KR102403278B1 - Vehicle number recognition apparatus performing recognition of vehicle number through analysis and correction for a plurality of frames constituting a license plate video - Google Patents

Abstract

Disclosed is a vehicle number recognition apparatus for recognizing a vehicle number through analysis and correction of a plurality of frames constituting a vehicle license plate image. The vehicle number recognition device according to the present invention acquires a vehicle license plate image composed of a plurality of frames by photographing a vehicle license plate of a vehicle based on a camera, and analyzes and corrects each of a plurality of frames constituting the license plate image By providing a vehicle number recognition device that performs recognition of the vehicle number through , it is possible to improve the accuracy of vehicle number recognition.

Description

VEHICLE NUMBER RECOGNITION APPARATUS PERFORMING RECOGNITION OF VEHICLE NUMBER THROUGH ANALYSIS AND CORRECTION FOR A PLURALITY OF FRAMES CONSTITUTING A LICENSE PLATE VIDEO}

The present invention relates to a vehicle number recognition apparatus for recognizing a vehicle number through analysis and correction of a plurality of frames constituting a vehicle license plate image.

Recently, as various technologies related to e-commerce have appeared, e-commerce services for the convenience of users are attracting attention.

In particular, with the advent of Internet-connectable portable devices such as smart phones and tablet PCs, the spread of e-commerce related services using portable devices has been actively carried out.

Recently, as the number of self-driving cars is increasing, a car commerce service called 'drive-through', a service that allows drivers to enter a store and purchase desired products directly from within the vehicle, without getting out of the vehicle is attracting attention.

In order to operate a car commerce service, when a vehicle enters a store, a technology capable of automatically recognizing the vehicle number of an entering vehicle needs to be introduced.

In this regard, the existing vehicle number recognition technology used a method of recognizing the vehicle number by performing image processing from the photographed photo image after photographing the vehicle number plate photo through the camera. There was a disadvantage in that the recognition accuracy was lowered in that the vehicle number was recognized using only the vehicle number.

In addition, in the method utilizing the existing vehicle number recognition technology, if the vehicle license plate picture is taken at an angle, there is also a disadvantage in that it is difficult to properly recognize the license plate number from the picture taken at an angle.

Therefore, in order to solve this disadvantage, a method of detecting the license plate number from each of a plurality of frames constituting the image by photographing the license plate as an image rather than a photo, and performing correction on the image of the license plate photographed at an angle By utilizing the method, additional research is needed on a technology that can improve the accuracy of vehicle number recognition.

Republic of Korea Patent Publication No. 10-1935010 (2019.01.04) Republic of Korea Patent Publication No. 10-2010-0046113 (2010.05.06)

The present invention captures a license plate of a vehicle based on a camera to obtain a license plate image composed of a plurality of frames, and recognizes a license plate number through analysis and correction of each of a plurality of frames constituting the license plate image By providing a vehicle number recognition device that performs the following, it is possible to improve the accuracy of vehicle number recognition.

The vehicle number recognition device for performing the recognition of the vehicle number through analysis and correction of a plurality of frames constituting the vehicle license plate image according to an embodiment of the present invention captures the vehicle license plate of the vehicle through a camera and n(n) is a natural number of 3 or more) an image acquisition unit for obtaining a license plate image composed of frames, and by extracting a license plate region from each of the n frames constituting the license plate image to generate n license plate region images Vehicle license plate area image generating unit, for each of the n number of license plate area images, based on the two-dimensional coordinate value of the uppermost left point corresponding to each vehicle license plate area image and the two-dimensional coordinate value of the uppermost right point of the horizontal axis A tilt angle calculation unit for calculating a tilt angle representing the degree of inclination, based on the tilt angle calculated for each of the n number of license plate area images, each of the n number of license plate area images is horizontal with respect to the horizontal axis By tilting as much as possible, the vehicle license plate area image correction unit for correcting the n number of license plate area images, when the n number of license plate area images are corrected, by extracting the vehicle number from each of the corrected n number of vehicle license plate area images, A vehicle number detection unit for detecting n vehicle numbers, a representative vehicle number extraction unit for extracting at least one non-overlapping representative vehicle number constituting the detected n vehicle numbers from the detected n vehicle numbers, the For each of the at least one representative vehicle number, a counting unit for counting the number of duplicates existing in the detected n number of vehicle numbers, a vehicle number having the largest number of overlaps among the at least one representative vehicle number is the first number for the vehicle When the first vehicle number is determined, and a vehicle number determining unit to determine the first vehicle number, it includes a vehicle number display unit for displaying the first vehicle number on the screen.

