WO2015167202A1

WO2015167202A1 - Apparatus and method for generating photographic image of license plate number, and license plate number reading system using same

Info

Publication number: WO2015167202A1
Application number: PCT/KR2015/004219
Authority: WO
Inventors: 안순현; 박상춘
Original assignee: 렉스젠(주)
Priority date: 2014-04-30
Filing date: 2015-04-28
Publication date: 2015-11-05
Also published as: KR20150125463A; CN106462770A

Abstract

The present invention relates to an apparatus for generating a photographic image of a license plate number, a method for controlling same, and a license plate number reading system using an apparatus for generating a photographic image of a license plate number. Provided in the present invention is the apparatus for generating a photographic image of a license plate number, comprising: a photographing unit for forming a photographic image of a license plate number corresponding to a license plate, by using an image of a photographed license plate and the license plate number read from the image of the license plate; and an output unit for outputting the license plate image and photographic image.

Description

Vehicle Number Visualization Apparatus, Method and Vehicle Number Reading System Using the Same

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a vehicle number realization apparatus, a method, and a vehicle number reading system using the same, but is not limited thereto. More specifically, an image of a vehicle photographed by a vehicle number reading system (hereinafter, referred to as a "photo image") The present invention relates to a vehicle number actualization device, a method and a vehicle number reading system using the same, which can more effectively verify the accuracy of the vehicle number read in the).

Recently, vehicle number reading systems such as public security, crime prevention system, and law enforcement system have been widely used by CCTV cameras. In a vehicle number reading system, a vehicle including a license plate of a vehicle is photographed, and a vehicle number is read from a photographed image and used for an appropriate purpose. For example, in a law enforcement system such as a speeding enforcement system, a license plate is read to find out which vehicles are speeding up, and in public security and crime prevention systems, a criminal vehicle is found.

However, current vehicle number reading system using the image processing in the vehicle number reading system is the best technology level of 95 ~ 98%, this degree of reading is recognized by experts as a limitation of image processing technology. In other words, 100% perfect vehicle number reading is practically impossible due to the noise included in the image due to various shooting environments. In addition, 100% perfect license plate reading is practically impossible due to the damage, contamination and artificial obstruction of the license plate of the vehicle.

However, 2-5% of unread is a major obstacle to the original purpose of using a license plate reading system. For example, in Suwon, South Korea, about 80 vehicle number reading systems are installed and operated. About 700,000 vehicles are photographed per day. The vehicle number reading system operates in an optimal state (95% recognition rate). Assuming, about 35,000 vehicles fail to read the vehicle number. Failure to read about 35,000 vehicle numbers per day, for example in speeding and security systems, can cause major problems. Therefore, in order to more effectively achieve the original purpose of the vehicle number reading system, proper processing of unread vehicles is required.

However, many of the unread vehicles of 2 to 3% can read the vehicle number with the human eye. This is a technical limitation of vehicle number reading by image processing. Therefore, it is necessary for humans to manually read 2 to 3% of unread cars. There is also a need for means for verifying the accuracy of the vehicle number read by the human eye. In addition, there may be a vehicle that has been misread even though the vehicle license plate has been read out of the vehicle number automatically read out by using image processing in the image of the vehicle photographed. Therefore, even in such a case, a means for verifying the accuracy of the automatically read license plate number is required.

SUMMARY OF THE INVENTION The present invention has been made to solve the above-described problems, and the present invention provides a vehicle number actualization apparatus, a method and a vehicle number reading system using the same, which can efficiently verify the accuracy of a manually read or automatically read vehicle number.

In another aspect, the present invention provides a vehicle number actualization device, a method and a vehicle number reading system using the same that can be efficiently manually read.

According to an embodiment of the present invention, the present invention uses the license plate image (photo image) and the vehicle number (vehicle number) automatically read or manually read from the license plate image, the vehicle number to the license plate A live-action section for generating a converted image of a corresponding type; Provided with a vehicle number actualization device including an output unit for outputting the license plate image and the embodiment image. In the live action unit, it is preferable to perform a live action using license plate type information of the license plate. The license plate type information is preferably at least one of the color of the license plate, the size of the license plate, the arrangement of letters of the license plate, the aspect ratio of the license plate, and the font of the license plate. The license plate type information extracting unit extracting the license plate type information from the license plate image may be further included. The live action image is preferably stored in match with the license plate image.

