present invention, a vehicle parking management method for managing a vehicle entering and leaving a parking lot comprises an image photographing step of photographing an image of a parking space in the parking lot; an image storage step of storing the photographed image; an image comparing step of comparing the stored image with a currently photographed image; and a parking checking step of determining the presence or absence of the vehicle when the stored image and the currently photographed image are different from each other. Security, vehicle identification and parking control can be performed simultaneously.

Description

[0002] Parking control method for vehicle [

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a parking management method, and more particularly, to a parking management method of a vehicle based on a real-time comparison technique of images photographed by a camera.

In recent years, most of the buildings have underground parking lots to accommodate the automobiles. As the income increases, the number of cars increases rapidly, and the number of cars parked under the buildings is naturally increasing.

The problem is that as the number of parked vehicles increases, the problems to cope with the parking problems are also increasing. Specifically, when the customer can not memorize the parking position, it is possible to provide parking information in real time on the basis of the parking lot, the floor for each floor and the line for each zone, The parking lot technology is required for the current parking lot including the technology for securing the data in case of the occurrence of the accident, and the management method according to the case where the external vehicle is parked in the parking lot without permission.

Considering this, it is natural to recruit a manpower for the management of parking vehicles. However, since a large amount of expenditure has to be paid, a parking control system with automatic guidance and monitoring of incoming and outgoing cars to replace them is continuously appearing.

Especially in the recent parking control system technology, a technique of calculating the parking time and the parking charge at the time of departure is developed by photographing the license plate portion of each of the entering vehicles, recognizing the license plate, storing and analyzing the license plate in the database, And whether or not the vehicle has entered the parking line by using an ultrasonic sensor, an infrared sensor, or a loop coil to confirm whether or not the vehicle has entered the parking line.

However, the above-mentioned techniques have a disadvantage that it is possible to use another information collecting sensor in addition to the use of a plurality of cameras as a basis, and accordingly, a large amount of additional construction cost is still added to the problem .

Korean Patent Registration No. 10-0384438 (May 22, 2003) Korean Patent Registration No. 10-0869937 (November 17, 2008)

An object of the present invention to solve the problems derived from the prior art is to provide a parking control system based on vehicle tracking technology capable of simultaneously performing security security, car number recognition and parking control even with a small amount of cameras without sensor technology, .

Meanwhile, the object of the present invention is not limited to the above-mentioned objects, and other objects not mentioned can be clearly understood from the following description.

According to an embodiment of the present invention, there is provided a vehicle parking management method for managing a vehicle entering and leaving a parking lot, the method comprising: capturing an image of a parking space in a parking lot; An image storing step of storing the photographed image; An image comparison step of comparing the stored image with a currently photographed image; And a parking check step of determining the presence or absence of the vehicle if the stored image and the currently photographed image are different from each other; .

In addition, the image comparing step compares the difference between the stored image and the currently photographed image to determine whether or not the difference is greater than or equal to a certain pixel amount based on a color difference between the stored image and the currently photographed image.

Further, the parking surface is formed with a plurality of IDs, and an ID for identifying the respective areas is given.

The parking lot may include a plurality of cameras spaced apart from each other by a predetermined distance so as to take at least one or more parking planes at a time, and a control unit that receives and stores the currently photographed images in real time, And a control server for controlling the server.

In addition, the camera is characterized in that an infrared ray photographing function is included so that color difference can be discriminated even at night.

In addition, the camera may further include: vehicle number information including a number plate of the vehicle, a color of the vehicle and a vehicle size, which can be confirmed through an image taken when the vehicle enters the parking lot, To the control server.

In addition, the control server may check the parking zone of each parking surface by collating the stored image transmitted in real time with the currently photographed image, and identify each of the IDs assigned to the parking surface, And notifies the parking lot manager and the car owner of the vehicle of the vehicle information and the movement information of the other vehicle moving in the parking lot.

According to the present invention, the following effects can be expected.

First, the parking situation can be guided by the floor and the area, so that it is possible to guide the vehicle according to the free parking space which is checked in real time even when parking is congested.

Secondly, it is possible to inquire the car in case of a parking accident or a safety accident, and it is possible to promptly confirm even if the parking position is forgotten.

Third, since it can be used in conjunction with an existing camera, there is an advantage that cost for installing the camera can be greatly reduced.

Fourth, since the maximum number of parking control effect is obtained with a minimum number of cameras, there is an advantage that the efficiency of parking management can be improved.

