KR102614194B1 - Parking control system using vehicle detection and license plate recognition technology using PTZ camera - Google Patents

Abstract

The parking control system utilizing vehicle detection and vehicle number recognition technology using a PTZ camera according to the present invention is a PTZ camera equipped with pan, tilt, and zoom functions that photographs parking spaces in the parking lot. Camera (1) and; PTZ camera control module (2) that controls the PTZ camera (1); The video captured through the PTZ camera (1) is transmitted through the PTZ camera control module (2), and deep learning technology is applied to the captured video to determine the presence or absence of parked vehicles in the parking lot, location information of parked vehicles in the captured video, and parking information. An image processing and analysis unit (3) that detects the license plate number of the vehicle; In connection with the image processing and analysis unit 3, when a parked vehicle is detected in a parking space in the captured image, an entry record includes the entry time of the parked vehicle, location information of the parked vehicle in the captured image, and the vehicle number of the parked vehicle. The information is stored in the database 5, and whether the vehicle number corresponding to the location of the parked vehicle is recognized at the location of the parked vehicle in the image stored in the entry record information, and whether the vehicle is parked is always determined to determine whether to leave the vehicle. and an entry/exit detection unit 4 that stores exit record information including the exit time of the exit vehicle, the vehicle number of the exit vehicle, and location information of the parking space in which the exit vehicle was parked, in the database 5; The parking time of a specific license plate number vehicle is calculated by querying the entry and exit record information stored in the database 5, the parking fee is calculated according to the calculated parking time, and the parking fee is collected from the owner of the specific license plate number vehicle. Unmanned payment settlement machine (6); And a parking control computer (7) that is linked to the entry/exit detection unit (4) and the unmanned payment machine (6) to monitor the current number of available parking spaces for each parking lot and the parking fee payment status for each parking lot. The parking control system using vehicle detection and license plate recognition technology using a PTZ camera according to the present invention, which has this configuration, can secure parking space by establishing an on-street unmanned settlement system in the local government and creating an on-street parking lot of a certain size. It can also eliminate the waste of administrative power and parking stress caused by parking management.

Description

Parking control system using vehicle detection and license plate recognition technology using PTZ camera}

The present invention relates to a parking control system using vehicle detection and license plate recognition technology using low-cost PTZ cameras in environments where parking control systems cannot be installed, such as on-street parking lots or small parking lots.

Currently, major roads in downtown are experiencing difficulties with vehicle traffic due to illegal parking and long-term parking.

However, cracking down on illegal parking not only reduces the surrounding commercial area, but also causes vehicles to go from the main street into the alley. To improve this, parking on the street is allowed and managed as long as it does not interfere with vehicle traffic. I'm doing it.

In these street parking lots, a parking manager is always present and parking fees are collected only during the parking manager's working hours.

The parking manager carries a portable parking fee settlement terminal such as a PDA, issues an entry permit to each vehicle parked in each parking area, and collects the designated parking fee through the settlement terminal when the vehicle leaves.

However, it was difficult for parking managers to assign shift workers due to working conditions, so customers had to wander around looking for parking managers or wait for parking managers to appear when they were away or doing other work, which was inconvenient.

In addition, because parking attendants work outdoors, they are in a harsh working environment where they cannot leave their jobs even if weather or physiological problems arise.

In addition, in the case of on-street parking lots, the management area that parking managers can be responsible for is very limited and high fixed costs are incurred. In particular, parking fee collection by parking managers causes irregularities due to disputes with customers over parking fee setting and cash collection. The possibility cannot be ruled out.

In addition, in recent years, each local government has invested a large amount of budget to purchase land in an effort to solve the parking problem, such as the small-scale parking lot project that purchases and builds a certain amount of land in the downtown area and the Ssamzie parking lot project that purchases empty houses and turns them into parking lots. However, even after some sites were purchased, it took a considerable amount of time to negotiate parking lots with residents living around the site, or in some cases, the purchase was canceled. In addition, even when operated as a parking lot, it is impossible to install a parking control system, so parking fees are not collected, so parking rotation is not possible and long-term parking occurs, causing another parking problem.

