WO2023075015A1

WO2023075015A1 - Smart parking system based on intelligent image analysis

Info

Publication number: WO2023075015A1
Application number: PCT/KR2021/019012
Authority: WO
Inventors: 정철호
Original assignee: (주)제이엔디
Priority date: 2021-10-27
Filing date: 2021-12-14
Publication date: 2023-05-04
Also published as: KR20230060562A

Abstract

A smart parking system based on intelligent image analysis, of the present invention, comprises a parking determination unit for determining whether, in images acquired from CCTV, vehicles are parked or can be parked on respective parking surfaces of a parking lot, by using a model set in order to determine whether the vehicles are parked or can be parked on the parking surfaces on the basis of analysis of the images.

Description

Smart parking system based on intelligent video analysis

The present invention relates to a smart parking system based on intelligent video analysis.

As of 2018, the number of registered vehicles in Korea is 23.2 million, and it is increasing at an average annual rate of about 3%. Even though new buildings or shopping malls are considering parking spaces, parking spaces are still insufficient, and parking problems are always agonized when moving from place to place. In particular, illegal parking is prevalent in downtown areas due to lack of parking spaces, and fights over parking are common.

Although parking spaces that are parked 24 hours a day in a parking space are rare, it is difficult to identify vacant parking spaces, and research on services for integrated management of them continues, but it is difficult to integrate a lot of data.

Currently, parking guidance systems are applying technologies that count and guide available parking spaces based on sensors. Since these systems require sensors to be provided on all parking surfaces, the initial installation cost is very high and the sensors are damaged by vehicles. Sensors that are worn or installed outdoors have a problem in that durability is deteriorated due to poor environments such as rain and snow.

As a result, installation and maintenance costs can be reduced, and it is highly reliable and efficient even outdoors or in poor environments. There is a need for a parking system that can solve all of the difficult problems and the problem of sharing empty parking spaces.

An object of the present invention is not only to reduce installation costs and maintenance costs, but also to be highly reliable and efficient even outdoors or in poor environments, and to solve the problem of parking vehicles other than disabled vehicles in disabled parking areas, narrow parking for large vehicles such as campers. It is to provide an intelligent video analysis-based smart parking system that can solve both the problem of difficult parking on the side and the problem of needing to share an empty parking surface.

The above and other objects of the present invention can all be achieved by the present invention described below.

One aspect of the present invention relates to a smart parking system based on intelligent video analysis.

According to one embodiment, the intelligent video analysis-based smart parking system includes an image acquisition unit for acquiring an image of a parking lot from a CCTV; a parking sensing unit provided on a parking surface of the CCTV image blind spot to determine whether the vehicle is parked; And a parking management server; including, wherein the parking management server, based on the image of the image acquisition unit, a video analysis unit for generating analysis data by analyzing the parking area and the vehicle area; A parking determination unit for determining whether or not a parking surface in the image is parked or not, using a model set to determine whether each parking surface of the parking lot is parked or not, based on the analysis data; a parking information management unit that collects and manages parking information including the number of available parking spaces from the parking determination unit and the parking sensing unit; a parking information providing unit providing the parking information to a parking lot user; And a communication unit for communicating with the CCTV image acquisition unit or a server for managing the video of the CCTV image acquisition unit, the parking sensing unit, and a user terminal through a communication network; the model includes, for the image and analysis data of the parking lot As a model created by learning using an artificial intelligence algorithm, it is a model created by learning standard information on whether or not a vehicle is parked and whether or not a vehicle is parked on each parking surface of a parking lot image.

The learning may include one or more of machine learning and deep learning.

The parking sensing unit may include a sensor for determining whether a parking surface of the CCTV image blind spot is parked; And a communication unit for transmitting the parking information sensed by the sensor to the parking management server; may include.

The parking management server includes a user authentication unit authenticating the parking lot user; a user management unit managing information of the authenticated user; Reservation management unit for receiving a parking reservation from the parking lot user through a client program or the web; a fee calculation unit for calculating a parking fee for the parking lot user; and a fee payment unit for paying the parking fee. One or more of them may be further included.

The user authentication unit may authenticate the user by a method of inquiring at least one of the user's personal information, the user's reservation time, and the type of vehicle.

The user terminal includes a communication unit for communicating with the parking management server or the parking sensing unit through a communication network; A program unit that downloads and installs a client program related to a parking information providing service from the parking management server, executes the client program to authenticate a parking lot user, or authenticates a parking lot user by accessing the web and receives the parking information; an input unit for the authenticated user to input reservation information including a reservation parking lot, vehicle type, and license plate number through a client program or the web; a display unit for displaying on a screen information including at least one of parking information of each of the parking lots, a reservation time, and a reservation cancellation scheduled time; a payment unit for paying a parking fee; And a control unit for controlling at least one of communication with the parking management server or parking sensing unit, execution of the client program, input of reservation information, and display of information; One or more of them may be included.

