KR20180055083A - Management apparatus of parking spaces - Google Patents

Abstract

A parking space management system according to an embodiment of the present invention includes a photographing unit installed on a ceiling of a parking lot and photographing a parking space partitioned at a lower portion to generate image information, A vehicle identification module for extracting license plate image information of the detected vehicle from the parking surface image information and recognizing the vehicle number from the extracted license plate image information, A storage module that stores information on the parking lot surface state of the parking space and the recognized car number information, and a communication module that transmits the parking lot status information and the recognized car number information to the connected external terminal, And a control unit for controlling the main server and the main server, And a display unit for displaying the beam.

Description

{Management apparatus of parking spaces}

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a parking space management system, and more particularly, to a parking space management system using an image acquired by a camera installed on a ceiling of a parking lot, To a parking space management system capable of performing parking space management.

In large buildings, apartment complexes, etc., there is a large parking lot that can accommodate many vehicles for the convenience of tenants or visitors. As the parking lot becomes larger, it often happens that it is difficult to find the parking space or to find the position of the car that you parked. If you neglect to look ahead to find a place to park, an accident may occur. In other words, after entering a large parking lot, they have to go around to find a parking space. In some cases, when a person enters a parking space quickly to occupy a parking space, a contact accident may occur.

In general, a large parking lot is divided into pillars or walls, which helps the driver to locate the vehicle. In recent years, there is a system for detecting whether the vehicle is empty or not by detecting the vehicle at the top of the parking space.

One example of a parking management system according to the related art is Korean Registration No. 10-0999522 and Korean Patent Publication No. 10-2011-0020384. Such a technique is a method of determining whether a vehicle is parked in a parking space in various ways By detecting and displaying this information, it is possible to efficiently manage the parking lot and to allow the user to easily find an empty parking space. However, the conventional techniques have the disadvantage that the vehicles entering the parking lot must find the empty space by themselves.

Therefore, it is necessary to develop a parking lot management system that can reduce the number of cameras installed in the parking lot and simplify the facilities required for mounting the camera.

It should be understood that the foregoing description of the background art is merely for the purpose of promoting an understanding of the background of the present invention and is not to be construed as adhering to the prior art already known to those skilled in the art.

SUMMARY OF THE INVENTION The present invention has been conceived to solve such problems, and it is an object of the present invention to provide an image pickup apparatus, And a method of managing the parking area.

A parking space management system according to an embodiment of the present invention includes a photographing unit installed on a ceiling of a parking lot and photographing a parking space partitioned at a lower portion to generate image information, A vehicle identification module for extracting license plate image information of the detected vehicle from the parking surface image information and recognizing the vehicle number from the extracted license plate image information, A storage module that stores information on the parking lot surface state of the parking space and the recognized car number information, and a communication module that transmits the parking lot status information and the recognized car number information to a connected external terminal, And a control unit for controlling the main server and the main server, And a display unit for displaying the beam.

The vehicle recognition module detects a license plate candidate region through machine learning using the image information photographed by the photographing section, selects four vertexes of the license plate candidate region, derives the license plate region through slope image transformation, It may be to recognize the car number.

The binarization method may be an Ostu binarization method (Otsu's Threshold).

And a car number recognizer connected to the main server and recognizing the number of the entering vehicle installed at the car park entrance and transmitting the car number through the communication module.

The main server may further include a control module for controlling the blinking of the dimming lamps.

And a resident terminal connected to the communication module to display the vehicle status information of the parking space and the recognized car number information.

And a driver terminal installed in the entering vehicle and connected to the communication module to display vehicle status information on the parking surface and the recognized car number information to the driver.

Wherein the main server further comprises a calculation module for calculating an optimal parking area by using information of a parking area according to a car number and information of a car holder through information stored in the storage module, It may be to provide optimal parking area information to the entering vehicle driver.

And a parking guide panel installed at the entrance of the parking lot and connected to the communication module to display the parking status of the parking lot.

According to the parking face management system of the present invention, it is possible to directly detect the entrance / exit of the vehicle, recognize the car number, and guide the parking position without any additional screen correction.

In addition, it is possible to guide the driver to the optimal parking surface for entering the car, or to guide the parking position at the time of departure, so that the parking can be conveniently used.

