KR102703897B1 - Vehicle number recognition and vehicle access management method and system - Google Patents

Abstract

A vehicle number recognition and vehicle access management method, according to one embodiment of the present application, comprises the steps of: receiving captured video; applying the video to a plurality of models to extract objects and recognize a vehicle number; determining whether the vehicle number recognized by the models are the same; and transmitting a signal to allow or block the vehicle's access based on the determination result. According to one embodiment of the present invention, by applying the video to the plurality of models to extract objects and recognize the vehicle number, the license plates of vehicles entering and exiting can be accurately captured to ensure accurate fee collection.

Description

Vehicle number recognition and vehicle access management method and system {Vehicle number recognition and vehicle access management method and system}

The present application relates to a method and system for vehicle number recognition and vehicle access control, and more specifically, the present application relates to a method and system for vehicle number recognition and vehicle access control that performs vehicle number recognition using multiple models based on acquired images, confirms whether they are the same, and then determines whether they are the same.

In buildings and areas where many vehicles enter and exit, such as apartments and large shopping malls, there are parking lots for parking vehicles. Vehicles entering and exiting the parking lot use an automated parking lot management system that automatically recognizes the license plates of the vehicles entering and exiting, and determines whether or not to allow entry based on the vehicle.

In this regard, Korean Patent Publication No. 10-2604069 (Registration date: November 15, 2023) is a “Vehicle Entry and Exit Management System Capable of Blocking Vehicles with Counterfeit License Plates” that extracts the license plate number from the captured image and compares it with the previously stored license plate data to check if it matches and restricts or allows the entry and exit of the vehicle.

In addition, Korean Patent Publication No. 10-2562135 (Registration Date: July 27, 2023) “Parking Control Device and Parking Settlement Method” checks whether an error situation occurs in which the license plate number of an exiting vehicle fails to be recognized or the license plate number of an entering vehicle that is identical to the license plate number recognized from the exiting vehicle is not retrieved from the database, and, when an error occurs, processes parking settlement by comparing the feature points of a vehicle object extracted from the image of the exiting vehicle with the feature points in the image of the entering vehicle.

This method had the problem that it was difficult to restrict or allow entry and exit of vehicles whose vehicle numbers were not stored and registered. It is necessary to develop a technology that enables accurate collection of fees through normal entry and exit management of vehicles that were not previously stored and registered.

The problem to be solved by the present invention is to provide a method and system for vehicle number recognition and vehicle entry management capable of determining whether vehicle numbers recognized by multiple models are identical and allowing or blocking vehicle entry based on the determination result.

In addition, a problem to be solved by the present invention is to provide a method and system for vehicle number recognition and vehicle access control capable of recognizing numbers in parallel by applying images to arbitrarily different multiple DNN models.

In addition, a problem to be solved by the present invention is to provide a method and system for vehicle number recognition and vehicle entry management, which can store an image and allow entry by receiving a number entered by an administrator when it is determined that the vehicle numbers are not identical based on the result of determining whether the vehicle numbers recognized by each model are identical.

In addition, a problem to be solved by the present invention is to provide a method and system for vehicle number recognition and vehicle entry/exit management that can allow or block exit by matching an object in an image taken at the time of entry with an object in an image taken at the time of exiting when it is determined that the exiting vehicle number is not the same and determining whether they are the same.

The problems to be solved by the present invention are not limited to the problems described above, and problems not mentioned can be clearly understood by a person having ordinary skill in the art to which the present invention belongs from this specification and the attached drawings.

A method for recognizing a vehicle number and controlling vehicle access according to an embodiment of the present invention includes: a step of receiving a photographed image; a step of extracting an object by applying the image to a plurality of models and recognizing a vehicle number; a step of determining whether the vehicle numbers recognized for each model are the same; and a step of transmitting a signal for allowing or blocking vehicle access according to the determination result.

A vehicle number recognition and vehicle access control system according to an embodiment of the present invention includes: a photographing device for photographing an entering or exiting vehicle; a blocking device for allowing or blocking vehicle entry or exit according to a signal received from a server; and a server for receiving an image photographed by the photographing device, applying the image to a plurality of models to extract an object, recognizing a vehicle number, determining whether the vehicle numbers recognized for each model are the same, and transmitting the result to the blocking device and an administrator terminal.

The solutions to the problems of the present invention are not limited to the solutions described above, and solutions that are not mentioned can be clearly understood by a person skilled in the art to which the present invention pertains from this specification and the attached drawings.