On the other hand, on the basis of the license plate image and the visualized image output to the output unit, it may further include an input unit for receiving the accuracy of the read vehicle number. The input unit may be configured to further receive a manually read vehicle number based on the vehicle license plate image. When the vehicle number manually read by the input unit is input, it is preferable that the vehicle number is displayed as a realistic image on the output unit.

According to another embodiment of the present invention, the present invention includes a camera unit for photographing a predetermined area including a vehicle; An image processing unit which reads the vehicle number by image processing from the image photographed by the camera unit; An integrated operation unit which receives and manages the photographed image and the read vehicle number; Provides a vehicle number reading system including a manual reading unit for outputting a photographed image determined to need re-reading of the image read by the image processing unit, and receives a vehicle number manually read from a user based on the output image. do. The photographed image that requires the rereading is preferably at least one of an unread photographed image and a photographed image determined to be misread by the image processing unit. In the manual reading unit, when the manually read vehicle number is input, the vehicle number is preferably displayed as a realistic image.

According to another embodiment of the present invention, an information acquisition step of acquiring a license plate image and a vehicle number read from the license plate image; Using a license plate image and the read vehicle number, provides a vehicle number actualization method comprising a virtualization step of making the vehicle number into a realistic image of a form corresponding to the license plate.

An output step of outputting the license plate image and the action image; The method may further include an input step of inputting whether the read license number is correct using the license plate image and the embodiment image output in the output step.

In the due diligence step, it is preferable to perform the due diligence using license plate type information for the license plate. The license plate type information is preferably at least one of the color of the license plate, the size of the license plate, the arrangement of letters of the license plate, the aspect ratio of the license plate, and the font of the license plate.

On the other hand, it may include a comparison step of determining the accuracy of the read license plate number by comparing the license plate image and the live action image with a predetermined algorithm.

According to another embodiment of the present invention, the present invention provides a photo-realization unit for creating a template corresponding to the shape of the license plate image and the license plate image; An output unit for outputting the license plate image and the template; It has an input unit for receiving a manually read vehicle number on the basis of the license plate image, the output unit provides a vehicle number actualization device for outputting the vehicle number received from the input unit to the template. The live action unit may generate a template by image processing the license plate image. In addition, the live-action unit may receive a type of the template from the input unit to create a corresponding template.

An apparatus, method, and method for reading a vehicle number using the same according to the present invention will be described below.

First, according to the present invention, it is possible to efficiently verify the accuracy of the vehicle number manually read by the human eye and the vehicle number reading device in the captured image.

Secondly, according to the present invention, when the vehicle number is read by the human eye from the captured image, it is possible to input the vehicle number read by the naked eye more accurately compared to a predetermined storage device.

Third, according to the present invention, it is possible to more faithfully perform the original purpose of the vehicle number reading apparatus.

Fourthly, according to the present invention, it is now possible to use the vehicle number reading system more effectively in a new system besides the vehicle number reading system in general, such as speeding, security, and the like. For example, in general, a high pass system is generally introduced and operated as an electric tolling system (ETC) for expressing highway tolls, which requires the installation of a terminal in a vehicle. However, if the reading of the vehicle number is substantially 100% guaranteed in the vehicle number reading system, the driver can configure a toll payment system that can be used in a postpaid settlement method without purchasing a separate terminal.

1 is a view for explaining the principle of the vehicle number actualization device according to the present invention

2 is a diagram showing the configuration of an embodiment of a vehicle number reading system with a vehicle number realization apparatus according to the present invention;

3 is a diagram showing the configuration of the vehicle number actualization device of FIG.

4 is a view showing the license plate type of Korea

5 is a view showing an experimental method for confirming the effect of the vehicle number actualization device according to the present invention;

6 is a view showing another embodiment of a vehicle number actualization device according to the present invention;

7 is a view showing another embodiment of a vehicle number actualization device according to the present invention

Referring to the accompanying drawings, it will be described a preferred embodiment of the vehicle number actualization device, method and vehicle number reading system using the same according to the present invention.