Meanwhile, as a prior art of the present invention, "On-street parking lot unmanned settlement system and method" under application number "10-2006-0081445" has been applied for and registered. The on-street parking lot unmanned settlement system and method are installed on each of a plurality of parking surfaces. multiple loop coils to detect vehicle entry and exit, a long-distance camera to capture images of the entire on-street parking lot, and a plurality of number recognition cameras to capture images of vehicle license plates when entering and exiting a vehicle based on signals from the loop coil. , Vehicle number recognition means that recognizes the vehicle number from images captured by a remote camera and a number recognition camera corresponding to the parking surface when the vehicle enters the parking surface, from the time of entry of the vehicle detected by the loop coil to the current time. A database server containing a database storing the calculated parking time, the parking fee calculated from the parking time, the image of the vehicle captured by the remote camera and the number recognition camera, and the vehicle number recognized by the vehicle number recognition means; Parking surface: When the user enters his or her vehicle number or the parking surface where his or her vehicle is parked, the captured video, recognition number, entry time, and parking fee stored in the database server are displayed and collected when the user pays the parking fee. It includes an unmanned fee settlement means for doing so, and an integrated management server that receives data stored in the parking lot database server and integrates and manages information on each parking lot from the received data.

Republic of Korea Patent Registration No. "10-0751498" (2007.08.22)

Accordingly, in order to solve the above problems, the present invention is a PTZ camera that can secure parking space by establishing an unmanned on-road settlement system in local governments and creating on-street parking lots of a certain size, and eliminate the waste of administrative power and parking stress due to parking management. It provides a parking control system using vehicle detection and license plate recognition technology.

In addition, another purpose of the present invention is to provide customers with location information of parking lots in the area and information about the number of available parking spaces for each parking lot to customers through a parking information guidance system, thereby shortening customers' parking time and reducing customers' parking stress. It provides a parking control system using vehicle detection and license plate recognition technology using PTZ cameras.

Another purpose of the present invention is vehicle detection and license plate recognition technology using a PTZ camera to reduce customers' parking stress and fuel consumption wasted in finding parking spaces by encouraging customers to park based on clear parking information. The goal is to provide a parking control system utilizing .

In addition, another object of the present invention is to provide a parking control system using vehicle detection and license plate recognition technology using a PTZ camera that can utilize narrow parking spaces as effectively as possible by providing a parking lot fee settlement system that occupies a small installation space. It is done.

In addition, another object of the present invention is parking control using vehicle detection and vehicle number recognition technology using a PTZ camera that can store and recover unpaid parking fees in connection with the parking unpaid settlement system in the district when the customer does not pay the parking fee. To provide a system.

The parking control system using vehicle detection and vehicle number recognition technology using a PTZ camera according to the present invention to achieve the above purpose is a camera equipped with pan, tilt, and zoom functions, and is installed in the parking lot. A PTZ camera (1) that takes pictures of the parking space; PTZ camera control module (2) that controls the PTZ camera (1); The video captured through the PTZ camera 1 is transmitted through the PTZ camera control module 2, and deep learning technology is applied to the captured video to determine the presence or absence of a parked vehicle in the parking space, location information of the parked vehicle in the captured video, An image processing and analysis unit (3) that detects the license plate number of a parked vehicle; In connection with the image processing and analysis unit 3, when a parked vehicle is detected in a parking space in the captured image, an entry record includes the entry time of the parked vehicle, location information of the parked vehicle in the captured image, and the vehicle number of the parked vehicle. The information is stored in the database 5, and whether the vehicle number corresponding to the location of the parked vehicle is recognized at the location of the parked vehicle in the image stored in the entry record information, and whether the vehicle is parked is always determined to determine whether to leave the vehicle. and an entry/exit detection unit 4 that stores exit record information including the exit time of the exit vehicle, the vehicle number of the exit vehicle, and location information of the parking space in which the exit vehicle was parked, in the database 5; The parking time of a specific license plate number vehicle is calculated by querying the entry and exit record information stored in the database 5, the parking fee is calculated according to the calculated parking time, and the parking fee is collected from the owner of the specific license plate number vehicle. Unmanned payment settlement machine (6); and a parking control computer (7) that is linked to the unmanned payment machine (6) and manages and monitors the current parking lot entry and exit status and parking fee payment status. The image processing and analysis unit 3 includes a vehicle detection unit 31 in a parking slot that analyzes the captured image to recognize parking slots and vehicles in the video and detects whether a vehicle is parked in a specific parking slot, and a vehicle in the parking slot. A parked vehicle position coordinate detection unit 32 detects the location coordinates within the image of the parked vehicle detected by the detection unit 31, and applies a CNN (Convolutional Neural Network) algorithm to the captured image to extract the vehicle license plate from the captured image and ANN (Artificial Neural Network) includes a license plate extraction and number recognition unit 34 that recognizes the vehicle number in the vehicle license plate extracted using a character recognizer. In addition, the present invention includes a security and monitoring CCTV (8) that photographs the parking space for safety in the parking lot and securing evidence in the event of an accident, and a network video recorder (9) that stores the images captured by the security and monitoring CCTV (8). ) (NVR: Network Video Recorder) is further included, and the unmanned pay adjustment machine (6) receives parking fees from the owner of the vehicle through card payment or linkage with a payment site using a QR code, and the unmanned pay pay machine (6) It further includes a VAN and QR settlement server 10 that settles the card payment amount and the parking fee paid by QR code, and the VAN and QR settlement server 10 provides parking fee payment to the card company through a payment agency service. Request approval. In addition, the present invention receives and stores the details of non-payment of parking fees by vehicle number from the unmanned settlement machine (6), and when the parking fee is collected through the unmanned settlement machine (6) by the owner of the unpaid parking fee after a certain period of time, the unmanned settlement machine ( It further includes an unpaid settlement server (11) that provides the details of unpaid parking fees through 6) and induces the unpaid parking fees to be settled, and is connected to two or more parking control computers (7) to check the entry and exit status for each parking lot, It further includes an integrated parking control server 12 that centrally manages and monitors the fee receipt status.