The image analysis unit extracts a parking area and a vehicle area from the images collected from the image acquisition unit, calculates the overlapping ratio (R) of each parking area and an adjacent parking area, and determines whether the parking area of the parking surface to be judged is the adjacent vehicle. The overlapping ratio (P) with the region can be calculated and stored as analysis data.

The parking determination unit may determine whether the parking surface is parked or not by applying a predetermined criterion based on the analysis data.

The preset criterion may be Equation 1 below.

[Equation 1]

P < R × A

(In Equation 1, P is the overlapping ratio of the parking area of the parking surface to be determined with the adjacent vehicle area, R is the overlapping ratio of the parking area of the parking surface to be determined with the adjacent parking area, A is 0.317 to 0.517, If Equation 1 is satisfied, it is determined as an available parking surface, and if Equation 1 is not satisfied, it is determined as an unavailable parking surface).

The model determines that the corresponding parking surface is a non-parking parking surface by Equation 1, learns when a vehicle is parked on the parking surface, corrects the A value downward, and determines that the corresponding parking surface is parkable parking by Equation 1. If the vehicle is determined to be a surface and the vehicle is not parked on the parking surface for a predetermined period of time or more, the B value may be corrected downward by learning this.

The upward correction and the downward correction may be corrected by (P/R)×x.

The CCTV may further include an infrared camera to transmit the captured image to the image analysis unit of the parking management server, and the parking determination unit may apply the image of the infrared camera to the analysis of the vehicle area.

The parking management server further includes an address extraction unit for extracting address information of a place where the CCTV is photographed and address information of the parking sensing unit, and the parking management server provides parking information further including the extracted address to the user. The user terminal may provide a navigation function by linking the extracted address with a client program or a map on the web.

The parking sensing unit may include a sensor for determining whether a parking surface of the CCTV image blind spot is parked; a communication unit that transmits parking information sensed by the sensor to a parking management server, receives an authentication code of a reserved user from the parking management server, and communicates with a user terminal; Parking prevention unit capable of preventing parking on the corresponding parking surface; a driving unit for driving the parking prevention unit; And a control unit for controlling the sensor, communication unit and driving unit; and, the reservation management unit of the parking management server transmits the smart key including the authentication code to the terminal of the user who has made the reservation at the reserved time, and the control unit of the parking sensing unit may use the authentication code of the user terminal to form a wireless link with the parking sensing unit and control the driving unit.

The parking management server receives reservation information from the user terminal, and the reservation information includes one or more of a parking lot to be parked, a vehicle type, and a vehicle number, and provides the reserved parking lot within a specific time from the input time of the reservation information. If there is no entered information, the reservation may be canceled.

The entered information is transmitted in conjunction with the parking management system when the parking lot has a parking management system, and when the parking lot does not have a parking management system, the parking management system is provided using the authentication code of the user terminal. It may be determined whether or not a wireless link is formed with the sensing unit.

The number of available parking spaces may be calculated by subtracting the number of reserved users.

If the parking lot has a parking management system, the parking management server shares user information reserved in the parking management system in conjunction with the parking management system, and the parking management system can park in consideration of the number of reserved users. The space is calculated, and if the parking lot does not have a parking management system, the number of available parking spaces may be calculated by dividing into areas including more than a specific number of parking surfaces.

The artificial intelligence-based parking system further includes a camera for determining a license plate of a vehicle and a sticker attached to the vehicle at the entrance of a parking lot, and the parking management server further includes a movement tracking unit for tracking the movement of a vehicle entering the entrance and exit, , When the entered vehicle is parked in the disabled parking area, the license plate information of the determined vehicle is inquired through the server of the Ministry of Public Administration and Security or the integrated server of the local government linked therewith, and if the vehicle is not a disabled vehicle, a notification is sent to the manager. can

The parking surface of the parking lot may be alternately displayed in different colors of adjacent parking surfaces.

The parking surface of the parking lot may be provided with a door opening and closing prevention unit on one side of one parking surface and the other side of the one parking surface and the adjacent parking surface.

The present invention not only can reduce installation and maintenance costs, but also has high reliability and efficiency even outdoors or in poor environments, and the problem of parking vehicles other than disabled vehicles in parking spaces for the disabled, large vehicles such as campers, It has the effect of providing an intelligent image analysis-based smart parking system that can solve both the problem of difficult parking and the problem of needing to share an empty parking surface.

1 is a schematic diagram of a concept of a smart parking system based on intelligent video analysis according to an embodiment of the present invention.

2 is a configuration diagram showing functional blocks of a parking management server of an artificial intelligence-based parking system according to one embodiment of the present invention.

3 to 9 are diagrams for explaining an algorithm for analyzing a parking area and a vehicle area of an image analysis unit and a parking determination unit and determining whether or not parking is permitted according to one embodiment of the present invention.