1 is a block diagram of a parking control system based on a vehicle tracking technology according to the present invention,
FIG. 2 is a flowchart of a control method based on a vehicle tracking technology based parking control system according to the present invention,
FIG. 3 is a detailed view of the parking information transmission step of FIG.

BRIEF DESCRIPTION OF THE DRAWINGS The advantages and features of the present invention, and the manner of achieving them, will be apparent from and elucidated with reference to the embodiments described hereinafter in conjunction with the accompanying drawings. However, the present invention is not limited to the embodiments described below, but may be embodied in various different forms, and these embodiments are not intended to be exhaustive or to limit the scope of the present invention to the precise form disclosed, It is provided to inform the person completely of the scope of the invention. And the terminology used herein is for the purpose of illustrating embodiments and is not intended to be limiting of the present invention. The singular forms herein include plural forms unless the context clearly dictates otherwise.

Hereinafter, preferred embodiments of the present invention will be described in detail with reference to the accompanying drawings. Brief Description of Drawings FIG. 1 is a block diagram of a computer system according to an embodiment of the present invention; FIG. 2 is a block diagram of a computer system according to an embodiment of the present invention; FIG.

1 to 3 are views for explaining a parking monitoring system based on a vehicle tracking technology and a control method thereof according to the present invention.

2 is a flowchart of a control method based on a vehicle tracking technology based parking control system according to the present invention, and FIG. 3 is a flowchart of a control method based on a vehicle tracking technology based on a vehicle tracking technology according to the present invention. And a parking information transmitting step.

As shown in FIG. 1, the parking monitoring system based on the vehicle tracking technology that can be used as the data memory of the present invention includes an image capturing step S10, an image storing step S20, an image comparing step S30, ).

In the image capturing step (S10), an image including the parking surface, the entrance and exit roads, and the moving passage is photographed. At this time, the parking surface is formed in a large number so that a plurality of vehicles entering and exiting for parking can be maximally accommodated. An ID is assigned to each parking surface so that each area divided by the parking line can be identified and managed.

In the parking lot, there are installed a plurality of cameras installed so as to have a photographing range including at least one parking surface for photographing images. The cameras are installed to be spaced apart from each other by a predetermined distance, It is desirable to have the efficiency.

Next, an image storing step for storing the photographed image is performed. An image taken by each camera is performed through a large-capacity control server for continuously collecting, storing and storing a plurality of image data.

Thereafter, the stored image stored in the control server is subjected to an image comparison step in which it is compared with an image captured and transmitted in real time in the camera. In this case, the step of comparing the photographed images of the parking surface specifically during the image comparison step compares the difference between the stored images and the present photographed images to determine whether or not they are greater than or equal to a predetermined pixel amount. When there is a difference between the compared images, a parking confirmation step for determining the presence or absence of the vehicle is performed.

Here, the image comparison step and the parking confirmation step may be regarded as a single step. In addition to the determination of the parking zone identified in the present invention, the main information necessary for the management of the parking is collected and the parking efficiency of the parking lot is further increased through the information It is possible.

At this time, the parking guide display board can be installed in the parking lot entrance and the ceiling of the moving passage so that it is easy for the car owner of the car using the parking lot to easily confirm the result of the two steps of the control server in the parking lot. Specific functions are described later.

Returning again to the image comparison step and the parking confirmation step, the camera displays the vehicle's license plate number, the vehicle information including the color of the vehicle and the vehicle size, and the image captured after entering the parking lot of the vehicle, And transmits the movement information of the vehicle which can be confirmed through the control server to the control server. At this time, a method of recognizing the license plate which is considered to be the most important among the vehicle information will be described.

There is a disadvantage in that the existing license plate type discrimination method is sensitive to the distortion caused by the illumination, because the license plate recognition technique including the license plate type discrimination method is utilized in various fields such as Yosai, parking and management of the car entering and leaving, and crime prevention.

In order to compensate for this, we propose a method to correctly identify the license plate type even in darkened license plates by using the ratio of color coordinate components in license plate background. Experiments were carried out by obtaining the color coordinates of the backlit plates, and it was confirmed that the proposed method was robust to the influence of the illumination because the ratio of the color coordinates of the green plate was larger. Therefore, it can be effectively used in an environment susceptible to illumination.

Particularly, the method of discriminating the license plate type in the license plate recognition process includes a method using the position of a numerical column, and a method according to the pixel frequency of letters. When the type of the plate is determined by using the position of the number column, since the position of the number column does not greatly differ according to the type of the plate, all the column of numbers must be detected at the correct position. In addition, the method using the pixel frequency of letters requires accurate separation of letters and license plate backgrounds, and is sensitive to distortion caused by illumination.