The present invention captures a license plate of a vehicle based on a camera to obtain a license plate image composed of a plurality of frames, and recognizes a license plate number through analysis and correction of each of a plurality of frames constituting the license plate image By providing a vehicle number recognition device that performs, it is possible to improve the accuracy of vehicle number recognition.

1 is a view showing the structure of a vehicle number recognition apparatus according to an embodiment of the present invention.
2 is a view for explaining a vehicle number recognition apparatus according to an embodiment of the present invention.
3 is a flowchart illustrating a method of operating a vehicle number recognition apparatus according to an embodiment of the present invention.

Hereinafter, embodiments according to the present invention will be described in detail with reference to the accompanying drawings. These descriptions are not intended to limit the present invention to specific embodiments, and should be understood to include all modifications, equivalents, and substitutes included in the spirit and scope of the present invention. While describing each drawing, like reference numerals are used for similar components, and unless otherwise defined, all terms used in this specification, including technical or scientific terms, refer to those of ordinary skill in the art to which the present invention belongs. It has the same meaning as is commonly understood by those who have it.

In this document, when a part "includes" a certain component, it means that other components may be further included, rather than excluding other components, unless otherwise stated. In addition, in various embodiments of the present invention, each of the components, functional blocks or means may be composed of one or more sub-components, and the electrical, electronic, and mechanical functions performed by each component are electronic. A circuit, an integrated circuit, an ASIC (Application Specific Integrated Circuit), etc. may be implemented as various well-known devices or mechanical elements, and may be implemented separately or two or more may be integrated into one.

On the other hand, the blocks in the accompanying block diagram or steps in the flowchart are computer program instructions that are loaded in a processor or memory of equipment capable of data processing, such as a general-purpose computer, a special-purpose computer, a portable notebook computer, and a network computer, and perform specified functions. can be interpreted as meaning Since these computer program instructions may be stored in a memory provided in a computer device or in a memory readable by a computer, the functions described in the blocks of the block diagrams or the steps of the flowcharts are produced as articles of manufacture containing instruction means for performing the same. it might be In addition, each block or each step may represent a module, segment, or portion of code comprising one or more executable instructions for executing the specified logical function(s). It should also be noted that, in some alternative embodiments, it is also possible for the functions recited in blocks or steps to be executed out of the prescribed order. For example, two blocks or steps shown one after another may be performed substantially simultaneously or in the reverse order, and in some cases, some blocks or steps may be omitted.

1 is a view showing the structure of a vehicle number recognition apparatus according to an embodiment of the present invention.

Referring to FIG. 1 , the vehicle number recognition device 110 according to the present invention includes an image acquisition unit 111 , a vehicle license plate area image generation unit 112 , a tilt angle calculation unit 113 , and a vehicle license plate area image correction unit ( 114 ), a vehicle number detection unit 115 , a representative vehicle number extraction unit 116 , a count unit 117 , a vehicle number determination unit 118 , and a vehicle number display unit 119 .

The image acquisition unit 111 acquires a license plate image composed of n (n is a natural number greater than or equal to 3) frames by photographing the license plate of the vehicle 140 through a camera.

The license plate area image generating unit 112 generates n license plate area images by extracting the vehicle license plate area from each of the n frames constituting the license plate image.

At this time, according to an embodiment of the present invention, the license plate area image generating unit 112 uses a color specified in advance for the license plate area or using an object detection model such as YOLOv3 to configure the license plate image. By extracting the license plate region from each of the frames, it is possible to generate the n license plate region images.

For example, if it is assumed that 'n=5', the image acquisition unit 111 may acquire a license plate image composed of five frames by photographing the license plate of the vehicle 140 through the camera.

At this time, the license plate area image generating unit 112 extracts the vehicle license plate area from each of the five frames constituting the vehicle license plate image, so that 'Vehicle license plate area image 1, Vehicle license plate area image 2, Vehicle license plate area image 3 , it is possible to generate five license plate region images, such as the license plate region image 4, the license plate region image 5'.

The inclination angle calculation unit 113 for each of the n number of license plate area images, the horizontal axis and Calculate the inclination angle representing the degree of inclination of .

In this case, according to an embodiment of the present invention, the inclination angle calculating unit 113 may calculate the inclination angle based on Equation 1 below.