10 is a configuration diagram showing functional blocks of a parking sensing unit according to one embodiment of the present invention.

11 is a configuration diagram showing functional blocks of a user terminal according to one embodiment of the present invention.

12 illustrates a parking surface of an intelligent video analysis-based smart parking system according to another embodiment of the present invention.

13 illustrates a parking surface including a door opening/closing prevention unit of an intelligent video analysis-based smart parking system according to another embodiment of the present invention.

Hereinafter, with reference to the accompanying drawings, it will be described in more detail the specific examples of the present application. However, the technology disclosed in this application is not limited to the specific examples described herein and may be embodied in other forms.

However, the specific examples introduced herein are provided so that the disclosed content can be thorough and complete and the spirit of the present application can be sufficiently conveyed to those skilled in the art. In the drawing, in order to clearly express the components of each device, the size of the components, such as width or thickness, is shown somewhat enlarged. In addition, although only some of the components are shown for convenience of description, those skilled in the art will be able to easily grasp the remaining components of the components.

When describing the drawings as a whole, it has been described from an observer's point of view, and when an element is referred to as being located above or below another element, this means that the one element is located directly above or below another element, or an additional element between them. It includes all meanings that may be intervened. In addition, those skilled in the art will be able to implement the spirit of the present application in various other forms without departing from the technical spirit of the present application. Also, like reference numerals in a plurality of drawings denote substantially the same elements.

On the other hand, singular expressions described in this application should be understood to include plural expressions unless the context clearly indicates otherwise, and terms such as 'include' or 'having' refer to the described features, numbers, steps, It is intended to specify the presence of an operation, component, part, or combination thereof, but precludes the possibility of existence or addition of one or more other features, numbers, steps, operations, components, parts, or combinations thereof. It should be understood that it does not.

In this specification, a “unit” includes a unit realized by hardware, a unit realized by software, and a unit realized using both. Further, one unit may be realized using two or more hardware, and two or more units may be realized by one hardware. On the other hand, '*?* unit' is not limited to software or hardware, and '*?* unit' may be configured to be in an addressable storage medium or configured to reproduce one or more processors. there is. Therefore, as an example, '*?* part' refers to components such as software components, object-oriented software components, class components, and task components, processes, functions, properties, and programs. procedures, subroutines, segments of program code, drivers, firmware, microcode, circuitry, data, databases, data structures, tables, arrays and variables. Functions provided within components and '-units' may be combined into smaller numbers of components and '-units' or further separated into additional components and '-units'. In addition, components and '~units' may be implemented to play one or more CPUs in a device or a secure multimedia card.

A 'user terminal' referred to below may be implemented as a computer or portable terminal capable of accessing a server or other terminals through a network. Here, the computer includes, for example, a notebook, a desktop, a laptop, etc. equipped with a web browser, and the portable terminal includes, for example, wireless communication that ensures portability and mobility. As a device l International Mobile Telecommunication (IMT)-2000, CDMA (Code Division Multiple Access)-2000, W-CDMA (WCode Division Multiple Access), Wibro (Wireless Broadband Internet), LTT Long Term Evolution) communication-based terminal, smart It may include all types of handheld-based wireless communication devices such as phones and tablet PCs.

In addition, a 'network' is a wired network or mobile radio communication network such as a local area network (LAN), a wide area network (WAN) or a value added network (VAN), or It can be implemented in all kinds of wireless networks such as satellite communication networks.

In addition, in the present specification, 'X to Y' representing a range means 'X or more and Y or less'.

지능형 영상 분석 기반 스마트 주차 시스템Smart parking system based on intelligent video analysis

1 is a schematic diagram of a concept of a smart parking system based on intelligent video analysis according to an embodiment of the present invention. Referring to FIG. 1, the intelligent video analysis-based smart parking system includes an image acquisition unit 100 that acquires an image of a parking lot from a CCTV; A parking sensing unit 200 provided on the parking surface of the CCTV image blind spot to determine whether the vehicle is parked; and a parking management server 300.

2 is a configuration diagram showing functional blocks of a parking management server of an artificial intelligence-based parking system according to one embodiment of the present invention. Referring to FIG. 2 , the parking management server 300 includes an image analysis unit 305 for generating analysis data by analyzing a parking area and a vehicle area based on an image of the image acquisition unit; A parking determination unit 310 for determining whether or not a parking surface in the image is parked or whether parking is possible using a model set to determine whether or not parking is permitted for each parking surface of the parking lot based on the analysis data; a parking information management unit 320 that collects and manages parking information including the number of available parking spaces from the parking determination unit 310 and the parking sensing unit 200; Parking information providing unit 330 for providing the parking information to the parking lot user; and a communication unit 340 for communicating with the CCTV image acquisition unit or a server for managing images of the CCTV image acquisition unit, the parking sensing unit, and a user terminal through a communication network.