According to one embodiment of the present invention, it is possible to determine whether vehicle numbers recognized by multiple models are the same, and to allow or block entry of the vehicle based on the determination result.

In addition, according to one embodiment of the present invention, numbers can be recognized in parallel by applying an image to a plurality of arbitrarily different DNN models.

In addition, according to one embodiment of the present invention, if it is determined that the vehicle numbers are not the same based on the result of determining whether the vehicle numbers recognized by each model are the same, the image can be stored and the number entered by the manager can be received to allow entry.

In addition, according to one embodiment of the present invention, if it is determined that the exiting vehicle number is not the same, the object in the image taken at the time of entry and the object in the image taken at the time of exit can be matched to determine whether they are the same, and exit can be permitted or blocked.

FIG. 1 is a block diagram of a vehicle number recognition and vehicle entry/exit management system according to one embodiment of the present application.
FIG. 2 is a drawing specifically explaining the operation of an object identity judgment module according to one embodiment of the present application.
FIG. 3 is a flowchart showing a vehicle number recognition and vehicle entry/exit management method according to one embodiment of the present application.
FIG. 4 is a flowchart specifically illustrating a method for recognizing a vehicle number by applying an image to multiple models according to one embodiment of the present application.

The above-described objects, features and advantages of the present application will become more apparent through the following detailed description with reference to the attached drawings. However, since the present application may have various modifications and various embodiments, specific embodiments will be illustrated in the drawings and described in detail below.

Throughout the specification, the same reference numerals, in principle, represent the same components. In addition, components that have the same function within the scope of the same idea shown in the drawings of each embodiment are described using the same reference numerals, and redundant descriptions thereof are omitted.

If it is determined that a detailed description of a known function or configuration related to this application may unnecessarily obscure the gist of this application, the detailed description will be omitted. In addition, the numbers (e.g., first, second, etc.) used in the description of this specification are merely identifiers to distinguish one component from another.

In addition, the suffixes “module” and “part” for components used in the following examples are given or used interchangeably only for the convenience of writing the specification, and do not have distinct meanings or roles in themselves.

In the examples below, singular expressions include plural expressions unless the context clearly indicates otherwise.

In the examples below, terms such as “include” or “have” mean that a feature or component described in the specification is present, and do not exclude in advance the possibility that one or more other features or components may be added.

In the drawings, the sizes of components may be exaggerated or reduced for convenience of explanation. For example, the sizes and thicknesses of each component shown in the drawings are arbitrarily shown for convenience of explanation, and the present invention is not necessarily limited to what is shown.

In some embodiments, where the implementation is otherwise feasible, the order of specific processes may be performed differently from the order described. For example, two processes described in succession may be performed substantially simultaneously, or may be performed in an order opposite to the order described.

In the following examples, when components are said to be connected, this includes not only cases where the components are directly connected, but also cases where components are interposed between the components and connected indirectly.

For example, when it is said in this specification that components, etc. are electrically connected, it includes not only cases where the components, etc. are directly electrically connected, but also cases where components, etc. are interposed and indirectly electrically connected.

Hereinafter, the vehicle number recognition and vehicle entry/exit management method and system of the present application will be described with reference to FIGS. 1 to 4.

FIG. 1 is a block diagram of a vehicle number recognition vehicle access management system (hereinafter referred to as system (10)) according to one embodiment of the present application.

Referring to FIG. 1, the system (10) may include a photographing device (100), a blocking device (200), a server (300), and an administrator terminal (500).

Previously, there was a problem that it was difficult to restrict or allow entry and exit of vehicles whose license plates were not stored and registered. Therefore, the development of technology that enables accurate collection of fees through normal entry and exit management of vehicles that were not previously stored and registered is required.

To solve this problem, a system (10) according to an embodiment of the present invention can provide a service that extracts objects by applying images taken by multiple models and recognizes vehicle numbers to enable accurate fee settlement when a vehicle exits parking.

The system (10) can receive a captured image, apply the received image to multiple models to extract an object, and then recognize a vehicle number. The system (10) can determine whether the vehicle number recognized by each model is the same, and allow or block the entry of the vehicle based on the determination result.

Multiple models can be one of the DNN models, and objects can be extracted using multiple models, or vehicle license plates can be recognized through the extracted objects. Objects extracted using multiple models can include vehicle license plates, numbers, and Korean characters.