는 상기 기울기 각도로,

은 각 차량 번호판 영역 이미지에 대응되는 최상단 좌측 지점의 2차원 좌표 값,

는 각 차량 번호판 영역 이미지에 대응되는 최상단 우측 지점의 2차원 좌표 값을 의미한다.here,

is the inclination angle,

is the two-dimensional coordinate value of the uppermost left point corresponding to each license plate area image,

denotes a two-dimensional coordinate value of the uppermost right point corresponding to each license plate area image.

The vehicle license plate area image correction unit 114 tilts each of the n number of license plate area images to be horizontal with respect to the horizontal axis based on the inclination angle calculated for each of the n number of vehicle license plate area images. Corrects n license plate area images.

In relation to this, Figure 2 is a view for explaining the operation of the inclination angle calculation unit 113 and the vehicle license plate area image correcting unit 114, reference numeral 210 is a specific generated by the vehicle license plate area image generation unit 112 Indicates the license plate area image 210 in the frame, reference number 220 indicates the two-dimensional coordinate value 220 of the uppermost left point corresponding to the vehicle license plate area image 210, reference number 230 is the vehicle license plate area image ( A two-dimensional coordinate value 230 of the uppermost right point corresponding to 210 is indicated, and reference numeral 240 denotes a horizontal axis 240 .

At this time, as described above, if it is assumed that 'n=5', the inclination angle calculating unit 113 is 'vehicle license plate area images 1, 2, 3, , 4, 5' (210), the two-dimensional coordinate value 220 of the uppermost left point and the two-dimensional coordinate value 230 of the uppermost right point corresponding to each may be calculated.

'와 같이 산출할 수 있다.Then, the inclination angle calculation unit 113 is a two-dimensional coordinate value of the uppermost left point corresponding to each of the five license plate area images, 'vehicle license plate area images 1, 2, 3, 4, 5' (210) ( 220) and the two-dimensional coordinate value 230 of the uppermost right point by substituting into Equation 1, 'Vehicle license plate area images 1, 2, 3, 4, 5' 210, which are the five vehicle license plate area images, respectively The angle of inclination with the horizontal axis 240 for '

' can be calculated as

'를 기초로, 상기 5개의 차량 번호판 영역 이미지들인 '차량 번호판 영역 이미지 1, 2, 3, 4, 5'(210) 각각을 가로축(240)에 대해 수평이 되도록 기울임으로써, 상기 5개의 차량 번호판 영역 이미지들인 '차량 번호판 영역 이미지 1, 2, 3, 4, 5'(210)를 각각 '차량 번호판 영역 이미지 1-1, 2-1, 3-1, 4-1, 5-1'로 보정할 수 있다.After that, the vehicle license plate area image correction unit 114 calculates for each of the five vehicle license plate area images 'Vehicle license plate area images 1, 2, 3, 4, 5' 210 with the horizontal axis 240 and ' is the angle of inclination

Based on ', by tilting each of 'Vehicle license plate area images 1, 2, 3, 4, 5' 210, which are the five license plate area images, to be horizontal with respect to the horizontal axis 240, the five license plates Correction of 'vehicle license plate area images 1, 2, 3, 4, 5' 210, which are area images, to 'vehicle license plate area images 1-1, 2-1, 3-1, 4-1, 5-1', respectively can do.

When the n number of license plate area images are corrected, the vehicle number detection unit 115 detects n number of vehicle numbers by extracting a vehicle number from each of the corrected n number of license plate area images.

The representative vehicle number extraction unit 116 extracts at least one representative vehicle number that does not overlap with each other constituting the detected n number of vehicle numbers from the detected n number of vehicle numbers.

For example, as in the example described above, 'n=5' and the five license plate area images 'Vehicle license plate area image 1, vehicle license plate area image 2, vehicle number plate' by the vehicle license plate area image correction unit 114 Area image 3, vehicle license plate area image 4, vehicle license plate area image 5' is respectively 'Vehicle license plate area image 1-1, vehicle license plate area image 2-1, vehicle license plate area image 3-1, vehicle license plate area image 4-1 , Assume that the license plate area image 5-1' has been corrected.

At this time, the license plate detection unit 115 is the corrected five license plate region images 'Vehicle license plate region image 1-1, license plate region image 2-1, license plate region image 3-1, license plate region image 4- 1, by extracting the license plate number from each of the license plate area image 5-1', five license plate numbers such as '8501, 8502, 8501, 8503, 8504' can be detected.

Then, the representative vehicle number extraction unit 116 is the detected five vehicle numbers '8501, 8502, 8501, 8503, 8504 from the detected five vehicle numbers '8501, 8502, 8501, 8503, 8504' '8501, 8502, 8503, 8504', which are at least one representative vehicle number that does not overlap with each other constituting ', may be extracted.