The model is a model generated by learning the image and analysis data of the parking lot using an artificial intelligence algorithm, and is a model generated by learning standard information on whether or not the vehicle is parked and whether or not the vehicle is parked on each parking surface of the image of the parking lot. , the learning may include one or more of machine learning and deep learning.

The parking system of the present invention generates analysis data by analyzing the image by applying artificial intelligence to the image of the image acquisition unit 100 that acquires the image of the parking lot from CCTV, and from this, determines whether or not the parking surface is parked and whether or not parking is allowed. By judging, it is possible to more accurately determine whether parking is available and whether parking is permitted, and CCTVs can be newly installed with a communication unit, but the initial installation cost can be significantly reduced by collecting vast parking information by using existing CCTVs. There are advantages to being The parking lot image may be provided directly from a CCTV having a communication unit, or may be provided from a server or system that manages the CCTV, which may be appropriately controlled according to the environment.

Unlike existing rule-based smart systems, artificial intelligence is a system in which machines learn, judge, and become smarter on their own. The more artificial intelligence is used, the higher the recognition rate and the more accurate understanding of user preferences. Existing rule-based smart systems are gradually being replaced by “deep learning”-based “artificial intelligence” systems, and artificial intelligence technologies are machine learning and elements that utilize machine learning. made up of technologies. Machine learning is an algorithm technology that classifies/learns the characteristics of input data by itself, and element technology is a technology that uses machine learning algorithms such as “deep learning” to mimic the functions of the human brain, such as cognition and judgment. It consists of technical fields such as understanding, inference/prediction, knowledge expression, and motion control.

The model set to determine whether or not parking on each parking surface of the parking lot is a model generated by learning using an artificial intelligence algorithm for the image and analysis data of the parking lot, whether or not a vehicle is parked on each parking surface of the parking lot image and whether the vehicle is parked It may be a model generated by learning criterion information on availability. In addition, the model may be a model generated by further learning reference information on whether each parking surface is a parking surface capable of parking a vehicle through image analysis and correction described below in an image of a parking lot.

3 to 9 are diagrams for explaining an algorithm for analyzing a parking area and a vehicle area of an image analysis unit and a parking determination unit according to one embodiment of the present invention and determining whether or not parking is permitted, and FIGS. Referring to FIG. 9, the image analysis unit 305 and the parking determination unit 310 analyze the parking area and the vehicle area to generate analysis data, and an algorithm for determining whether or not the parking surface is parked will be described.

In a specific example, the image acquisition unit 100 may use an existing CCTV, and the existing CCTV has a variety of installation positions, heights, and angles of illumination.

In the parking system of the present invention, the image analysis unit 305 generates analysis data through image analysis performed on a specific basis in images collected from a wide variety of CCTVs. For example, the image analysis unit 305 extracts a parking surface from the collected images (see FIG. 3), and uses a space of a predetermined specific height (H, see FIG. 4) from the extracted parking surface as a parking area. and extracts the parking area as a quadrangular two-dimensional area including the area (see FIG. 5). In this case, the H may be 100 cm to 200 cm, specifically 130 cm to 180 cm.

In addition, the image analysis unit 305 extracts the vehicle from the collected images, and extracts the vehicle region as a quadrangular two-dimensional area including the extracted vehicle (see FIGS. 6 and 7 ).

From the parking area and vehicle area, the overlapping ratio (R) of each parking area with the adjacent parking area is calculated, and the overlapping ratio (P) of the parking area of the parking surface subject to judgment overlaps with the adjacent vehicle area is calculated to be used as analysis data. can be saved (See FIGS. 5 and 8)

Specifically, among the parking areas indicated by the central rectangular area in FIG. 5 , an area ratio of an area overlapping a blue rectangular area, which is an adjacent parking area, may be calculated as R. In addition, among the parking areas indicated by the central rectangular area in FIG. 8 , an area ratio of an area overlapping a blue rectangular area, which is a vehicle area parked in an adjacent parking area, may be calculated as (P).

The parking determination unit 310 may determine whether the parking surface is parked or not by applying a predetermined criterion based on the analysis data.

The preset criterion may be Equation 1 below.

[Equation 1]

P < R × A

The upward correction and the downward correction may be corrected by (P/R)×x.

Referring to FIG. 9, the algorithms of the video analysis unit 305 and the parking determination unit 310 can be equally applied to images or images as shown in FIG. 9 having different angles of CCTV images, and the parking area of an empty parking surface. R is calculated on the basis of, and based on this, it is determined based on the ratio of overlapping P values with adjacent parked vehicles, so it can be applied to CCTV images of all angles.

The CCTV further includes an infrared camera to transmit the captured image to the image analysis unit 305 of the parking management server 300, and the parking determination unit 310 applies the image of the infrared camera together to the vehicle area analysis. can do.