According to an exemplary embodiment of the present invention, the vehicle number read from the photographed image is converted into an image corresponding to the photographed license plate image (hereinafter referred to as "converted image" for convenience). Here, the read vehicle number includes a vehicle number read by image processing in the vehicle number reading device (hereinafter referred to as "auto reading") and a vehicle number read by the human eye (hereinafter referred to as "manual reading"). Concept.

First, referring to FIG. 1, the concept of a vehicle number and a visualized image of the vehicle number will be described below. For convenience of explanation, the following describes the license plate used in Korea as an example.

In order to verify the accuracy of the auto-read or manual-ready license plate number, a human visual comparison between the actual license plate image and the license-ready vehicle number ("text type vehicle number" in some cases for convenience) should be made by human eye. . For example, as illustrated in FIG. 1A, the actually photographed license plate image 1 may be compared with the read license plate number 3. In this case, the actual photographed license plate image 1 is compared with the vehicle number " 56 and 8797 "

However, in the present invention, as shown in FIG. 1B, in order to verify more effectively, the vehicle number read out from the license plate image 1 is converted into an image of a license plate form of the vehicle. The accuracy of the reading is determined by comparing the actual license plate image 1 photographed with the actualized image 5 converted into the form of the license plate of the vehicle. In other words, the read license plate number " 56 " 8797 " (3) is converted back into the photorealistic image 5 similar to the license plate image. The actual photographed license plate image 1 is compared with the actual image 5. (The details of the actual image of the license plate image will be described later.) That is, the actual photographed license plate image 1 is represented by text "56." This is compared with the actual image 5 without comparing with 8797 "(3).

An experiment was conducted to verify the performance of scheme (B) according to the invention. In the experiment, 100 wrong vehicle numbers were included in 5 vehicle numbers, and the comparison time, accuracy of comparison, and convenience of the comparators were tested in A and B methods. Experimental results confirmed that the method according to the present invention, that is, the B method is much more effective. (Detailed experimental results will be described later.)

Referring to Figure 2, an embodiment of the configuration of the vehicle number actualization apparatus according to the present invention will be described.

First, the vehicle number reading system will be described. The vehicle number reading system may include a camera unit 10 for capturing a predetermined region, and an image processing unit 20 for reading a vehicle number by image processing an image captured and connected to the camera unit 10. have. The plurality of image processing units 20 are connected to the integrated operating unit 30 by the communication network 12. The integrated operation unit 30 may include a predetermined database to perform integrated search and operation, a penalty notice function for a law violation vehicle, and the like. The camera unit 10 and the image processing unit 20 are installed at a specific site such as a road, and the integrated operating unit 30 is generally installed at a remote location. In addition, the image processing unit 20 may also be provided in the integrated operation unit 30. Meanwhile, the manual readout device 40 may be connected to the integrated operation unit 30. The manual reading apparatus 40 may receive a captured image determined to be necessary for re-reading from the image read by the image processing unit 20, and perform a function of reading the captured image with the naked eye. That is, in the manual reading apparatus 40, it is preferable that a human can manually read and input an unread vehicle number or an image which has been read but suspected to be misread by the image processing unit 20.

On the other hand, the vehicle number actualization device according to this embodiment is not limited, but is preferably provided in the integrated operation unit 30. Referring to Figure 3, the vehicle number actualization apparatus 100 will be described in detail as follows. Vehicle number actualization device 100 is provided with a storage unit 120 for storing the license plate image and the vehicle number read from the license plate image.

The license plate image may refer to an image photographed by the camera unit 10 (see FIG. 2) of a general license plate reading system, and may be an entire image including a license plate of a vehicle or a partial image extracted only from a license plate area of a vehicle. . By using a general vehicle number reading algorithm that is currently used, it is possible to extract only the license plate area of the vehicle, which is generally known to those skilled in the art, and thus the detailed description thereof will be omitted. The vehicle number is a vehicle number read from the photographed license plate image. That is, the vehicle number is a unique vehicle identification number consisting of letters, numbers, and symbols automatically read or manually read from the license plate image.