The counting unit 117 counts the number of duplicates existing in the detected n number of vehicle numbers for each of the at least one representative vehicle number.

The vehicle number determining unit 118 determines the vehicle number having the largest overlapping number among the at least one representative vehicle number as the first vehicle number for the vehicle 140 .

When the first vehicle number is determined, the vehicle number display unit 119 displays the first vehicle number on the screen.

For example, as in the example described above, 'n = 5', the five detected vehicle numbers are referred to as '8501, 8502, 8501, 8503, 8504', and the at least one representative vehicle number is referred to as '8501, 8502. , 8503, 8504', the counting unit 117 for each of the at least one representative vehicle number '8501, 8502, 8503, 8504', the detected five vehicle numbers '8501, 8502, The number of duplicates existing in 8501, 8503, and 8504' may be counted.

At this time, the vehicle number determination unit 118 sets the vehicle number '8501' with the largest overlapping number among the at least one representative vehicle number '8501, 8502, 8503, and 8504' as the first vehicle number for the vehicle 140. can decide

In this way, when the first vehicle number is determined to be '8501', the vehicle number display unit 119 may display '8501', which is the first vehicle number, on the screen.

That is, the license plate recognition device 110 acquires a license plate image composed of n frames by photographing the license plate of the vehicle 140 through a camera, and then from each of the n frames constituting the license plate image. By extracting the license plate area, n number of license plate area images are generated, and for each of the n number of license plate area images, the two-dimensional coordinate value of the uppermost left point corresponding to each license plate area image and the uppermost right point Calculate the inclination angle representing the degree of inclination with the horizontal axis based on the two-dimensional coordinate value, and based on the calculated inclination angle for each of the n number of license plate area images, each of the n number of license plate area images By tilting to be horizontal with respect to the horizontal axis, correcting the n number of license plate area images, and extracting the vehicle number from each of the corrected n number of license plate area images, n number of vehicle numbers are detected, At least one representative vehicle number that does not overlap with each other constituting the detected n vehicle numbers is extracted from the n vehicle numbers, and, for each of the at least one representative vehicle number, from the detected n vehicle numbers By counting the number of duplicates that exist, determining the vehicle number with the largest number of overlaps among the at least one representative vehicle number as the first vehicle number for the vehicle 140 and then displaying the first vehicle number on the screen, the driver It is possible to provide more accurate vehicle number recognition information to

According to an embodiment of the present invention, the vehicle number recognition apparatus 110 may further include a diffusion calculating unit 120 and a message display unit 121 .

After calculating the ratio of the overlapping number of each of the at least one representative vehicle number to the total number of the detected n number of vehicle numbers, the diffusion calculating unit 120 calculates the ratio of the detected n vehicle numbers based on the ratio. Computes the diffusivity indicating how evenly distributed it is.

In this case, according to an embodiment of the present invention, the diffusivity calculator 120 may calculate the diffusivity based on Equation 2 below.

Here, I is the diffusion road, X _i is the number of duplicates of the i-th representative vehicle number among the at least one representative vehicle number, n is the total number of the detected n vehicle numbers, and J is the at least one representative vehicle number means the total number of

The closer the value of the diffusivity calculated according to Equation 2 to '0', the more it means that the detected n number of vehicle numbers are concentrated, and the closer the value of the diffusivity is to '1', the closer the detected This means that n vehicle numbers are evenly distributed.

In this regard, according to the example described above, it is assumed that the five detected vehicle numbers are '8501, 8502, 8501, 8503, 8504', and the at least one representative vehicle number is '8501, 8502, 8503, 8504'. lets do it.

At this time, since J is the total number of the at least one representative vehicle number '8501, 8502, 8503, 8504', it becomes '4', and n is the five detected vehicle number '8501, 8502, 8501, 8503'. , since it is the total number of 8504', it becomes '5'.

And, X ₁ /5 is the ratio of the overlapping number of the first representative vehicle number '8501' to the total number of '8501, 8502, 8501, 8503, 8504', which are the detected five vehicle numbers, so '2/ 5', and X ₂ /5 is the ratio of the overlapping number of '8502', the second representative vehicle number, to the total number of '8501, 8502, 8501, 8503, 8504', which are the detected five vehicle numbers, becomes '1/5', and X ₃ /5 is the overlapping number of '8503', which is the third representative vehicle number for the total number of '8501, 8502, 8501, 8503, 8504', which are the detected five vehicle numbers. Since it is a ratio, it becomes '1/5', and X ₄ /5 is the 4th representative vehicle number '8504' for the total number of the detected five vehicle numbers '8501, 8502, 8501, 8503, 8504' Since it is a ratio of the number of duplicates, it becomes '1/5'.