The parking control system using vehicle detection and license plate recognition technology using a PTZ camera according to the present invention, which has this configuration, can secure parking space by establishing an on-street unmanned settlement system in the local government and creating an on-street parking lot of a certain size. It can eliminate the waste of administrative power and parking stress caused by parking management.

In addition, the present invention can shorten customers' parking time and reduce customers' parking stress by providing customers with parking location information in the area and information on the number of available parking slots for each parking lot through a parking information guidance system.

In addition, the present invention can reduce customers' parking stress and fuel consumption wasted while searching for a parking space by encouraging customers to park based on clear parking information.

In addition, the present invention provides a parking lot fee settlement system that occupies a small installation space, so that narrow parking spaces can be utilized as effectively as possible.

In addition, the present invention can store and recover the unpaid parking fee in connection with the parking unpaid settlement system in the district when the customer does not pay the parking fee.

Drawings 1a and 1b show PTZ cameras installed in a parking lot,
Figure 2 is a control block diagram of the present invention,
Figure 3 is a control block diagram of the image processing and analysis unit,
Figures 4a to 4e are diagrams for explaining the vehicle number recognition process,
Figure 5a is a flow chart to explain the recognition process of vehicles entering the parking lot,
Figure 5b is a flow chart to explain the recognition process of a vehicle taken out of a parking lot,
Figure 6 is a control block diagram for explaining the security and monitoring CCTV and network video recorder equipped with the present invention,
Figure 7 is a diagram for explaining the VAN and QR settlement server and the unpaid settlement server provided in the present invention;
Figure 8 is a diagram for explaining the integrated parking control server provided in the present invention.

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

The parking control system utilizing vehicle detection and vehicle number recognition technology using a PTZ camera according to the present invention includes pan, tilt, and zoom functions, as shown in Figures 1a to 1b and Figure 2. A PTZ camera (1), which is a functional camera that takes pictures of the parking space in the parking lot; PTZ camera control module (2) that controls the PTZ camera (1); The video captured through the PTZ camera 1 is transmitted through the PTZ camera control module 2, and deep learning technology is applied to the captured video to determine the presence or absence of a parked vehicle in the parking space, location information of the parked vehicle in the captured video, An image processing and analysis unit (3) that detects the license plate number of a parked vehicle; In connection with the image processing and analysis unit 3, when a parked vehicle is detected in a parking space in the captured image, an entry record includes the entry time of the parked vehicle, location information of the parked vehicle in the captured image, and the vehicle number of the parked vehicle. The information is stored in the database 5, and whether the vehicle number corresponding to the location of the parked vehicle is recognized at the location of the parked vehicle in the image stored in the entry record information, and whether the vehicle is parked is always determined to determine whether to leave the vehicle. and an entry/exit detection unit 4 that stores exit record information including the exit time of the exit vehicle, the vehicle number of the exit vehicle, and location information of the parking space in which the exit vehicle was parked, in the database 5; The parking time of a specific license plate number vehicle is calculated by querying the entry and exit record information stored in the database 5, the parking fee is calculated according to the calculated parking time, and the parking fee is collected from the owner of the specific license plate number vehicle. Unmanned payment settlement machine (6); and a parking control computer (7) that is linked to the unmanned payment machine (6) and manages and monitors the current parking lot entry and exit status and parking fee payment status.