The parking information management unit 320 may manage parking information provided from the parking determination unit 310 and the parking sensing unit 200 . The parking information may include, for example, one or more of the number of available parking spaces, availability of bus or truck parking, availability of electric vehicle charging, and availability of parking for the disabled. In the case of a partitioned or independent parking space, the number of available parking spaces may be based on the parking space of the corresponding area. It can be calculated based on the above space. The reference spaces may be given one parking lot address to be described later, and a navigation function on a client program or web may be applied based on the address. In addition, the number of available parking spaces may be calculated in consideration of the number of user reservations, which will be described later. Specifically, when the parking lot has a parking management system, the parking management server shares user information reserved in the parking management system in conjunction with the parking management system, and the parking management system considers the number of reserved users. When the parking lot is not provided with a parking management system, the number of available parking spaces may be calculated by dividing into areas including more than a specific number of parking surfaces.

The parking information providing unit 330 provides the parking information received from the parking determination unit 310 and the parking sensing unit 200 to the terminal 400 to a user who requests the parking information through a client program or the web. can

The communication unit 340 serves to communicate with the CCTV image acquisition unit 100 or a server managing images of the CCTV image acquisition unit, the parking sensing unit 200, and the user terminal 400 through a communication network.

The parking management server 300 includes a user authentication unit 350 authenticating the parking lot user; a user management unit 360 for managing information of the authenticated user; Reservation management unit 370 for receiving a parking reservation from the parking lot user through a client program or the web; A fee calculation unit 380 for calculating a parking fee of the parking lot user; and a fee payment unit 390 for paying the parking fee. One or more of them may be further included.

The user authentication unit 350 may authenticate the user by a method of inquiring at least one of the user's personal information, the user's reservation time, and vehicle type.

The user management unit 360 may store and manage information such as personal information, vehicle type, vehicle number, mobile phone number, and account number of users using the parking system of the present invention. The above information can be collected through a client program installed in the user terminal 400 to be described later.

The reservation management unit 370 receives reservation information from the user terminal 400, and the reservation information includes one or more of a parking lot to be parked, a vehicle type, and a vehicle number, and within a specific time from the time the reservation information is provided. If there is no information about entering the reserved parking lot, the reservation may be canceled.

The entered information is transmitted in conjunction with the parking management system when the parking lot has a parking management system, and when the parking lot does not have a parking management system, the information is received using the authentication code of the user terminal described later. It may be determined based on whether a wireless link with the parking sensing unit is formed.

The fee calculator 380 may calculate a parking fee for a parking lot user. The time for calculating the fee in the parking system of the present invention is transmitted in conjunction with the parking management system when the parking lot has a parking management system, and when the parking lot does not have a parking management system, the user terminal described later. By using the authentication code of the parking sensing unit and a wireless link can be formed to determine whether or not the parking prevention unit is driven.

The fee payment unit 390 is not particularly limited as long as it is a mobile payment method that is easy for those skilled in the art. For example, Korean Patent Registration No. 10-1148038 may be referred to. By paying the fee before exiting, there is an advantage in minimizing the vehicle being pushed when exiting the parking lot.

The parking sensing unit 200 includes a sensor 210 for determining whether the parking surface of the CCTV image blind spot is parked; And a communication unit 230 for transmitting the parking information sensed by the sensor to the parking management server 300; may include.

The sensor 210 includes a first sensor 211 provided at the center of the floor of each parking space to determine whether the vehicle is parked; a second sensor 213 provided on the parking line to determine whether a parked vehicle invades the parking line; and a signal generating unit (not shown) that notifies when the violation of the parking line is determined by the second sensor 213 . In this case, when the violation of the parking line is determined by the second sensor 213, parking information may be calculated by deducting the number of available parking spaces.

The first sensor 211 is a sensor provided at the center of the floor of the parking space. The number of available parking spaces in a corresponding parking lot area may be calculated from signals sensed by the sensor.

The second sensor 213 is provided on the parking line, specifically, a sensor provided in a buried type, and is not limited as long as it is a sensor that is easy for those skilled in the art to determine that the vehicle has invaded the parking line. There is an advantage in providing a more accurate number of available parking spaces to the parking management server 300 by deducting the number of available parking spaces when the violation of the parking line is determined by the second sensor 213 . In this case, since the number of unparkable spaces is determined by the number of available parking spaces, the user can accurately determine the unavailable spaces in advance, thereby maximizing parking efficiency.

In particular, the first sensor 211 may include a combination of a radar sensor and a geomagnetic sensor. The radar sensor has an advantage in that it has an excellent recognition rate because it can detect distance information of an object without being affected by changes in the external environment such as weather. , leaves, or snow may misjudge it. Accordingly, the first sensor 211 of the present invention determines that the distance measured from the radar sensor is within a specific distance as an obstacle such as fallen leaves or snow, and determines whether the vehicle is parked based on information from the geomagnetic sensor, thereby providing more accurate It has the advantage of being able to determine whether or not to park.