1 is a view schematically showing a photographing method by a camera in a parking space management system according to an embodiment of the present invention.
2 is a block diagram of a parking space management system according to an embodiment of the present invention.
FIG. 3 is a diagram illustrating a state in which an image photographed by a photographing unit according to an exemplary embodiment of the present invention derives a license plate candidate region through machine learning.
4 is a diagram illustrating a number recognition process according to an embodiment of the present invention.
5 is a diagram illustrating a number recognition process according to an embodiment of the present invention.
FIG. 6 is a view showing a screen displaying a photographed image, parking availability, and car number information displayed on a display unit according to an embodiment of the present invention.
FIG. 7 is a view illustrating a screen displaying a photographed image, parking availability, and car number information displayed on a residential terminal according to an embodiment of the present invention.

The terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the invention. The singular forms as used herein include plural forms as long as the phrases do not expressly express the opposite meaning thereto. Means that a particular feature, region, integer, step, operation, element and / or component is specified, and that other specific features, regions, integers, steps, operations, elements, components, and / And the like.

Unless otherwise defined, all terms including technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs. Commonly used predefined terms are further interpreted as having a meaning consistent with the relevant technical literature and the present disclosure, and are not to be construed as ideal or very formal meanings unless defined otherwise.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, a parking space management system according to a preferred embodiment of the present invention will be described with reference to the accompanying drawings.

1 is a view schematically showing a photographing method by a camera in a parking space management system according to an embodiment of the present invention. As shown in FIG. 1, in a parking surface management system according to an embodiment of the present invention, a high-resolution camera is installed on a parking lot ceiling so that the inside of the parking lot can be photographed. It is possible to acquire the divided image information of the parking faces of both sides through the installed camera and to use this information to judge the state of the parked vehicle on the parking surface and the parked vehicle number.

FIG. 2 is a diagram illustrating a configuration of a parking space management system according to an embodiment of the present invention. 2, a parking space management system according to an embodiment of the present invention includes a photographing unit 10, a main server 20, a display unit 30, a car number recognizer 40, A residential terminal 60, a car terminal 70, and a parking guide sign 80. The residential terminal 60,

In the present invention, the photographing unit 10 may be installed in a parking lot ceiling to photograph a parking space partitioned at a lower portion thereof to generate image information. The main server 20 carries out vehicle detection for grasping the state of the parked vehicle on the basis of the image information generated by the photographing unit 10 and recognizes the vehicle number to extract the vehicle number of the parked vehicle. In addition, the user can check the image information collected in the photographing unit 10 through the display unit 30 in real-time streaming.

The photographing unit 10 may include a high-resolution CCD camera so as to photograph a plurality of parking faces on both sides of an incoming and outgoing car. In the case of photographing a parked face with a camera, it is possible to secure both of the parking faces on both sides of the photographing range by using a camera using a wide-angle lens.

Basically, the part that can be photographed by the camera can be photographed at a wide angle including the entrance and exit lane and the parking area. The photographed image is not distorted by itself and can be used immediately without any restoration work.

Assuming that there are three parking spaces between the pillars of the building, the parking spaces that can be photographed by the photographing unit 10 can be photographed in total of six, three on both sides of the entrance and exit. The number of parking faces that can be photographed may vary depending on the photographing angle of the camera and the number of parking spaces between the columns. Considering this, the total number of the photographing units 10 installed on the ceiling of the parking lot can be determined.

The main server 20 not only recognizes the availability of the parking space, the car number, records the information, but also transmits and receives information with other equipment connected to the outside using the image information acquired by the photographing unit, As shown in FIG. To this end, the main server 20 includes a detection module 21, a vehicle recognition module 22, a calculation module 23, a storage module 23, a communication module 25, and a control module 26.

The detection module 21 determines the fullness or the tolerance state of the parking surface through the image information generated by the photographing unit 10 by photographing. The detection module 21 detects the vehicle by applying a vehicle detection algorithm to the parking plane image information. Here, the vehicle detection algorithm is an algorithm for determining whether or not the vehicle is parked. Therefore, a motion detection algorithm for detecting the motion of the vehicle or an image detection algorithm for comparing the background image of the parking plane with the parking image can be used. Both algorithms can be used for more accurate vehicle detection. The parking status of the parking space is determined through the parking status information of each parking space. For example, it is possible to check whether or not the parking space is full by comparing brightness of the photographed parking area image with that of the reference parking area after removing the sleeping sound.