A Deep Neural Network (DNN) is an artificial neural network (ANN) with multiple hidden layers between the input layer and the output layer. Like general artificial neural networks, DNNs can model complex non-linear relationships. DNNs can include CNNs (Convolutional Neural Networks), RNNs (Recurrent Neural Networks), CRNNs (Convolutional Recurrent Neural Networks), OCRs (Optical Character Recognition), EASTs (Efficient and Accurate Scene Text Detectors), YOLO (You Only Look Once), Faster R-CNN (Region-based Convolutional Neural Networks), SSDs (Single Shot Multibox Detectors), etc.

There may be at least one camera device (100) and the camera device (100) can capture images of vehicles entering and exiting the parking lot. The camera device (100) can transmit captured images to the server (300).

Vehicles may include passenger cars, sports cars, special purpose vehicles, two-wheeled vehicles, three-wheeled vehicles, etc., that have license plates attached.

In one embodiment, when there is only one photographing device (100), the server (300) can distinguish and determine whether the vehicle is entering or exiting based on preset criteria of the image captured by the photographing device (100). The preset criteria may be criteria for distinguishing whether the vehicle is entering or exiting, such as coordinates or a reference line.

In another embodiment, when there are two or more photographing devices (100), the server (300) can distinguish and determine whether it is an entry or exit based on the information of the photographing devices (100). The information of the photographing devices (100) may be a unique identification number of the device.

The blocking device (200) can control the entry and exit of a vehicle by allowing or blocking the entry and exit of the vehicle according to a signal received from the server (300). In addition, the blocking device (200) can control the entry and exit of the vehicle by allowing or blocking the entry and exit of the vehicle when it receives information from the administrator terminal (500). The information can be the vehicle number or a signal for allowing or blocking the entry and exit of the vehicle.

The server (300) can receive an image captured from the photographing device (100), apply the received image to multiple models, extract an object of the vehicle number, and recognize the vehicle number. The server (300) can determine whether the recognized vehicle numbers are the same for each model. The server (300) can transmit the determination result and a signal to allow or block the vehicle's entry to the blocking device (200) and the administrator terminal (500).

The server (300) may include a communication unit (310), a vehicle number recognition module (330), a vehicle number identity determination module (350), an object identity determination module (370), and a memory unit (390).

The communication unit (310) can perform communication with any external device including a server. The system (10) can connect to a network through the communication unit (350) and transmit and receive various data. The communication unit (350) can largely include a wired type and a wireless type. Here, in the case of the wireless type, a communication method of the WLAN (Wireless Local Area Network) series such as Wi-Fi can be used. Alternatively, in the case of the wireless type, a communication method of the cellular communication series such as LTE or 5G can be used. However, the wireless communication protocol is not limited to the examples described above, and any appropriate wireless type of communication method can be used. In the case of the wired type, representative examples include LAN (Local Area Network) or USB (Universal Serial Bus) communication, and other methods are also possible.

The communication unit (310) can transmit the judgment results generated from the vehicle number identical judgment module (350) and the object identical judgment module (370) to the blocking device (200) and the administrator terminal (500). In addition, the communication unit (310) can transmit an image captured by the photographing device (100) and stored in the memory unit (390) to the administrator terminal (500). Alternatively, the communication unit (310) can receive the vehicle number input from the administrator terminal (500).

The license plate recognition module (330) may include multiple models, and may extract objects and recognize license plates by applying images to multiple models.

The license plate recognition module (330) can arbitrarily select multiple different DNN models. The license plate recognition module (330) can apply the received image to multiple different models and recognize the license plate by executing them in parallel. In one embodiment, the license plate recognition module (330) can select CNN and RNN among the DNN models, extract objects for each model, and recognize the license plate.

The vehicle number identical judgment module (350) can calculate the similarity of vehicle numbers recognized by multiple models included in the vehicle number recognition module (330) and determine whether they are identical based on preset criteria. The preset criteria can be set by the administrator.

As an example, if the number recognized by the first vehicle number recognition module is '123ga 4567' and the number recognized by the second vehicle number recognition module is '723geo 4567', the vehicle number identity determination module (350) calculates the similarity between the first and second recognition numbers as '75%', and if the preset standard is '95%', the vehicle number identity determination module (350) can determine that they are not identical.

In addition, the vehicle number identical judgment module (350) can determine that the entry or exit vehicle numbers recognized by each model are not identical based on the vehicle number identical judgment result.

In this way, by recognizing vehicle numbers using different models and determining whether the vehicle numbers recognized by each model are the same, the vehicle number recognition accuracy can be improved. This has the effect of assisting in accurately recognizing vehicle numbers and making vehicle toll collection more accurate.