Referring to FIG. 11, the user terminal 400 includes a communication unit 410 for communicating with the parking management server 300 or the parking sensing unit 200 through a communication network; A program that downloads and installs a client program related to the parking information providing service from the parking management server 300 and executes the client program to authenticate the parking lot user or authenticate the parking lot user by accessing the web and receive the parking information section 420; an input unit 430 for the authenticated user to input reservation information including a reservation parking lot, vehicle type, and vehicle number through a client program or the web; a display unit 440 for displaying on a screen information including at least one of parking information of each of the parking lots, reservation time, and scheduled cancellation time; a payment unit 450 for paying a parking fee; And a control unit 460 for controlling at least one of communication with the parking management server or parking sensing unit, execution of the client program, input of reservation information, and display of information; One or more of them may be included.

The communication unit 410 serves to communicate with the parking management server 300 or the parking sensing unit 200 through a communication network.

The program unit 420 downloads and installs a client program related to the parking information providing service from the parking management server 300, and executes the client program to authenticate a parking lot user and receive the parking information. In addition, it is possible to authenticate a parking lot user by accessing the web and to receive the parking information. The user authentication and provision of parking information are as described above.

In a specific example, the program unit 420 may designate a destination and a parking lot within a certain transaction and request parking information related thereto. Regarding this, Korean Patent Registration No. 10-1691312 may be referred to.

The input unit 430 allows an authenticated user to input reservation information and user information including a reserved parking lot, vehicle type, and license plate number through a client program or the web.

The display unit 440 may display parking information, reservation time, scheduled cancellation time for each parking lot near the destination, implement a navigation function, and provide CCTV image information from the parking management server 300 in a specific example. can be received and displayed.

The payment unit 450 is not particularly limited as long as it is a mobile payment method that is easy for those skilled in the art. For example, Korean Patent Registration No. 10-1148038 may be referred to. By paying the fee before exiting, there is an advantage in minimizing the vehicle being pushed when exiting the parking lot.

The control unit 460 may control one or more of communication with the parking management server or parking sensing unit, execution of the client program, input of reservation information, and display of information.

In another specific example, the parking management server 300 further includes an address extraction unit 395 for extracting address information of a place where the CCTV is photographed and address information of the parking sensing unit, and the parking management server 300 ) provides parking information further including the extracted address to the user terminal 400, and the user terminal 400 may provide a navigation function by linking the extracted address with a client program or a map on the web. .

As described in the parking information management unit 320, the address of the parking lot may be assigned based on area division, independence, and setting.

In another specific example, the parking sensing unit 200, the sensor 210 for determining whether the parking surface of the CCTV image blind spot is parked; A communication unit 231 that transmits the parking information sensed by the sensor to the parking management server 300, receives the authentication code of the user who has made a reservation from the parking management server 300, and communicates with the user terminal 400; Parking prevention unit 250 that can prevent parking on the corresponding parking surface; a driving unit 270 for driving the anti-parking unit 250; And a control unit 290 for controlling the sensor 210, the communication unit 231 and the drive unit 270; including, the reservation management unit 370 of the parking management server 300 is the terminal 400 of the user who has made a reservation The smart key including the authentication code is transmitted at the reserved time, and the control unit 290 of the parking sensing unit 200 uses the authentication code of the user terminal 400 to communicate with the parking sensing unit 200 wirelessly. A link may be formed and the driving unit 270 may be controlled.

Specifically, when a user who has reserved a parking lot moves to the corresponding parking lot, the reservation management unit 370 transmits a smart key including an authentication code to the user's terminal 400 at the reserved time, and the user clicks the smart key. By doing so, the parking sensing unit 200 and the user's terminal 400 form a wireless link. In this case, by driving the driving unit 270 with the user's terminal 400, various controls such as opening and closing the anti-parking unit 250, erecting/laying the anti-parking unit 250 according to the shape of the anti-parking unit 250 are possible. The anti-parking unit 250 can be used without limitation as long as it is possible to control operations such as erection/laying by communication.

As a result, it is possible to maximize the efficiency of the parking system by preventing other users who have not made a reservation from parking first, and there is an advantage that the user's parking time can be calculated through this wireless link. In addition, as described above, when a specific time (for example, 5 to 30 minutes) preset in the system from the reservation time is exceeded, the reservation is canceled, so there is an advantage in preventing indiscriminate reservations and no shows. there is. The anti-parking unit of the present invention can be applied as long as it can give an anti-parking effect through the driving unit.

The intelligent video analysis-based smart parking system further includes a camera (not shown) for determining a license plate of a vehicle and a sticker attached to the vehicle at the entrance of the parking lot, and the parking management server 300 is connected to the traffic line of the vehicle entering the entrance. Further provided with a movement tracker (not shown) for tracking, and when the entered vehicle is parked in a parking area for the disabled, the number plate information of the determined vehicle is transmitted through the Ministry of Public Administration and Security server or the integrated server of the local government that is linked therewith. and send a notification to the manager if the vehicle is not a disabled vehicle.