In order to compensate the above points, an algorithm for discriminating the license plate type that can accurately discriminate the license plate type even if the license plate color is changed by illumination is as follows.

First, an edge is extracted through a preprocessing process on an input image, and then a process of detecting the license plate and the type of the license plate is performed by using the size, the aspect ratio, and the number of the previously acquired segments of the license plate candidate region formed by the edge. After one more segment acquisition process according to the license plate type, the result of the recognized character is output in each segment.

The edge extraction process for license plate recognition is performed by performing a Canny edge operation after applying a DoG (Differenceof Gaussian) filter. DoG filters can reduce false positives because they are strong against noise and can extract sharp edges. When the Canny edge operation is performed on the same image, there is a strong point that the image to which the DoG filter is applied has less noise than that of the non-DoG filter.

 After repeatedly performing morphology operations such as erosion and expansion on the extracted image of the edge, the outline is detected and the rectangles formed by each outline are selected as the license plate candidate regions. Then, the final plate is detected by using the width, the aspect ratio, or the number of segments obtained inside the candidate region. In addition, the number of intersections with the vertical line is calculated when the inside of the license plate is projected in the horizontal direction, and if it does not exceed the certain standard, it is excluded from the license plate candidate.

Next, the type of vehicle license plate is determined. Currently, there are four types of vehicle license plate: green sphere, green new sphere, white sphere and white sphere. In this case, it is possible to identify the type of the license plate by using the characteristics of each type, and to perform different binarization accordingly, thereby improving the error recognition rate of characters.

At this time, the four kinds of license plates can be divided into two types of license plates, that is, green license plates and white license plates. The license plate classified by color is again classified into new type and old type. In the case of a green plate, the number of segments is divided into a spherical shape when the number of segments is 8 or more, and a new shape when the number of segments is less than that of a segment.

In this way, the segment acquisition process is performed one more time after the license plate type is discriminated, so that the character recognition can be accurately performed by applying the binarization differently according to the license plate type. Apply binarization for green plates and reverse binarization for white plates to treat letters inside plates as white and backgrounds as black. Also, binarization is applied by varying the threshold value according to the type of the license plate and the font size, thereby preventing the bundling of characters.

 From the obtained segments, the character recognition is performed by comparing the standard character standard image according to the license plate type and selecting the character having the highest degree of similarity. In the binarized segment image, the degree of similarity is determined by assigning a weighting to the case in which the black background and the white character are the best match, the case in which the characters are the closest match, and the case in which the characters are in the closest match.

In this way, it is possible to effectively recognize the license plate even in an environment that is highly influenced by illumination, but the camera can more accurately confirm the vehicle information by including the infrared ray photographing function so that the color difference can be discriminated even at night .

Next, a method of predicting the moving path of the vehicle will be described. The map of each floor constituting the parking lot is divided into a tile map starting from at least four and increasing in number by 4 N power units as accumulation is reduced to 1/2, . That is, the map divided into four is stored in the map DB, and the control unit reads the map of the area where the camera where the moving vehicle is photographed is read from the map DB and displays it on the parking guide display board as in the enlarged part.

At this time, if the camera in which the vehicle is photographed exists over a plurality of locations, the tile map of the corresponding parking layer is loaded and displayed in the map DB.

Then, the control server analyzes the image signal of the camera taken of the vehicle, and predicts the traveling direction of the vehicle using the direction and speed when the problem vehicle moves from the first area to the second area.

After the above analysis, the moving route of the vehicle is displayed on the map. In addition, images photographed by a camera installed in the moving path of the vehicle are sequentially displayed on the image window.

When two or more vehicles are photographed, the image of the vehicle, the photographing position, and the photographing date and time are displayed on the map in the order of comprehensive consideration of the size of the vehicle or the safety accident in the parking lot caused by the vehicle in the vehicle information stored in the control server . That is, the vehicle having the highest risk level is selected based on the information stored in the control server, and the image of the selected vehicle is displayed in the image window. The moving route of the vehicle and the sequentially photographed images are displayed in the continuous image window.

In addition, when a problematic vehicle causing a safety accident at a predetermined position is photographed, the camera installed in the photographing region is set around the corresponding position so that the camera is operated by the camera of the control server Preset. In other words, you can adjust the direction of each camera by adjusting the direction of the camera displayed on the map. The images taken by each camera are displayed in the continuous image window, and the problem vehicle is automatically photographed along the traveling direction.