The vehicle recognition module 22 uses the image information generated by the photographing section 10 to recognize the vehicle number of the parked vehicle. The image information may be at least one still image information corresponding to a still image among a plurality of frames of the photographed moving image. The vehicle recognition module 22 recognizes the vehicle number of the parked vehicle through the number recognition algorithm. The number recognition algorithm may be a method of recognizing text information from image information having a car number in the parking plane image information. As a matter of course, only when the entrance of the parking space of the vehicle is confirmed as a result of the detection of the detection module 21, the vehicle number is recognized. Details of the car number recognition process will be described later.

The storage module 24 is provided with a memory for storing the photographed image information, the parking lot state information of the parking space, and the car number information, and the information stored in the memory can be set to be accessible at any time. The memory included in the storage module 24 may be a flash memory type, a hard disk type, a multimedia card micro type, a card type memory (for example, SD or XD memory, etc.), a random access memory (RAM), a static random access memory (SRAM), a read-only memory (ROM), an electrically erasable programmable read-only memory (EEPROM), a programmable read- ), A magnetic memory, a magnetic disk, and / or an optical disk.

The communication module 24 is for transmitting and receiving necessary information to an external device connected to the main server, and can be connected in a wired or wireless manner. WLAN (Wi-Fi), Wibro (Wireless broadband), Wimax (World Interoperability for Microwave Access), WCDMA (Wideband Code Division Multiple Access), HSDPA High Speed Downlink Packet Access), LTE (Long Term Evolution), and the like. The information transmitted through the communication module 24 may be the image information photographed through the photographing unit 10, the state of the vehicle on the parking lot, and the car number information. Or may be the optimal parking area information selected through a calculation module described later.

The calculation module 23 calculates the optimum parking place on the basis of the image information stored in the storage module, the state of the parking lot on the parking lot, the car number information, the parking time, and the parking time. For example, it is possible to sequentially determine the parking side closest to the closest entrance from the place where the vehicle user resides, to determine whether the parking side is full or not, and calculate the optimum available parking side. As another example, a parking lot that is frequently used by a vehicle user is selected as a preferred parking lot, and a big data processing method based on the selected parking lot is used to check whether the parking lot is the most frequently parked, It is also possible to do.

The control module 26 controls the pneumatic differential 50 installed in the photographing unit 10 among the external devices connected to the main server and displays the state of the paved vehicle on the parking surface.

The display unit 30 can display the image information photographed through the photographing unit 10, the vehicle state of the parking lot on the parking surface, and the car number information through the information stored in the storage module 24. [ The parking lot manager can confirm the current state of the parking lot, the current state of the entrance and exit car, etc. through the display unit 30. Also, it is possible to check the image acquired by the photographing unit 10 through real-time streaming, or to store a real-time image through a self-storage device.

The display unit 30 may be a liquid crystal display (LCD), a thin film transistor liquid crystal display (TFT LCD), an organic light emitting diode (OLED), a flexible display Display, and a three-dimensional display (3D display).

The car number recognizer 40 can be connected to the communication module 25 of the main server 20 in a wired or wireless manner and recognizes the number of the vehicle installed or departing from the parking lot entrance or exit and transmits it to the main server 20 ). It is a device that recognizes a car number through a still image captured by a camera when entering the car, and it is possible to make accurate entry and exit by comparing the recognized car number with the registered car number when entering and leaving the car.

The pneumatic differential 50 outputs an indication to the driver who enters the parking lot to easily tell whether there is a parking space available. The pneumatic differential 50 is installed on the upper part of the photographing unit 10 and can be configured to display the state of the paved vehicle on all sides of 360 degrees. The LED 50 is mounted inside the pavilion 50, .

The residential terminal 60 is connected to the main server 20 via the communication module 25 and is installed in a residence of a parking lot such as an apartment, And the parking lot number, and the parked car number information. Through the residential terminal, the user of the vehicle can confirm on which parking side the vehicle is parked. Also, it is possible to confirm the entry time and parking time of the vehicle. The residential terminal may be, for example, a fixed terminal such as a month pad installed in an apartment or a mobile terminal such as a smart phone or a tablet PC.