In consideration of this, if the diffusivity is calculated according to Equation 2, the diffusivity may be calculated as '0.72'.

When it is determined that the calculated diffusion exceeds a preset threshold value, the message display unit 121 displays a message instructing to re-recognize the license plate of the vehicle 140 on the screen.

For example, as in the example described above, it is assumed that the calculated spread is '0.72' and the threshold value is '0.56'. At this time, the message display unit 121 determines that the calculated diffusion '0.72' exceeds the threshold value '0.56', thereby displaying a message instructing to re-recognize the license plate of the vehicle 140 on the screen. can be displayed.

According to an embodiment of the present invention, the vehicle number recognition apparatus 110 may further include a history information database 122 and a history information storage unit 123 .

The history information database 122 stores vehicle number recognition history information for a plurality of vehicles.

Here, the vehicle number recognition history information means information composed of a recognized vehicle number and a time for the recognition time of the vehicle number.

For example, information may be stored in the history information database 122 as shown in Table 1 below.

Vehicle number recognition history information Vehicle number recognition history information 1 (Vehicle number: 3295, time for recognition: 14:30) Vehicle number recognition history information 2 (Vehicle number: 6736, time for recognition time: 15:00) ...

When the first vehicle number is determined, the history information storage unit 123 stores the first vehicle number in the history information database 122 and first vehicle number recognition history information composed of the time for the recognition time of the first vehicle number and the first vehicle number. generated and stored in the history information database 122 .

For example, as in the example described above, it is assumed that the first vehicle number is determined to be '8501', and the time for the recognition time of the first vehicle number '8501' is '21:30'. At this time, the history information storage unit 123 is the time for the recognition time of the first vehicle number '8501' and the first vehicle number '8501' in the history information database 122 as shown in Table 2 below. After generating the first vehicle number recognition history information composed of '21:30' as 'vehicle number recognition history information 20', the 'vehicle number recognition history information 20', which is the first vehicle number recognition history information, is shown in Table 2 below. It can be stored in the history information database 122, such as

Vehicle number recognition history information Vehicle number recognition history information 1 (Vehicle number: 3295, time for recognition: 14:30) Vehicle number recognition history information 2 (Vehicle number: 6736, time for recognition time: 15:00) ... Vehicle number recognition history information 20
(Vehicle number: 8501, time for recognition: 21:30)

According to an embodiment of the present invention, the vehicle number recognition device 110 includes an OTP generating function storage unit 124 , an authentication event generating unit 125 , an authentication request unit 126 , an authentication performing unit 127 , and history information. It may further include a transmission unit 128 .

The OTP generation function storage unit 124 stores an OTP (One Time Password) generation function shared in advance with the manager terminal 130 .

Here, OTP means a one-time authentication number that is randomly generated instead of a fixed password, and the OTP generation function is a function that can be set in advance by a developer, and can be set to generate a one-time authentication number based on the current time value. have.

When the authentication event generator 125 receives a request for transmission of second vehicle number recognition history information among vehicle number recognition history information for the plurality of vehicles stored in the history information database 122 from the manager terminal 130 , , an authentication event for authentication of the manager terminal 130 is generated.

When the authentication event occurs, the authentication request unit 126 generates a random matrix having a size of t x t (t is a natural number greater than or equal to 2) and transmits it to the manager terminal 130 while transmitting a feedback matrix corresponding to the random matrix to request

At this time, according to an embodiment of the present invention, the manager terminal 130 pre-stores the OTP generation function in a memory, and when the random matrix is received from the vehicle number recognition device 110, the Based on the stored OTP generation function, k (k is a natural number greater than or equal to 2) generates k one-time authentication numbers, and after generating a row vector having the k one-time authentication numbers as components, a Kronecker between the row vector and the random matrix After calculating a first feedback matrix by calculating a product (Kronecker product), the first feedback matrix may be transmitted to the vehicle number recognition apparatus 110 .