In another embodiment, referring to FIG. 12 , adjacent parking surfaces of the parking lot may alternately display different colors from each other. By guiding one color to front parking and another color to rear parking, there is an advantage in securing space when opening and closing the vehicle door to get on and off when the vehicle crosses between front and rear parking.

In another specific example, referring to FIG. 13, the parking surface of the parking lot may be provided with a door opening/closing prevention unit 900 on one side of one parking surface and the other side of the one parking surface and the adjacent parking surface. As shown in FIG. 13, the opening and closing prevention unit 900 may be provided in areas corresponding to 1/3 and 2/3 in the longitudinal direction of the parking surface, and by making the driver aware of the difficulty of opening and closing the door only by installing a simple single rod, the front It has the advantage of efficiently inducing parking and rear parking. The opening/closing prevention unit 900 is formed of a material such as plastic or resin that can be bent when pressed in a horizontal direction, and thus can prevent damage to the vehicle that may occur.

Although the specific examples of the present invention have been described above, the present invention is not limited to the above specific examples and can be manufactured in various different forms, and those skilled in the art to which the present invention belongs can understand the technical spirit of the present invention. However, it will be understood that it may be embodied in other specific forms without changing its essential features. Therefore, it should be understood that the specific examples described above are illustrative in all respects and not limiting.

100, 101, 102: image acquisition unit 200, 201, 202: parking sensing unit

210: sensor 211: first sensor

213: second sensor 230, 231: communication unit

250: anti-parking unit 270: driving unit

290: control unit 300: parking management server

305: image analysis unit

310: Parking determination unit 320: Parking information management unit

330: Parking information provision unit 340: Communication unit

350: user authentication unit 360: user management unit

370: reservation management unit 380: rate calculation unit

390: bill payment unit 395: address extraction unit

400: user terminal 410: communication unit

420: program unit 430: input unit

440: display unit 450: payment unit

460: control unit 900: door opening and closing prevention unit

Claims

Image acquisition unit for acquiring the image of the parking lot from CCTV; a parking sensing unit provided on a parking surface of the CCTV image blind spot to determine whether the vehicle is parked; And a parking management server; including,

The parking management server,

an image analysis unit generating analysis data by analyzing a parking area and a vehicle area based on the image of the image acquisition unit;

A parking determination unit for determining whether or not a parking surface in the image is parked or not, using a model set to determine whether each parking surface of the parking lot is parked or not, based on the analysis data;

a parking information management unit that collects and manages parking information including the number of available parking spaces from the parking determination unit and the parking sensing unit;

a parking information providing unit providing the parking information to a parking lot user; and

Including; the CCTV image acquisition unit or a server for managing the video of the CCTV image acquisition unit, the parking sensing unit, and a communication unit for communicating with the user terminal through a communication network,

The model is a model generated by learning the image and analysis data of a parking lot using an artificial intelligence algorithm, and is a model generated by learning standard information on whether or not a vehicle is parked and whether or not a vehicle is parked on each parking surface of a parking lot image. Smart parking system based on intelligent video analysis.
According to claim 1,

The learning is an intelligent video analysis-based smart parking system that includes one or more of machine learning and deep learning.
According to claim 1,

The parking sensing unit,

A sensor for determining whether the CCTV image blind spot parking surface is parked; and

An intelligent video analysis-based smart parking system comprising a; communication unit for transmitting the parking information sensed by the sensor to the parking management server.
According to claim 1,

The parking management server,

a user authentication unit authenticating the parking lot user;

a user management unit managing information of the authenticated user;

Reservation management unit for receiving a parking reservation from the parking lot user through a client program or the web;

a fee calculation unit for calculating a parking fee for the parking lot user; and

a fee payment unit for paying the parking fee;

An intelligent video analysis-based smart parking system further comprising one or more of the following.
According to claim 4,

The user authentication unit,

An intelligent video analysis-based smart parking system that authenticates a user by inquiring at least one of the user's personal information, user's reservation time, and vehicle type.
According to claim 1,

The user terminal,

a communication unit for communicating with the parking management server or parking sensing unit through a communication network;

A program unit that downloads and installs a client program related to a parking information providing service from the parking management server, executes the client program to authenticate a parking lot user, or authenticates a parking lot user by accessing the web and receives the parking information;

an input unit for the authenticated user to input reservation information including a reservation parking lot, vehicle type, and license plate number through a client program or the web;

a display unit for displaying on a screen information including at least one of parking information of each of the parking lots, a reservation time, and a reservation cancellation scheduled time;

a payment unit for paying a parking fee; and

a controller for controlling at least one of communication with the parking management server or parking sensing unit, execution of the client program, input of reservation information, and display of information;

An intelligent video analysis-based smart parking system including one or more of the following.
According to claim 1,