On the other hand, the storage unit 120 is preferably connected to the live-action unit 110, the live-action unit 110 functions to make a read image of the vehicle image corresponding to the type of license plate. Preferably, the actualization unit 110 is connected to an output unit 140 for outputting the license plate image and the actualized image. Preferably, the output unit 140 may verify the accuracy of the license plate number by comparing the license plate image 1 with the visualized image 5 (see FIG. 1B).

When configured as described above, when verifying the accuracy of the automatic reading or manual reading the vehicle number, the vehicle license plate image and the actualization image is output to the output unit 140, the user can more effectively verify the accuracy .

Each component will be described in detail as follows.

First, the storage unit 120 will be described. The storage unit 120 preferably stores the image photographed by the camera unit of the vehicle number reading system. In addition, the storage unit 120 preferably stores the vehicle number automatically read or manually read from the license plate image.

Next, the realistic image 110 and the actual image will be described in more detail. As mentioned above. The live-action image is an image made to correspond to the read license plate number and the photographed license plate image of the vehicle. The realistic image is preferably made to resemble the actual photographed license plate shape. However, there are several license plate types (forms) in Korea's current legislation. That is, there are license plate types as shown in FIG. In other words, according to the "Notice on Standards of Vehicle Registration Plates" (Ministry of Land, Transport and Maritime Affairs Notification No. 2013-30), there are currently six types of license plates established after November 1, 2006. According to the above notice, the type of license plate currently in use is composed of various forms of different colors, sizes, aspect ratios, character arrangements, fonts, and the like.

Therefore, if the type of license plate is not one, the type of license plate (type), for example, the color of the license plate, the size of the license plate, the arrangement of the letters of the license plate, the aspect ratio of the license plate and the font of the license plate (hereinafter referred to as "number plate format information") It is desirable to create a photorealistic image accordingly. That is, when the type of license plate is not one, when creating a photorealistic image using the vehicle number read by the live-viewing unit 110, the corresponding license plate type information of the vehicle is used as well as the vehicle number. It is desirable to make a photorealistic image from the image of the license plate form.

The license plate type information of the vehicle may be obtained in a step of performing a vehicle number reading algorithm in the vehicle number reading system. Alternatively, the license plate type information extracting unit 150 may be separately provided in the vehicle number actualization device 100, and the license plate type information extracting unit 150 may extract the license plate type information.

Even when it is unsuccessful to read the license plate number by image processing from the photographed license plate image, it is possible to extract license plate type information such as size, aspect ratio, color, etc. of the license plate by image processing. Because, in order to image the license plate image for vehicle number reading, detecting the license plate portion from the image of the vehicle, analyzing the shape (size, color, etc.) of the license plate, and segmenting, reading and recognizing characters using the same Proceeds. However, the step of extracting the license plate type information of the vehicle is a preliminary step of the vehicle number reading, and is relatively easy. Because reading the license plate number from the license plate photographed by the computer should be able to read the letters and numbers that can be recognized as text, but the license plate color, license plate size, etc. is the ratio of R, G, B. This is because it is relatively easy to recognize. Meanwhile, the extracted license plate type information of the vehicle is preferably stored in the storage unit 120 by matching the license plate image and the read vehicle number.

On the other hand, by looking at the photographed license plate image displayed on the output unit 140, it is also possible for a person to manually select the license plate type information of the vehicle through the input unit 130. In this case, the type (type) of each license plate is determined in advance, so that a person does not input the license plate size or color, but instead creates a basic template (hereinafter referred to as a "template") in advance and sets a template (of the license plate type). It is preferable because it is convenient to receive only).

As described above, when the number of license plates is not one, it is necessary to create a template corresponding to the type (type) of each license plate in advance, and to create a photorealistic image by further combining the vehicle number as an image on the template. desirable.

Basically, it is desirable to make a template in advance as many kinds of license plate exist as the current law. The actual picture unit 110 determines the type of license plate by using the license plate type information, and adds the vehicle number as an image to the template. That is, the live-action unit 110 may use the vehicle number and license plate type information read from the license plate image of the vehicle to reconstruct the vehicle number into a realistic image. Of course, as described above, it is also possible to use the template by receiving the type of license plate from the user.