Here, the Kronecker product refers to a matrix specifically expressing the tensor product of two matrices. Assuming that a matrix M of m x n as in Equation 3 below and a matrix N of p x q as in Equation 4 below are given, It can be expressed as Equation 5 of

At this time, when the first feedback matrix is received from the manager terminal 130 as a feedback matrix corresponding to the random matrix, the authentication performing unit 127 uses the OTP generation function stored in the OTP generation function storage unit 124 . After generating the k one-time authentication numbers on the basis of, authentication for the manager terminal 130 by checking whether a Kronecker product between a row vector having the k one-time authentication numbers as a component and the random matrix is calculated as the first feedback matrix carry out

That is, when the Kronecker product between the row vector and the random matrix is calculated as the first feedback matrix, since the manager terminal 130 can see that the terminal that normally shares the OTP generation function is correct, the authentication performing unit 127 ) may complete the authentication of the manager terminal 130 when the Kronecker product between the row vector and the random matrix is calculated as the first feedback matrix.

When the authentication for the manager terminal 130 is completed, the history information transmission unit 128 extracts the second vehicle number recognition history information stored in the history information database 122 and then sends the second information to the manager terminal 130. Vehicle number recognition history information may be transmitted.

3 is a flowchart illustrating a method of operating a vehicle number recognition apparatus for recognizing a vehicle number through analysis and correction of a plurality of frames constituting a vehicle license plate image according to an embodiment of the present invention.

In step S310, a vehicle license plate image is obtained by photographing the license plate of the vehicle through a camera, and consisting of n (n is a natural number of 3 or more) frames.

In step S320, by extracting a license plate region from each of the n frames constituting the license plate image, n number of license plate region images are generated.

In step S330, for each of the n number of license plate area images, the degree of inclination with the horizontal axis based on the two-dimensional coordinate value of the uppermost left point and the two-dimensional coordinate value of the uppermost right point corresponding to each vehicle license plate area image Calculate the inclination angle representing .

In step S340, based on the inclination angle calculated for each of the n number of license plate area images, by tilting each of the n number of license plate area images to be horizontal with respect to the horizontal axis, the n number of license plate area Correct the images.

In step S350, when the n number of license plate area images are corrected, the n number of vehicle numbers are detected by extracting a vehicle number from each of the corrected n number of license plate area images.

In step S360, at least one representative vehicle number that does not overlap with each other constituting the detected n vehicle numbers is extracted from the detected n vehicle numbers.

In step S370, for each of the at least one representative vehicle number, the number of duplicates existing in the detected n number of vehicle numbers is counted.

In step S380, the vehicle number having the largest overlapping number among the at least one representative vehicle number is determined as the first vehicle number for the vehicle.

In step S390, when the first vehicle number is determined, the first vehicle number is displayed on the screen.

At this time, according to an embodiment of the present invention, in step S330, the inclination angle may be calculated based on Equation 1 above.

In addition, according to an embodiment of the present invention, the method of operating the vehicle number recognition device calculates the ratio of the overlapping number of each of the at least one representative vehicle number to the total number of the detected n number of vehicle numbers, calculating a spread indicating how evenly the detected n number of license plates are distributed based on the ratio, and when it is determined that the calculated spread exceeds a preset threshold value, the license plate of the vehicle The method may further include displaying a message instructing to recognize again on the screen.

In this case, according to an embodiment of the present invention, calculating the diffusivity may include calculating the diffusivity based on Equation 2 above.

In addition, according to an embodiment of the present invention, the operating method of the vehicle number recognition apparatus includes vehicle number recognition history information for a plurality of vehicles (the vehicle number recognition history information is at the time of recognition of the recognized vehicle number and vehicle number) maintaining a history information database in which the information consisting of time) is stored, and when the first vehicle number is determined, the time for the recognition time of the first vehicle number and the first vehicle number in the history information database It may further include the step of generating the first vehicle number recognition history information consisting of and storing it in the history information database.

At this time, according to an embodiment of the present invention, the method of operating the vehicle number recognition device includes maintaining an OTP generation function storage unit in which an OTP generation function previously shared with a manager terminal is stored, and the history information from the manager terminal generating an authentication event for authentication of the manager terminal when a request for transmission of second vehicle number recognition history information among the vehicle number recognition history information for the plurality of vehicles stored in the database is received, the authentication event When generated, generating a random matrix having a size of t x t (t is a natural number greater than or equal to 2) and transmitting it to the manager terminal, requesting transmission of a feedback matrix corresponding to the random matrix, from the manager terminal to the random matrix A first feedback matrix as a corresponding feedback matrix (the first feedback matrix is, in the manager terminal, k (k is a natural number greater than or equal to 2) one-time authentication numbers are generated based on the OTP generation function pre-stored in the manager terminal When a row vector having the k one-time authentication numbers as a component is generated and then the Kronecker product between the row vector and the random matrix is calculated and received), the OTP stored in the OTP generation function storage unit After generating the k one-time authentication numbers based on a generation function, by checking whether the Kronecker product between a row vector having the k one-time authentication numbers as components and the random matrix is calculated as the first feedback matrix, After performing authentication and authentication for the manager terminal is completed, extracting the second vehicle number recognition history information stored in the history information database, and then transmitting the second vehicle number recognition history information to the manager terminal It may include further steps.