In addition, as shown in Figure 2, the present invention further includes an information board 13 that guides the parking area in the parking lot and informs the details of unpaid parking fees when settling the parking fee.

As shown in Figure 3, the image processing and analysis unit 3 is a vehicle detection unit 31 in a parking space that analyzes the captured image, recognizes the parking space and the vehicle in the image, and detects whether the vehicle is parked in a specific parking space. ), a parked vehicle position coordinate detection unit 32 that detects the position coordinates in the image of the parked vehicle detected by the vehicle detection unit 31 in the parking space, and a CNN (Convolutional Neural Network) algorithm is applied to the captured image to capture the image. It includes a license plate extraction and number recognition unit 34 that extracts the license plate number from the image and recognizes the vehicle number plate within the extracted vehicle license plate using an artificial neural network (ANN) character recognizer.

As shown in Figures 4a and 4b, the vehicle detection unit 31 in the parking space displays the vehicle recognized in the captured image as a bounding box, and determines whether the center point of the bounding box is within a specific parking space. It determines whether or not a vehicle can be parked in a specific parking space.

The vehicle detection unit 31 in the parking space detects a vehicle in an image frame using a deep learning algorithm and displays the detected vehicle as a bounding box. The deep learning algorithm uses YOLOv4 (You Only Look Once, Version 4). do.

The vehicle detection involves finding a vehicle, which is an object, in an image frame, and the bounding box is expressed as a dotted rectangle within the image frame, as shown in Figure 4b.

The vehicle detection unit 31 in the parking space displays the vehicle recognized in the captured image as a bounding box.

The image processing and analysis unit 3 displays the vehicle detected in the image frame captured by the PTZ camera 1 as a bounding box, and calculates the position coordinates of the bounding box in the image to determine whether the vehicle is parked in a specific parking space. Wow, the location coordinates in the image of the parked vehicle are detected, the vehicle license plate is extracted from the lower left corner area or the lower right corner area of the bounding box, and the vehicle license plate number is recognized from the extracted vehicle license plate.

As shown in Figures 4d and 4e, the vehicle license plate extraction and number recognition unit 34 enlarges the lower left and right corner areas of the bounding box by a certain magnification to identify the vehicle located at the lower left corner or lower right corner of the bounding box. The license plate area is extracted, and the vehicle number is recognized from the extracted vehicle license plate area.

The process of operating the present invention will be described with reference to Figures 5a to 5b as follows.

The vehicle detection unit 31 in the parking space searches the parking area in the captured image to detect the vehicle. If the vehicle is not detected, the search process is repeated to detect the vehicle (S1, S2, S3).

Additionally, the vehicle detection unit 31 in the parking space displays the vehicle detected in the captured image as a boundary box (S2).

Meanwhile, when a vehicle in the parking area is searched, the parked vehicle position coordinate detection unit 32 calculates the center point of the bounding box, calculates the position coordinates of the vehicle in the captured image, and checks whether the parked vehicle is in the parking space (S4).

Next, the license plate extraction and license plate recognition unit 34 estimates the location of the license plate within the bounding box based on the center point of the bounding box (S5).

Next, the license plate extraction and number recognition unit 34 enlarges the lower edge of one side of the bounding box to recognize the vehicle license plate, but if the vehicle license plate is not recognized, it enlarges the other lower edge of the bounding box to recognize the vehicle license plate. ( S6,S7,S8,S9)

The vehicle license plate extraction and number recognition unit 34, which recognizes the vehicle license plate, stores entry time of the parked vehicle, location information of the parked vehicle in the captured image, and entry record information including the vehicle number of the parked vehicle into the database 5. Save it in (S10)

If the license plate extraction and number recognition unit 34 does not recognize the license plate at both lower corners of the boundary box, it processes the vehicle as a vehicle whose license plate is not recognized (S11).

Meanwhile, the vehicle license plate extraction and number recognition unit 34 searches for the location information of the parked vehicle included in the entry record information to detect the exit vehicle, as shown in Figure 5b (S12).

Next, after detecting the vehicle at the location of the parked vehicle stored in the parking record information in the captured video, the lower corners on both sides of the boundary box are enlarged to recognize the vehicle license plate (S13, S14).