Here, when a problem vehicle having a contact accident or the like occurs, a notification message indicating the occurrence of the problem vehicle is displayed on the map, and the occurrence position of the problem vehicle is displayed on the map. The control server reads the map of the floor where the problem vehicle is generated from the map DB and displays it on the parking guide display board.

Previously, the problem vehicle is displayed in a different color, and the problem vehicle list includes the vehicle number and the shooting time of the problem vehicle so far. An image of the problem vehicle is also displayed together with an image window (hereinafter referred to as an image window) of the vehicle provided under the map of the parking guide display board.

With the above-described configuration, it is possible to predict the moving path of the vehicle, and when a problem vehicle occurs, the problem vehicle can be automatically photographed by predicting the traveling direction of the problem vehicle.

Now, the control server verifies the parking zone of each parking surface by collating the stored image transmitted in real time with the currently photographed image through the above-described processes, and not only can the information of the entering vehicle for parking, It is possible to derive parking information by collecting and collecting various information including movement information that can prevent a safety accident including vehicle-to-vehicle contact.

Thereafter, the IDs assigned to the parking surface in the control server, the vehicle information parked on the parking surface, and the movement information of the other vehicle moving in the parking lot are notified to the parking lot manager and the parking lot notice of the car More steps can be added.

The information notification method can be variously formed. In addition to a map in which the respective layers formed for predicting the travel path of the vehicle are miniaturized through the parking guide electric sign board and the travel path prediction of the vehicles running, It is possible to display simple personal information such as presence or absence of a vehicle, a vehicle number which can be obtained through a license plate recognition algorithm, and an emergency contact which can be confirmed through the vehicle number, It is possible to guide the position according to the ID.

Furthermore, it is also possible to provide a wireless communication device capable of grasping the parking status of the parking lot in a car which is not parked yet, and to guide the parking path that can be parked the fastest among the routes by grasping the route of the vehicle detected by the control server will be.

In addition, the efficiency of parking can be sufficiently increased even before entering the parking lot by using a parking-related application capable of grasping the parking status of the parking lot by a personal portable terminal such as a smart phone.

 The foregoing has outlined rather broadly the features and technical advantages of the present invention in order to better understand the scope of the appended claims. Those skilled in the art will appreciate that the present invention may be embodied in other specific forms without departing from the spirit or essential characteristics thereof. It is therefore to be understood that the above-described embodiments are illustrative in all aspects and not restrictive. The scope of the present invention is defined by the appended claims rather than the detailed description, and all changes or modifications derived from the claims and equivalents thereof are to be construed as being included within the scope of the present invention.

S10:
S20: Image storage step
S30: Image comparison step
S40: Parking confirmation step

Claims (7)

A vehicle parking management method for managing a vehicle entering and leaving a parking lot,
An image capturing step of capturing an image of a parking surface in a parking lot;
An image storing step of storing the photographed image;
An image comparison step of comparing the stored image with a currently photographed image; And
A parking check step of determining the presence or absence of the vehicle when the stored image and the currently photographed image are different;
Wherein the vehicle parking management method comprises the steps of:
The method according to claim 1,
The image comparison step may include:
Comparing the difference between the stored image and the currently photographed image to determine whether or not the difference is greater than or equal to a predetermined pixel amount based on a color difference between the stored image and the currently photographed image.
The method according to claim 1,
Wherein the parking surface comprises:
And a plurality of IDs for identifying the respective areas are provided.
The method according to claim 1,
The parking lot
A plurality of cameras spaced apart from each other by a predetermined distance so as to collect at least one or more parking faces,
And a control server for receiving and storing the currently photographed images in real time in the camera and checking the stored images with the stored images.
5. The method of claim 4,
The camera comprises:
And an infrared ray photographing function is included so that color difference can be discriminated even at night.
5. The method of claim 4,
The camera comprises:
Vehicle information including a number plate of the vehicle which can be confirmed through the image taken when the parking lot of the vehicle enters the parking lot, vehicle color and color of the vehicle,
Wherein the management server transmits the movement information of the vehicle that can be confirmed through the image taken after entering the parking lot of the vehicle to the control server.
The method according to claim 6,
The control server,
A method of checking a parking zone of each parking surface by collating the stored image transmitted in real time with the currently photographed image,
Wherein the controller informs the parking lot manager and the car owner of the vehicle of the respective IDs assigned to the parking surface and the vehicle information parked on the parking surface and the movement information of the other vehicle moving in the parking lot How to manage.

Images