The image analysis unit extracts a parking area and a vehicle area from the images collected from the image acquisition unit, calculates the overlapping ratio (R) of each parking area and an adjacent parking area, and determines whether the parking area of the parking surface to be judged is the adjacent vehicle. An intelligent video analysis-based smart parking system that calculates the overlapping ratio (P) with the area and stores it as analysis data.
According to claim 1,

The intelligent video analysis-based smart parking system for determining whether or not the parking surface is parked or not by applying a preset criterion based on the analysis data.
According to claim 1,

The preset criterion is an intelligent video analysis-based smart parking system of Equation 1 below:

[Equation 1]

P < R × A

(In Equation 1, P is the overlapping ratio of the parking area of the parking surface to be determined with the adjacent vehicle area, R is the overlapping ratio of the parking area of the parking surface to be determined with the adjacent parking area, A is 0.317 to 0.517, If Equation 1 is satisfied, it is determined as an available parking surface, and if Equation 1 is not satisfied, it is determined as an unavailable parking surface).
According to claim 9,

The model is

According to Equation 1, the corresponding parking surface is determined to be a non-parking parking surface, and when the vehicle is parked on the parking surface, the A value is downwardly corrected by learning this,

Based on Equation 1, the parking surface is determined to be a parking available parking surface, and if the vehicle is not parked on the parking surface for a certain period of time or more, it learns this and corrects the B value downward. Smart parking system based on intelligent video analysis.
According to claim 10,

The up-correction and down-correction are corrected by (P / R) × x, an intelligent image analysis-based smart parking system (where x is 0.05 to 0.15).
According to claim 1,

The CCTV further includes an infrared camera to transmit the captured image to the image analysis unit of the parking management server, and the parking determination unit applies the image of the infrared camera to the vehicle area analysis. Smart parking system based on intelligent image analysis.
According to claim 4,

The parking management server further includes an address extraction unit for extracting address information of a place where the CCTV is photographed and address information of the parking sensing unit,

The parking management server provides parking information further including the extracted address to the user terminal,

The user terminal provides a navigation function by linking the extracted address with a client program or a map on the web.
According to claim 4,

The parking sensing unit,

A sensor for determining whether the parking surface of the CCTV image blind spot is parked;

a communication unit that transmits parking information sensed by the sensor to a parking management server, receives an authentication code of a reserved user from the parking management server, and communicates with a user terminal;

Parking prevention unit capable of preventing parking on the corresponding parking surface;

a driving unit for driving the parking prevention unit; and

A controller for controlling the sensor, the communication unit, and the driving unit;

The reservation management unit of the parking management server transmits the smart key including the authentication code at the reserved time to the terminal of the user who made the reservation,

The control unit of the parking sensing unit uses the authentication code of the user terminal to form a wireless link with the parking sensing unit and controls the driving unit.
According to claim 14,

The parking management server receives reservation information from the user terminal,

The reservation information includes one or more of a parking lot to be parked, a vehicle type, and a vehicle number,

Intelligent video analysis-based smart parking system in which the reservation is canceled if there is no information entering the reserved parking lot within a specific time from the time of entering the reservation information.
According to claim 15,

The entered information,

If the parking lot has a parking management system, it is transmitted in conjunction with the parking management system,

If the parking lot does not have a parking management system, intelligent video analysis-based smart parking system for determining whether to form a wireless link with the parking sensing unit using the authentication code of the user terminal.
According to claim 4,

The intelligent video analysis-based smart parking system that calculates the number of available parking spaces by subtracting the reserved user.
According to claim 17,

If the parking lot has a parking management system, the parking management server shares user information reserved in the parking management system in conjunction with the parking management system, and the parking management system can park in consideration of the number of reserved users. calculate space,

If the parking lot does not have a parking management system, an intelligent image analysis-based smart parking system that calculates the number of available parking spaces by dividing into areas containing more than a specific number of parking surfaces.
According to claim 12,

The artificial intelligence-based parking system further includes a camera at the entrance of the parking lot to determine the license plate of the vehicle and the sticker attached to the vehicle,

The parking management server further includes a traffic line tracking unit for tracking the traffic line of a vehicle entering the entrance and exit,

When the entered vehicle is parked in a parking space for the disabled, the license plate information of the determined vehicle is inquired through the server of the Ministry of Public Administration and Security or the integrated server of the local government linked thereto to determine whether the vehicle is a disabled vehicle, and if the vehicle is not a disabled vehicle, an intelligent notification is sent to the manager. Video analysis based smart parking system.
According to claim 1,

The parking surface of the parking lot is an intelligent image analysis-based smart parking system in which adjacent parking surfaces are alternately displayed in different colors.
According to claim 1,

The parking surface of the parking lot is an intelligent video analysis-based smart parking system equipped with a door opening and closing prevention unit on one side of one parking surface and the other side of the one parking surface and the adjacent parking surface.