At this time, if the recognized vehicle number matches the entry record information, the vehicle license plate extraction and number recognition unit 34 determines that the vehicle is maintaining entry (S15, S16).

Meanwhile, if the vehicle license plate is not recognized at the location of the parked vehicle stored in the parking record information in the captured video, the PTZ camera control module 2 extracts the vehicle license plate and extracts the PTZ camera 1 according to the control signal from the number recognition unit 34. ) zooms out the zoom lens, and the image processing and analysis unit 3 checks whether the vehicle is detected at the location of the parked vehicle stored in the entrance record information (S15, S17).

At this time, if the vehicle is detected by the image processing and analysis unit 3 at the position of the parked vehicle stored in the entry record information, the entry/exit detection unit 4 determines that the vehicle is maintaining entry. (S18, S19,S16)

On the other hand, when the vehicle license plate and vehicle are not detected at the location of the parked vehicle stored in the entry/exit record information in the captured image, the entry/exit detection unit 4 recognizes that the vehicle has been exited, determines the exit time of the exit vehicle, and the vehicle of the exit vehicle. The exit record information, including the number and location information of the parking space where the exit vehicle was parked, is stored in the database (5) (S19, S20).

In addition, as shown in Figure 6, the present invention includes a security and monitoring CCTV (8) that photographs the parking space for safety in the parking lot and securing evidence in the event of an accident, and an image captured by the security and monitoring CCTV (8). It further includes a network video recorder (9) (NVR: Network Video Recorder) that stores.

As shown in Figure 7, the unmanned payment settlement machine 6 receives parking fees from the owner of the vehicle through card payment or linkage with a payment site using a QR code. The present invention provides a card for each unmanned payment settlement machine 6. It further includes a VAN and QR settlement server 10 that settles the payment amount and the parking fee paid by QR code.

The VAN and QR settlement server 10 requests parking fee payment approval from the card company through a payment agency service.

The unpaid parking fee details for each vehicle number are transmitted and stored from the unmanned pay adjustment machine (6), and when the parking fee is collected through the unmanned pay adjustment machine (6) by the owner of the vehicle for which the parking fee has not been paid after a certain period of time, the parking fee is parked through the unmanned pay adjustment machine (6). It further includes a non-payment settlement server 11 that provides non-payment details and induces settlement of unpaid parking fees.

In addition, as shown in Figure 8, the present invention further includes an integrated parking control server 12 that is connected to two or more parking control computers 7 and centrally manages and monitors the entry and exit status for each parking lot and the fee collection status. Includes.

The integrated parking control server 12 provides location information for each parking lot and information on the number of available parking slots for each parking lot to the terminal in real time through a communication network.

The parking control system using vehicle detection and license plate recognition technology using a PTZ camera according to the present invention, which has this configuration, can secure parking space by establishing an on-street unmanned settlement system in the local government and creating an on-street parking lot of a certain size. It can also eliminate the waste of administrative power and parking stress caused by parking management.

In addition, the present invention can shorten customers' parking time and reduce customers' parking stress by providing customers with parking location information in the area and information on the number of available parking slots for each parking lot through a parking information guidance system.

In addition, the present invention can reduce customers' parking stress and fuel consumption wasted while searching for a parking space by encouraging customers to park based on clear parking information.

In addition, the present invention provides a parking lot fee settlement system that occupies a small installation space, so that narrow parking spaces can be utilized as effectively as possible.

In addition, the present invention can store and recover the unpaid parking fee in connection with the parking unpaid settlement system in the district when the customer does not pay the parking fee.

1. PTZ camera 2. PTZ camera control module
3. Image processing and analysis unit 31. Vehicle detection unit in parking space
32. Parked vehicle location coordinate detection unit 34. Vehicle license plate extraction and number recognition unit
4. Vehicle entry/exit detection unit 5. Database
6. Unmanned payment machine 7. Parking control computer
8. Security and monitoring CCTV 9. Network video recorder
10. VAN and QR settlement server 11. Non-payment settlement server
12. Integrated parking control server

Claims (4)