The vehicle terminal 70 is connected to the main server and can display image information photographed through the photographing unit, the state of the vehicle on the parking lot, and the car number information. In addition, it is possible to receive and display an optimal parking surface calculated from the recognized car number at the time of entry from the main server.

The parking guide sign 80 is a display device which is installed at the entrance side of the parking lot and displays the number of parking spaces of the parking space in the parking space. The display device 80 is connected to the communication module 25, The number of parking spaces in each zone is calculated through the calculation module 23 and received and displayed. This information allows the driver of the vehicle to obtain information on the available parking area at the time of entry.

FIG. 3 is a diagram illustrating a state in which an image photographed by a photographing unit according to an embodiment of the present invention derives a license plate candidate region through machine learning. Referring to FIG. 3, the range that can be photographed by the photographing unit 10 can be photographed including an entrance / exit lane, a parking plane, and the like as shown in FIG. The obtained image information can be stored on the parking lot surface state and parked vehicle number information through the main server. When the information received through the car number recognizer is combined, it is possible to calculate the information such as the entrance and exit car status, the parking lot car status, the parked car number information, and the parking position by car number through the calculation module 23 . In order to recognize the car number, the license plate recognition candidate area is firstly designated. In the present invention, it is possible to designate the license plate recognition area through machine learning. Machine learning is an algorithm that predicts results for new data based on learned knowledge through training data. As shown in FIG. 3, the portion indicated by orange in the image acquired through the photographing portion corresponds to a license plate candidate region to be determined to be positive, and a person, a mark, a column and the like are discriminated as negative. It corresponds to a false image. In this way, it is possible to derive a license plate candidate area for new image data.

The number recognition algorithm detects the license plate area from the license plate candidate area, normalizes the license plate size, performs image enhancement of the license plate, detects the number area of the license plate number, and recognizes the number plate. First, four vertexes are manually selected to detect the license plate area in the portion photographed by the photographing unit 10. [ You can set the license plate area to fit the license plate through the tilt image. At this time, approximate to a rectangle through projection transformation. In order to improve the image of the license plate, the noise of the license plate is removed and the brightness is improved. Sharpness can be improved through high frequency emphasis in order to improve sharpness. After the image segmentation through the binarization method, the machine learning based number recognition algorithm is performed.

In image processing, binarization refers to changing all pixels that are lighter than a given threshold to white, and all pixels that are not. The optimal threshold is called T, which minimizes the sum of the ratio of the background pixels that are darker than T and the ratio of object pixels darker than T when one T value is fixed. However, this does not suggest how to obtain the actual threshold value, only the theoretical goal and the point of orientation. To know if T is optimal, it is necessary to know in advance which pixel is the object and the background. But if you know this in advance, you do not have to get a T because you already have optimal binarization. Otzu's method is the most practical method to find the brightness distribution of the input image and find the threshold value. Otsu's binarization method is a binarization method that minimizes the intra-class variance between two classes or maximizes the inter-class variance when classifying the image pixels into two classes based on the threshold T.

When the ratio of pixels darker than the threshold value T in the input image is α, the brightness average is μ 1, the variance is σ ₁ ² , the ratio of pixels lighter than T is β, the brightness average is μ 2 and the variance is σ ₂ ² , β = 1), intra-class variance and inter-class variance are calculated as follows.

intra-class variance = ασ ₁ ² + βσ ₂ ² (1)

inter-class variance = αβ (μ1-μ2) ² (2)

Minimizing Eq. (1) is equivalent to maximizing Eq. (2). Either criterion can be used, but it is efficient to maximize Eq. (2) in the calculation. That is, Otsu's binarization algorithm is a method of binarizing the image by calculating the equation (2) while gradually changing the threshold value T from 0 to 255 and finding the T value that maximizes the equation (2).

Machine learning is an algorithm that predicts results for new data based on learned knowledge through training data. Therefore, it is possible to acquire information on new data based on the recognized number information.

