KR101295000B1 - System for recognizing credit card number using area property of card number and method thereof - Google Patents

Abstract

PURPOSE: An image analysis-based credit card number reading system which utilizes area characteristics of a credit card number and a reading method thereof increases the reading rate of a credit card by comparing the image of embossed numbers on a credit card with unit area-specific characteristic information. CONSTITUTION: A card image storage unit (21) contains an image of a credit card. A print number reading unit (22) reads the first four print numbers out of a total of 16 print numbers from the card image. An embossed number reading unit (23) reads the first four embossed numbers located over the four print numbers out of a total of 16 embossed numbers until it can match the read print numbers with the four embossed numbers. If the four print numbers and the four embossed numbers turn out to be identical, the embossed number reading unit reads the remaining 12 embossed numbers. A number comparing unit (24) determines whether the four print numbers and the four embossed numbers are identical to each other by comparing them. [Reference numerals] (2) Computer device; (21) Card image storage unit; (22) Printing number recognizing unit; (23) Embossed number recognizing unit; (24) Number comparing unit; (25) Card number validity investigating unit; (26) Identifier number replacing unit; (27) Reflecting portion recognizing unit

Description

System for recognizing credit card number using area property of card number and method

The present invention relates to a credit card number recognition system and a credit card number recognition method, and more particularly, to a credit card number recognition system and a credit card number recognition system that recognizes a credit card number by analyzing an image of the credit card. It is about a method.

Operators such as credit card companies and payment companies manage customer credit card information as needed. In addition, operators in various fields collect credit card information to customers as needed to store their credit card numbers.

Credit cards are printed with a variety of information, including a 16-digit card number, expiration date and name. The 16-digit card number is printed with an embossed number. Immediately below the first four-digit embossed number, the same number as the four-digit embossed number is printed in small letters.

Here, the printing format of the credit card and the layout of the card information are uniform throughout the world. For example, the font, size and position of characters on a credit card are all unified by the standard format. Therefore, even if the credit card company is another credit card, the general purpose card reader can read the card information from the credit card and process the card payment.

On the other hand, the operator of the credit card after checking the physical card issued after issuing the card before issuing to the customer can enter and manage the card information.

In addition, the operator who needs to manage the customer's credit card information can manage the customer's credit card number with the customer's consent. For example, the employee in charge may request the card information from the customer and enter the number of the credit card presented by the customer to store the card information. The card information may be periodically used to charge a customer.

However, when the operator manages the card information of many customers, manually inputting the customer's card information is very cumbersome and time consuming. In addition, as the number of inputs of card information manually input increases, the possibility of mis-input of the card number increases due to an input error of the person in charge.

The present invention was created under the recognition of the prior art as described above, and identifies the image of the number embossed on the credit card, and recognizes the number of the credit card using the area-specific characteristics of the credit card number recognition system and credit It is an object to provide a card recognition method.

Another object of the present invention is to recognize the first four digits printed number and the first four digits embossed number from the image of the credit card, and then the remaining 12 digits embossed number if the recognized print number and the embossed number match To recognize it.

According to an aspect of the present invention, there is provided a credit card number recognition system for recognizing a credit card number, comprising: card image storage means for storing a card image photographed from a credit card; Print number recognizing means for recognizing a print number on which a first four digit number is printed from a card number of 16 digits from the card image; Recognize the first four-digit embossed number located above the print number among the sixteen-digit embossed numbers of the card image until it matches the recognized print number, and if the print number and the first four-digit embossed number match, the 16 Embossed number recognizing means for recognizing the remaining 12-digit embossed numbers among the digit embossed numbers; And a number comparison unit comparing the recognized print number with the first four digit embossed number to determine whether they match.

In the present invention, the computer terminal performs an image process of polarizing while adjusting the reference values of brightness and saturation with respect to the image of the individual number identified from the card image, and then card number from the image of the individual number processed image. Recognize.

Here, the print number recognition means may include a relative distance recognition scheme for recognizing an area of the print number located at a relative distance from an IC (chip) position of the card image; And recognizing an area of the print number by using any one of an absolute position recognition method of recognizing an area of the print number located in a predefined area in the card image. An image of the individual number identified from the image is compared with an image of a previously stored number to recognize the matching print number.

According to an aspect of the present invention, the computer terminal divides an image of an individual embossed number into a plurality of unit areas, and determines whether each of the divided unit areas is blank or filled, and recognizes the number as 0 to 9. .