On the other hand, as described above, it is also possible to obtain all license plate format information such as the size, aspect ratio, color, etc. of the license plate by image processing in the photographed license plate image, but using the current regularity described above, license plate format It is also possible to determine which license plate type the license plate is photographed with only part of the information. If this is outlined based on the current notice, it is as follows. Of course, in the case where the notice on the license plate is changed, the manner of determining the type of license plate may be changed. The following description will be made based on the current notice.

The license plate has two colors. The letters of the yellow license plate have two lines of top and bottom, and the vertical and horizontal forms are mixed. The white plate has only the horizontal arrangement. Thus, the license plate color can be used to determine the arrangement of the letters on the license plate.

In addition, the vehicle number includes various information such as region, usage symbol, vehicle type, etc., and it is possible to construct a template using this information. For example, a yellow license plate means a business vehicle, and a white license plate means a car, but taxis and buses are yellow license plates, as well as one of the "bar, four, ah, ruler" in the application symbol. It is configured to include. In addition, in the case of a business, a large license plate is in the form of a license plate consisting of two lines at the top and bottom, and a small license plate includes a vertical (area name) and a horizontal arrangement, and must also include a region name. Accordingly, in consideration of the above-described feature, it is possible to determine which license plate type the license plate photographed with only a part of license plate type information.

Next, the output unit 140 will be described. The output unit 140 simultaneously outputs the actual photographed license plate image and the actual image (see FIG. 1B) to provide an interface for the user to easily compare the two images. Since the actual task is a simple and tedious task, it is desirable to provide a user interface for the user to use. In addition, it is preferable to include a function for checking or correcting the read vehicle number or additionally input an index according to the characteristics of the vehicle according to the comparison result.

On the other hand, as described above, it is preferable to match the actual license plate image and the actual image photographed and stored in the storage unit 120. In addition, the comparison result of the actual license plate image and the actual image is preferably stored in the storage unit 120.

Referring to the effect of the vehicle number actualization device and the method according to the present invention are as follows. In order to determine the effect of the present invention, the following experiment was conducted. As shown in FIG. 5, the experimental method compares the actual license plate image 1 with the vehicle number 3 in a read text form (method 1 of FIG. 5), and the actual license plate image 1 (1). ) And the method of comparing the photorealistic image 5 (method 2 of FIG. 5) were performed. In other words, two verification methods were performed on the same experimenter for 100 license plate images including five misreading vehicle numbers, and the reading time was measured and the number of misreading license plates detected and the convenience of reading were selected. .

구 분division	소요시간 (Sec)Required time (Sec)		2가지 방법에서 편리성 선택 Choice of convenience in two ways
	text data 비교text data comparison	이미지번호판 비교Image License Plate Comparison	2가지 방법에서 편리성 선택 Choice of convenience in two ways

피실험자1Subject 1	169 169	136136	조건 2Condition 2
" 2" 2	146146	134134	""
" 3"3	188188	142142	""
" 4" 4	176176	140140	""
" 5"5	165165	125125	""
" 6"6	180180	145145	""
" 7"7	152152	131131	""
" 8" 8	190190	158158	""
" 9"9	210210	167167	""
" 10"10	162162	131131	""
평 균Average	173.8173.8	140.9140.9	조건1:0, 조건2:10Condition 1: 0, condition 2:10

As can be seen from the above experimental results, the method of method 2 had a reduction of about 20% of the reading time in the comparison of the reading times of the method 1 and the method 2, and it decreased when there were more samples such as 100 or more cases. The effect is expected to be even greater. In addition, in the two methods, 100% of the subjects chose Method 2 for the convenience of reading. In addition, all of the test subjects could feel that eye fatigue was greatly reduced in Method 2. Since the number of samples was 100, a number of misread plates was found in both methods 1 and 2. However, if the number of samples is increased, the method 2 is considered to be excellent in terms of accuracy.