5 and 6 illustrate a number recognition process according to an embodiment of the present invention. Referring to FIGS. 5 and 6, the number recognition process will be described. The number plate candidate region is detected from the photographed image. License plate 4 Specify the initial candidate point for the vertex and designate the license plate area through the tilt image. The slope of an image is the difference in brightness between adjacent pixels. In the region where the erection difference is small in the designated region, it appears dark in the slope image, and in the region where the erection difference is large, the number and the border portion are bright in the slope image. If a license plate area is specified, it is transformed through the projection transformation method. When the conversion is completed, the part requiring recognition is designated, and the sharpness of the image is improved through high frequency emphasis. Finally, the license plate is separated and recognized through binarization.

FIG. 7 is a view showing a screen displaying a photographed image, parking availability, and car number information displayed on a display unit according to an embodiment of the present invention. The main server 20 can grasp the parking lot status through the display unit. Basically, it is possible to confirm whether the vehicle number is entered or not, and the image information of the parked vehicle on each parking surface can be displayed. In addition, not only the vehicle number can be confirmed visually through each image information, but also the vehicle number recognized by the vehicle recognition module 22 can be confirmed. When a specific vehicle number vehicle is selected, the image information of the vehicle and the vehicle parking position can be confirmed.

FIG. 8 is a view showing a photographing image displayed on a resident's terminal according to an embodiment of the present invention, parking availability, and car number information. As a residential terminal as shown in FIG. 6, a wall pad can be used, and image information and a column number in a parked state of the vehicle can be displayed on the wall pad, and a resident can confirm the entry details, It is possible.

Since the acquired image information can be directly used without a separate correction operation as in the present invention, a separate apparatus is not provided, so that the apparatus can be configured more simply and cost can be reduced.

In addition, it is possible to confirm the parking position through the resident terminal at the time of departure, to be able to leave immediately, and to receive the optimum parking position through the vehicle terminal at the time of entry, so that it is possible to park conveniently.

While the present invention has been described in connection with what is presently considered to be practical exemplary embodiments, it is to be understood that the invention is not limited to the disclosed embodiments, but, on the contrary, You will understand.

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 meaning and scope of the claims and their equivalents should be interpreted as being included in the scope of the present invention .

10: photographing section 20: main server
30: display unit 40: car number recognizer
50: Permanent differential 60: Residential terminal
70: vehicle terminal 80: parking guide sign

Claims (9)

A photographing unit installed on a ceiling of a parking lot and photographing a parking space partitioned at a lower portion to generate image information;
A sensing module connected to the photographing unit to determine a state of fullness or tolerance of the parking space in the photographed image information, and a controller for extracting license plate image information of the detected vehicle from the parking space image information, A vehicle identification module for recognizing the vehicle number, a storage module for storing the photographed image information, the vehicle state information on the parking surface, and the recognized vehicle number information, and the vehicle state information and the recognized vehicle number information are connected A main server including a communication module for transmitting to an external terminal; And
And a display unit connected to the main server and displaying the photographed image information, the state information of the parking space on the parking surface, and the recognized vehicle information.
The method according to claim 1,
The vehicle recognition module detects a license plate candidate region through machine learning using the image information photographed by the photographing section, selects four vertexes of the license plate candidate region, derives a license plate region through slope image transformation, And the vehicle number is recognized.
The method of claim 2,
Wherein the binarization method is an Ostu binarization method (Otsu's Threshold).
The method according to claim 1,
And a car number recognizer connected to the main server and recognizing the number of the entering vehicle installed at the car park entrance and transmitting the car number through the communication module.
The method according to claim 1,
Wherein the main server further comprises a control module for controlling the blinking of the dimming lamps, wherein the main server further comprises a control module for displaying a parking status of the parking surface on an upper portion of the photographing portion.
The method according to claim 1,
And a resident terminal connected to the communication module to display the vehicle state information on the parking surface and the recognized car number information,
The method according to claim 1,
Further comprising a driver's terminal installed in the entering vehicle and connected to the communication module to display vehicle status information on the parking surface and the recognized car number information to the driver.
The method of claim 5,
Wherein the main server further comprises a calculation module for calculating an optimal parking area by using information of a parking zone according to a car number and information of a car holder through information stored in the storage module,
And provides the optimal parking area information to the driver of the vehicle entering the parking lot through the driver's terminal.
The method according to claim 1 or 2,
And a parking guide panel installed at the entrance of the parking lot and connected to the communication module to display the parking status of the parking lot.

Images