At this time, according to an embodiment of the present invention, the manager terminal pre-stores the OTP generation function in a memory, and when the random matrix is received from the vehicle number recognition device, the The first feedback matrix is generated by generating the k one-time authentication numbers based on the OTP generation function, generating the row vector having the k one-time authentication numbers as components, and calculating a Kronecker product between the row vector and the random matrix. After the calculation, the first feedback matrix may be transmitted to the vehicle number recognition device.

In the above, the operation method of the vehicle number recognition apparatus according to an embodiment of the present invention has been described with reference to FIG. 3 . Here, the operating method of the vehicle number recognition apparatus according to an embodiment of the present invention may correspond to the configuration of the operation of the vehicle number recognition apparatus 110 described with reference to FIGS. 1 and 2 , so a more detailed description thereof is to be omitted.

The operating method of the vehicle number recognition apparatus according to an embodiment of the present invention may be implemented as a computer program stored in a storage medium for execution through combination with a computer.

In addition, the operating method of the vehicle number recognition apparatus according to an embodiment of the present invention may be implemented in the form of a program command that can be executed through various computer means and recorded in a computer-readable medium. The computer-readable medium may include program instructions, data files, data structures, etc. alone or in combination. The program instructions recorded on the medium may be specially designed and configured for the present invention, or may be known and available to those skilled in the art of computer software. Examples of the computer-readable recording medium include magnetic media such as hard disks, floppy disks and magnetic tapes, optical media such as CD-ROMs and DVDs, and magnetic media such as floppy disks. - includes magneto-optical media, and hardware devices specially configured to store and execute program instructions, such as ROM, RAM, flash memory, and the like. Examples of program instructions include not only machine language codes such as those generated by a compiler, but also high-level language codes that can be executed by a computer using an interpreter or the like.

As described above, in the present invention, specific matters such as specific components, etc., and limited embodiments and drawings have been described, but these are only provided to help a more general understanding of the present invention, and the present invention is not limited to the above embodiments. , various modifications and variations are possible from these descriptions by those of ordinary skill in the art to which the present invention pertains.

Therefore, the spirit of the present invention should not be limited to the described embodiments, and not only the claims described below, but also all of the claims and all equivalents or equivalent modifications to the claims will be said to belong to the scope of the spirit of the present invention. .

110: vehicle number recognition device
111: image acquisition unit 112: vehicle license plate area image generation unit
113: tilt angle calculation unit 114: vehicle license plate area image correction unit
115: vehicle number detection unit 116: representative vehicle number extraction unit
117: count unit 118: vehicle number determination unit
119: vehicle number display unit 120: diffusivity calculation unit
121: message display unit 122: history information database
123: history information storage unit 124: OTP generation function storage unit
125: authentication event generation unit 126: authentication request unit
127: authentication performing unit 128: history information transmitting unit
130: manager terminal 140: vehicle

Claims (7)

An image acquisition unit for photographing a license plate of a vehicle through a camera to acquire a license plate image composed of n (n is a natural number greater than or equal to 3) frames;
a vehicle license plate area image generator for generating n number of license plate area images by extracting a vehicle license plate area from each of the n frames constituting the license plate image;
For each of the n license plate area images, the inclination angle representing the degree of inclination with the horizontal axis based on the two-dimensional coordinate value of the uppermost left point and the two-dimensional coordinate value of the uppermost right point corresponding to each vehicle license plate area image an inclination angle calculator to calculate ;
Based on the inclination angle calculated for each of the n number of license plate area images, by tilting each of the n number of license plate area images to be horizontal with respect to the horizontal axis, a vehicle to correct the n number of license plate area images License plate area image correction unit;
When the n number of license plate area images are corrected, by extracting a vehicle number from each of the corrected n number of license plate area images, a vehicle number detection unit for detecting the n number of license plates;
a representative vehicle number extraction unit for extracting at least one non-overlapping representative vehicle number constituting the detected n vehicle numbers from the detected n vehicle numbers;
a counting unit for counting the number of duplicates existing in the detected n number of vehicle numbers for each of the at least one representative vehicle number;
a vehicle number determining unit for determining, as a first vehicle number for the vehicle, a vehicle number having the largest overlapping number among the at least one representative vehicle number; and
When the first vehicle number is determined, the vehicle number display unit for displaying the first vehicle number on the screen
including,
The inclination angle calculation unit
A vehicle number recognition apparatus for performing recognition of a vehicle number through analysis and correction of a plurality of frames constituting the vehicle license plate image for calculating the inclination angle based on Equation 1 below.
[Equation 1]