The method (Method 2) of the present invention is much more efficient because, in the case of Method 1, a person is required to interpret an image (actually photographed license plate image) versus text (reading license number-head in the head). Each will be recognized. In the case of the method 2, it is considered that the person intuitively recognizes the image (the actual photographed license plate image) versus the image (the image of the actual vehicle number read out).

6 and 7, another embodiment of the present invention will be described. In the above-described embodiment, an embodiment of verifying the accuracy of the auto-reading or manual-reading vehicle number using the actualized image is described. That is, an embodiment of verifying the accuracy of a plurality of vehicle numbers which are already automatically read and stored in the storage unit 120 (see FIG. 3) or when the vehicle numbers are already manually read and stored in the storage unit. However, in the present embodiment, in the case of manual reading, a person inputs a vehicle number manually read and simultaneously verifies it. In other words, the realistic picture apparatus according to the present embodiment may be used as a manual reading apparatus (see 40 of FIG. 2).

As shown in FIG. 6, when the license plate image of the vehicle is visually read and inputted, the actual image 1 is displayed on the captured image display unit 142 located in a predetermined area of the output unit 140 such as a monitor. do. The human person reads the image 1 visually and inputs it through an input unit such as a keyboard or a touch screen. When a person inputs a vehicle number (text-number, symbol, etc.) that is visually read, the input text is displayed on the input content display unit 144 located in a predetermined area of the output unit 140. In this case, the text input by a person is displayed on the input content display unit 144 as an image of a license plate rather than a text form. That is, the background of the input content display unit 144 is configured as a template in the form of a license plate, and the inputted vehicle number (text) is realized and displayed in real time so that the user can simultaneously compare the input. Alternatively, as shown in FIG. 7, the vehicle number (text) input by a person is displayed in text form on the input content display unit 144, but may be displayed in the form of a license plate on the live action image display unit 146 located in another area. have.

6 and 7, when the vehicle number (text) inputted to the output unit 140 is displayed, a method of reconfirming whether the input contents are correct by hearing using a text to speech (TTS) engine. It is possible.

Another embodiment of the present invention will be described. As described above, the present invention can be used to verify the accuracy of the auto read or manual read license number. However, when the vehicle number is not automatically read by the image processing in the image photographed by the vehicle number reading system, it is preferable to verify the accuracy of the entire manual reading. However, in the case of automatic reading, the accuracy may be verified for all of them, but only the vehicle number suspected of a false reading may be verified only when the accuracy of the reading is suspected. In case of suspected misreading, when reading the license plate number from the image of the license plate photographed by image processing, when the accuracy of the reading is suspected by determining the inherent uncertainty of the algorithm of the image processing itself, for example, the uncertainty of noise removal, etc. Only accuracy verification in accordance with the present invention can be performed. In addition, the uncertainty may be determined not only by the inherent uncertainty of the algorithm of the image processing itself, but also by the legal form of the vehicle number recognized as text. For example, using license plate type information, it is possible to determine whether a vehicle's license plate is partially recognized or misrecognized. For example, if a yellow license plate is used, the name of the area (Seoul, Gyeonggi, etc.) and the usage symbol must be used. It must contain one of the words "ba, sa, ah, ja". Thus, if the yellow license plate does not have a local name or the usage number does not include "bar, sa, ah, ja", it can be presumed to have been misread. The present invention can be appropriately modified without being limited to the above-described embodiments.

Referring to the vehicle number due diligence method according to the present invention.

First, a vehicle license plate image and a vehicle number read from the vehicle license plate image are obtained. (Hereinafter, referred to as "information acquisition step") In the information acquisition step, the license plate image may be transmitted from the camera unit, which is already stored. You can also use video. Also, the read vehicle number may be an auto read or a manual read vehicle number.

Next, using the license plate image and the read license plate number, the license plate image is made into a photorealistic image of a form corresponding to the license plate.

On the other hand, it is preferable to include the step of outputting the license plate image and the actualization image to a predetermined output means (hereinafter "output stage"). In addition, it is preferable to include a step (hereinafter referred to as an "input step") of inputting the correctness of the read vehicle number from the user using the license plate image and the embodiment image output in the output step.