A PTZ camera (1) equipped with pan, tilt, and zoom functions that takes pictures of parking spaces in a parking lot;
PTZ camera control module (2) that controls the PTZ camera (1);
The video captured through the PTZ camera 1 is transmitted through the PTZ camera control module 2, and deep learning technology is applied to the captured video to determine the presence or absence of a parked vehicle in the parking space, location information of the parked vehicle in the captured video, An image processing and analysis unit (3) that detects the license plate number of a parked vehicle;
In connection with the image processing and analysis unit 3, when a parked vehicle is detected in a parking space in the captured image, an entry record includes the entry time of the parked vehicle, location information of the parked vehicle in the captured image, and the vehicle number of the parked vehicle. The information is stored in the database 5, and whether the vehicle number corresponding to the location of the parked vehicle is recognized at the location of the parked vehicle in the image stored in the entry record information, and whether the vehicle is parked is always determined to determine whether to leave the vehicle. and an entry/exit detection unit 4 that stores exit record information including the exit time of the exit vehicle, the vehicle number of the exit vehicle, and location information of the parking space in which the exit vehicle was parked, in the database 5;
The parking time of a specific license plate number vehicle is calculated by querying the entry and exit record information stored in the database 5, the parking fee is calculated according to the calculated parking time, and the parking fee is collected from the owner of the specific license plate number vehicle. Unmanned payment settlement machine (6);
And a parking control computer (7) that is linked to the unmanned payment machine (6) and manages and monitors the current parking lot entry and exit status and parking fee payment status,
The image processing and analysis unit 3 includes a vehicle detection unit 31 in a parking space that analyzes the captured image to recognize the parking space and the vehicle in the image and detects whether the vehicle is parked in a specific parking space;
A parked vehicle position coordinate detection unit 32 that detects the position coordinates in the image of the parked vehicle detected by the vehicle detection unit 31 in the parking space,
and a vehicle license plate extraction and number recognition unit that applies a CNN (Convolutional Neural Network) algorithm to the captured image to extract the vehicle license plate in the captured image and recognizes the vehicle number in the extracted vehicle license plate using an ANN (Artificial Neural Network) character recognizer. Contains (34),
Security and monitoring CCTV (8) that takes pictures of the parking space to ensure safety in the parking lot and secure evidence in case of an accident,
It further includes a network video recorder (9) (NVR: Network Video Recorder) that stores images captured by the security and monitoring CCTV (8),
The unmanned payment machine 6 receives parking fees from the vehicle owner through card payment or linkage with a payment site using a QR code,
It further includes a VAN and QR settlement server (10) that settles the card payment amount for each unmanned settlement machine (6) and the parking fee paid by QR code,
The VAN and QR settlement server 10 requests parking fee payment approval from the card company through a payment agency service,
The unpaid parking fee details for each vehicle number are received and stored from the unmanned pay adjustment machine (6), and when the parking fee is collected through the unmanned pay adjuster (6) by the owner of the vehicle whose parking fee has not been paid after a certain period of time, the parking fee is parked through the unmanned pay adjuster (6). It further includes an unpaid settlement server 11 that provides unpaid fee details and induces settlement of unpaid parking fees,
It further includes an integrated parking control server (12) that is connected to two or more parking control computers (7) to centrally manage and monitor the entry and exit status for each parking lot and the fee receipt status,
In addition to providing guidance on available parking areas in the parking lot, it further includes an information board (13) that provides information on non-payment details when parking fees are settled.
The vehicle detection unit 31 in the parking space displays the vehicle recognized in the captured image as a bounding box,
Determine whether or not the vehicle is parked in a specific parking space by determining whether the center point of the bounding box is within a specific parking space,
The vehicle detection unit 31 in the parking space detects a vehicle in an image frame using a deep learning algorithm and displays the detected vehicle as a bounding box. The deep learning algorithm uses YOLOv4 (You Only Look Once, Version 4). do,
The image processing and analysis unit 3 calculates the position coordinates of the bounding box in the image, detects the position coordinates in the image of the parked vehicle, extracts the vehicle license plate from the lower left corner area or the lower right corner area of the bounding box, , recognize the vehicle number from the extracted vehicle license plate,
The vehicle license plate extraction and number recognition unit 34 enlarges the lower left corner area or the lower right corner area of the bounding box by a certain magnification to extract the license plate area of the vehicle located at the lower left corner or lower right corner of the bounding box. , recognize the vehicle number from the extracted vehicle license plate area,
The integrated parking control server 12 utilizes vehicle detection and vehicle number recognition technology using a PTZ camera, which provides location information for each parking lot and information on the number of available parking spaces for each parking lot to the terminal in real time through a communication network. Parking control system.
delete delete delete

Images