here,

is the inclination angle,

is the two-dimensional coordinate value of the uppermost left point corresponding to each license plate area image,

denotes the two-dimensional coordinate value of the uppermost right point corresponding to each license plate area image. delete According to claim 1,
After calculating the ratio of the overlapping number of each of the at least one representative vehicle number to the total number of the detected n number of vehicle numbers, it is determined how evenly the detected n number of vehicle numbers are distributed based on the ratio a diffusivity calculator that calculates the indicated diffusivity; and
When it is determined that the calculated diffusion exceeds a preset threshold value, a message display unit for displaying a message instructing to re-recognize the license plate of the vehicle on the screen
A vehicle number recognition device for performing recognition of the vehicle number through analysis and correction of a plurality of frames constituting the vehicle license plate image further comprising a. 4. The method of claim 3,
The diffusivity calculator
A vehicle number recognition apparatus for performing vehicle number recognition through analysis and correction of a plurality of frames constituting the vehicle license plate image for calculating the diffusion degree based on Equation 2 below.
[Equation 2]

Here, I is the diffusion road, X _i is the number of duplicates of the i-th representative vehicle number among the at least one representative vehicle number, n is the total number of the detected n vehicle numbers, and J is the at least one representative vehicle number means the total number of According to claim 1,
a history information database in which license plate recognition history information for a plurality of vehicles is stored; and
When the first vehicle number is determined, the first vehicle number recognition history information consisting of the time for the recognition time of the first vehicle number and the first vehicle number is generated in the history information database and stored in the history information database history information storage
A vehicle number recognition device for performing recognition of the vehicle number through analysis and correction of a plurality of frames constituting the vehicle license plate image further comprising a. 6. The method of claim 5,
an OTP generation function storage unit storing an OTP (One Time Password) generation function shared with the administrator terminal in advance;
When a request for transmission of second vehicle number recognition history information among the vehicle number recognition history information for the plurality of vehicles stored in the history information database is received from the manager terminal, an authentication event for authentication of the manager terminal is generated an authentication event generating unit;
When the authentication event occurs, an authentication request unit for requesting transmission of a feedback matrix corresponding to the random matrix while generating a random matrix having a size of txt (t is a natural number greater than or equal to 2) and transmitting it to the manager terminal;
A first feedback matrix from the manager terminal to a feedback matrix corresponding to the random matrix - The first feedback matrix is k (k is 2 or more based on the OTP generation function pre-stored in the manager terminal in the manager terminal) A natural number) one-time authentication numbers are generated, and after a row vector having the k one-time authentication numbers as a component is generated, a Kronecker product between the row vector and the random matrix is calculated and calculated. - When received , After generating the k one-time authentication numbers based on the OTP generating function stored in the OTP generating function storage unit, the Kronecker product between the row vector having the k one-time authentication numbers as components and the random matrix is the first an authentication performing unit for performing authentication on the manager terminal by checking whether the feedback matrix is calculated; and
When the authentication for the manager terminal is completed, after extracting the second vehicle number recognition history information stored in the history information database, a history information transmission unit for transmitting the second vehicle number recognition history information to the manager terminal
A vehicle number recognition device for performing recognition of the vehicle number through analysis and correction of a plurality of frames constituting the vehicle license plate image further comprising a. 7. The method of claim 6,
The manager terminal
The OTP generation function is pre-stored in a memory, and when the random matrix is received from the vehicle number recognition device, the k number of one-time authentication numbers are generated based on the OTP generation function pre-stored in the memory, , After generating the row vector having the k one-time authentication numbers as a component, calculating the Kronecker product between the row vector and the random matrix to calculate the first feedback matrix, and then converting the first feedback matrix to the vehicle number recognition device A vehicle number recognition device for performing vehicle number recognition through analysis and correction of a plurality of frames constituting the transmitted vehicle license plate image.

Images