In addition, this invention is not limited to the above-mentioned embodiment. For example, in the above-described embodiment, it has been described that the accuracy of the read vehicle number is determined by the human eye. In other words, the accuracy of the read license plate number was verified by comparing the license plate photographed image with the visualized image of the read license plate number. However, according to the present invention, it is also possible to compare the license plate photographed image and the actualized image of the read vehicle number using an appropriate algorithm instead of the human eye. Since the license plate image is an image and the read vehicle number is data in text form, it is practically impossible to algorithmically compare the license plate image with the read vehicle number using an algorithm. However, license plate photographed images and photorealistic images are image versus image. Therefore, it is possible to compare an image with an appropriate algorithm. That is, according to another embodiment of the present invention, instead of comparing the license plate image with the actual image of the read license plate image, the present invention compares the license plate image and the realistic image with an algorithm, thereby reading the read image. It may be configured to include the step of determining the accuracy of the vehicle number (hereinafter referred to as "compare step").

Claims

A live-action unit for making the vehicle number into a realistic image of a form corresponding to the license plate by using the photographed license plate image and the vehicle number automatically read or manually read from the license plate image;

Vehicle number actualization device comprising an output unit for outputting the license plate image and the actual image.
The apparatus of claim 1, wherein the photo-projection unit performs a photo-realization using license plate type information of the license plate.
The apparatus of claim 2, wherein the license plate format information is at least one of a color of a license plate, a size of a license plate, a text arrangement of the license plate, an aspect ratio of the license plate, and a font of the license plate.
The apparatus of claim 2 or 3, further comprising a license plate type information extracting unit for extracting the license plate type information from the license plate image.
The apparatus of claim 1, wherein the actual image is stored in match with the license plate image.
The apparatus of claim 1, further comprising an input unit configured to receive an accuracy of the read vehicle number based on the vehicle license plate image and the live action image output on the output unit.
The apparatus of claim 1, wherein the input unit further receives a vehicle number manually read from a user based on the license plate image.
The apparatus of claim 7, wherein when the vehicle number manually read from the input unit is input, the vehicle number is displayed as a photorealistic image on the output unit.
A camera unit for photographing a predetermined area including a vehicle;

An image processing unit which reads the vehicle number by image processing from the image photographed by the camera unit;

An integrated operation unit which receives and manages the photographed image and the read vehicle number;

And a manual reading unit which outputs a photographed image determined to be necessary for re-reading among the images read by the image processing unit, and receives a manually read vehicle number based on the output photographed image.
The vehicle number reading system according to claim 9, wherein the photographed image requiring the reread is at least one of a photographed image which is not read by the image processing unit and a photographed image which is determined to be misread.
The vehicle number reading system according to claim 10, wherein the manual reading unit displays the vehicle number as a photorealistic image when a manually read vehicle number is input.
An information acquisition step of obtaining a license plate image and a license plate number read from the license plate image;

And using a license plate image and the vehicle number to generate a photo-realistic image of the vehicle number in a form corresponding to the license plate.
The method of claim 12, further comprising: an output step of outputting the license plate image and the action image; And an input step of receiving an accuracy of the read vehicle number by using the vehicle license plate image and the embodiment image output in the output step.
The vehicle number due diligence method of claim 12, wherein in the due diligence step, due diligence is performed using license plate type information of the license plate.
15. The method of claim 14, wherein the license plate format information is at least one of a color of a license plate, a size of a license plate, a text arrangement of the license plate, an aspect ratio of the license plate, and a font of the license plate.
The method of claim 12, further comprising comparing a vehicle license plate image with the visualized image by a predetermined algorithm to determine the accuracy of the read vehicle number.
A live-action unit for creating a template corresponding to the shape of the license plate image and the license plate image;

An output unit for outputting the license plate image and the template;

On the basis of the license plate image has a input unit for receiving a manually read vehicle number,

And the output unit outputs a vehicle number received from the input unit to the template.
18. The apparatus of claim 17, wherein the photo realization unit generates a template by image processing the license plate image.
18. The apparatus of claim 17, wherein the live-action unit receives the type of the template from the input unit and creates the template.