In addition, the vehicle number identical judgment module (350) can send a notification or a captured image to the administrator terminal (500) when it determines that the vehicle numbers are not identical.

In one example, if the first vehicle number recognition module recognizes '123ga4567' and the second vehicle number recognition module recognizes '723ga4567', the vehicle number identity determination module (350) determines that they are not identical and may send a notification or a captured image to the administrator terminal (500).

In this way, if a notification is sent in real time to the administrator terminal (500) based on the judgment result, the administrator can respond immediately in real time, and as a result, vehicles entering and exiting the parking lot can enter smoothly without being pushed.

In the case where the vehicle number identical judgment module (350) determines that the exit vehicle numbers recognized by each model are not identical, the object identical judgment module (370) can determine whether the object in the image stored at the time of entry and the object in the image captured at the time of exit are identical by matching them. In addition, the object identical judgment module (370) can transmit the judgment result and a signal to allow or block the entry of the vehicle to the administrator terminal (500) and the blocking device (200).

Specifically, reference may be made to FIG. 2. FIG. 2 is a drawing for specifically explaining the operation of the object identical judgment module (370) according to one embodiment of the present application. The object identical judgment module (370) can compare objects extracted from the images stored at the time of entry and the images captured at the time of exit, and determine whether they are identical.

Comparing objects in this way can use less server usage than extracting objects, recognizing numbers, and comparing the recognized numbers, which has the effect of reducing server usage costs.

The memory unit (390) may include a stored image in cases where the vehicle number identification judgment module (350) determines that the vehicle number of an incoming vehicle recognized by model is not identical.

The administrator terminal (500) can receive the determination of whether or not it is the same from the server (300) and can receive the vehicle number through the administrator. The administrator terminal (500) can transmit the entered vehicle number to the memory unit (390).

Hereinafter, with reference to FIGS. 3 and 4, a vehicle number recognition and vehicle entry/exit management method according to one embodiment of the present application will be described.

FIG. 3 is a flowchart showing a vehicle number recognition and vehicle entry/exit management method according to one embodiment of the present application.

The server can receive video of a vehicle entering or exiting a parking lot from at least one camera device (S100).

The server can recognize the vehicle number by applying the received image to multiple models (S200).

The server can determine whether the vehicle numbers recognized by model are identical (S300). In step S300, the server can determine whether they are identical through the similarity of the recognized vehicle numbers, and can determine whether they are identical based on preset criteria by calculating the similarity of the vehicle numbers recognized by multiple models.

As an example, if the server recognizes that the numbers are '123 is 4567' and '723 is 4567', it can calculate the similarity of the recognized numbers as '75%', and if the preset standard is '95%', it can determine that they are not identical because they do not meet the preset standard.

The server can transmit a signal to allow or block entry of the vehicle based on the judgment result, and the blocking device can allow or block entry of the vehicle through the received signal (S450).

If the server determines at step S300 that the recognized vehicle numbers for each model are not the same, the server can store the video (S410).

This has the effect of reducing server usage and server costs by allowing the server to store images based on the judgment results.

The server can send notifications and stored images to the administrator terminal (S411).

The server can receive a number entered from the administrator terminal (S413). In step S413, when the server receives the number, it can transmit a signal for allowing or blocking the entry and exit of the vehicle, and can control entry and exit using the signal received by the blocking device (S450).

If it is determined at step S300 that the vehicle numbers recognized by each model are not the same, the server can match the object in the image stored at step S410 with the object in the image captured at the time of exit (S430).

The server can determine whether the matched objects are identical (S431). If the objects are determined to be identical in step S431, the server can transmit a signal to allow or block the entry of the vehicle, and control the entry and exit through a blocking device (S450). In addition, after determining whether the objects are identical, the server can delete the stored video in step S410.

FIG. 4 is a flowchart specifically illustrating a method for recognizing a vehicle number by applying an image to multiple models according to one embodiment of the present application.

Referring to Fig. 4, the server can arbitrarily select multiple different DNN models (S210).

The server can apply the received image to multiple different models and execute them in parallel to extract objects (S230).

The server can recognize the vehicle number based on the extracted object (S250).

In one embodiment, at step S210, the server selects CNN and RNN among the DNN models and applies the received image to each model to execute in parallel to extract objects and recognize vehicle numbers.

By executing multiple different models in parallel and then extracting objects, the accuracy and speed of license plate recognition can be improved.

According to a method and system for vehicle number recognition and vehicle access control according to an embodiment of the present invention, an object is extracted using multiple models, a vehicle number is recognized using the extracted object, and then the presence or absence of identical vehicle numbers recognized by multiple models is determined. As a result, cases where vehicles enter and exit a parking lot due to incorrect recognition of the vehicle number can be prevented, and normal toll collection can be performed. In addition, since entry and exit can be controlled even for vehicles without stored vehicle numbers, more vehicles can be accommodated and economical effects can be obtained.

The features, structures, effects, etc. described in the embodiments above are included in at least one embodiment of the present invention, and are not necessarily limited to only one embodiment. Furthermore, the features, structures, effects, etc. exemplified in each embodiment can be combined or modified and implemented in other embodiments by a person having ordinary knowledge in the field to which the embodiments belong. Therefore, the contents related to such combinations and modifications should be interpreted as being included in the scope of the present invention.

In addition, although the above has been described focusing on the embodiments, this is only an example and does not limit the present invention, and those with ordinary knowledge in the field to which the present invention pertains will know that various modifications and applications not exemplified above are possible without departing from the essential characteristics of the present embodiment. In other words, each component specifically shown in the embodiments can be modified and implemented. And, the differences related to such modifications and applications should be interpreted as being included in the scope of the present invention defined in the appended claims.

10: Vehicle number recognition and vehicle access control system
100: Filming Device
200: Blocking device
300: Server
310: Communications Department
330: Vehicle Number Recognition Module
350: Vehicle number identification module
370: Object Identicality Determination Module
390: Memory section
500: Admin terminal

Claims (5)

In a method in which a server performs license plate recognition using multiple models and manages vehicle entry and exit,
A step of receiving an image captured from a shooting device;
A step of distinguishing entry or exit using the unique identification number of the above-mentioned photographing device;
A step of recognizing a vehicle number by applying an image to the first model and the second model;
A step of calculating the similarity of the vehicle numbers recognized by the first and second models and determining whether they are the same according to preset criteria; and
A step of transmitting a signal to allow or block vehicle entry based on the above judgment result;
The step of recognizing the vehicle number by applying the image to the first and second models above is:
The first model and the second model are steps of randomly selecting different DNN models; and
A step of recognizing a vehicle number in parallel by applying the image to the first model and the second model; including;
The step of transmitting a signal to allow or block vehicle entry based on the above judgment result is as follows:
As a result of the above judgment, if the similarity of the vehicle numbers recognized by the first and second models is calculated and determined to be not the same according to the preset criteria,
Step of saving the above image;
A step of transmitting the above image to the administrator terminal;
A step of receiving a number entered from the above administrator terminal; and
A step of transmitting an entry permit signal; comprising;
As a result of the above judgment, if the similarity of the exiting vehicle numbers recognized by the first and second models is calculated and determined to be not the same according to the preset criteria,
A step of matching an object in a video saved at the time of entry with an object in a video taken at the time of exit to determine whether they are the same;
A step for transmitting a signal for allowing or blocking exit based on the result of determining whether the above objects are identical by matching them; and
A method for recognizing a vehicle number and managing vehicle entry and exit, comprising: a step of deleting a video stored upon entry;
In a system that performs license plate recognition using multiple models and manages vehicle entry and exit,
A photographing device that photographs vehicles entering and exiting;
A blocking device that allows or blocks the entry and exit of a vehicle according to a signal received from a server; and
A server that receives a video captured from the above-mentioned shooting device, distinguishes entry or exit using the unique identification number of the shooting device, recognizes the vehicle number by applying the video to the first model and the second model, calculates the similarity of the vehicle numbers recognized by the first model and the second model, determines whether they are the same based on preset criteria, and transmits the result to the blocking device and the administrator terminal;
The above server,
The first and second models randomly select different DNN models and then apply the image to the first and second models to recognize the vehicle number in parallel.
If it is determined as a result of the above determination of whether or not the vehicle numbers recognized by the first and second models are not the same, the image is saved and the image is transmitted to the administrator terminal, and the vehicle number entered from the administrator terminal is received and an entry permission signal is transmitted to the blocking device.
A vehicle number recognition and vehicle access control system that, if it is determined as a result of the determination of the above identity that the exiting vehicle numbers recognized by the first and second models are not identical, matches an object in an image stored at the time of entry with an object in an image captured at the time of exit to determine whether they are identical, and then transmits an exit permission and blocking signal and deletes the image stored at the time of entry based on the result of the determination of whether the objects are identical.


delete delete delete