Here, the computer terminal stores a table in which information on whether the individual unit areas are blank or filled is stored for a plurality of unit areas corresponding to each of the numbers 0 to 9.

In addition, the computer terminal compares whether the unit area is empty or not for each image of the respective embossed numbers, obtains a degree of coincidence with respect to the numbers 0 to 9, and recognizes the number with the highest coincidence as the embossed number.

Furthermore, the embossed number recognizing means repeats polarization image processing for each individual digit of the four-digit embossed number while adjusting the reference value of brightness and saturation until the four-digit embossed number matches the print number.

Further, the embossed number recognizing means performs polarization image processing for each individual digit of the 12-digit embossed number using reference values of brightness and saturation used when the 4-digit embossed number corresponding to the print number is recognized. .

The computer terminal may further include a card number validity checker that checks the validity of the card number when the 16-digit embossed number is recognized as the card number.

The computer terminal recognizes that if the validity check fails, the validity check is repeatedly performed by replacing the embossed number having the lowest match among the 12-digit embossed numbers with a number having a matching degree of next rank. It further includes a number replacement.

According to another embodiment of the present invention, the computer terminal, after shading the remaining portions other than the bright portion above the threshold value due to the reflection of light on the image of the individual relief number, by using the position information of the bright portion And reflecting portion recognition means for obtaining respective correspondences to the numbers 0 to 9, and recognizing the number having the highest coincidence as the embossed number.

Here, the computer terminal determines whether the unit area is empty, determines the recognized 12-digit number and the position information of the bright part, compares the recognized 12-digit number, and determines whether they match.

According to another aspect of the present invention, the computer terminal may be configured as a client terminal or a network server, and when configured as the network server, receives a card image through a network and responds to a recognition result of the card number.

In order to achieve the above technical problem, a credit card number recognition method for recognizing a credit card number according to the present invention is a credit card number recognition method processed by a computer terminal that recognizes a card number from a credit card image. (a) a card image storing step of storing a card image photographed from a credit card; (b) a printing number recognition step of recognizing a printing number on which a first four digit number is printed among 16 digit card numbers from the card image; (c) a four-digit embossed number recognizing step of recognizing a first four-digit embossed number located above the printed number among the sixteen-digit embossed numbers of the card image until it matches the recognized print number; (d) a number comparison step of comparing the recognized print number with the first four digit embossed number to determine whether they match; And (e) a 12-digit embossed number recognizing step of recognizing the remaining 12-digit embossed number of the 16-digit embossed number if the print number and the first 4-digit embossed number match.

According to an aspect of the present invention, the recognition rate of the card number is increased by recognizing the card number in contrast to the characteristic information for each unit area for the image of the number embossed on the credit card.

According to another aspect of the present invention, the first four digits are recognized from the card image to determine whether they match and then the remaining card numbers are recognized to increase the validity of the card number.

According to another aspect of the present invention, by obtaining a matching degree for each digit of the recognized card number, and if the validation of the card number is failed, the card number by repeating the validation of the card number by replacing with another number using the matching degree card number To complement the validity.

In addition, by performing image processing to extract only the bright part of the embossed number reflected in the light, the recognition rate and effectiveness of the card number are improved.

The following drawings attached to this specification are illustrative of preferred embodiments of the present invention, and together with the detailed description of the invention to serve to further understand the technical spirit of the present invention, the present invention is a matter described in such drawings It should not be construed as limited to.
1 is a schematic diagram of a credit card number recognition system according to an embodiment of the present invention.
2 and 3 are exemplary diagrams of a credit card recognized by a credit card number recognition system according to an embodiment of the present invention.
4A to 4J are exemplary views of analyzing a unit area of an embossed number in an image of a credit card according to an embodiment of the present invention.
FIG. 5 is an exemplary diagram showing a table indicating the degree of correspondence for each number of the analyzed image of relief numbers according to an embodiment of the present invention.
6 is an exemplary view of recognizing the embossed number of the credit card using the reflection of light according to another embodiment of the present invention.
7 is a schematic flowchart of a credit card recognition method for recognizing a credit card number according to an embodiment of the present invention.

Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings. Prior to this, terms or words used in the specification and claims should not be construed as having a conventional or dictionary meaning, and the inventors should properly explain the concept of terms in order to best explain their own invention. Based on the principle that can be defined, it should be interpreted as meaning and concept corresponding to the technical idea of the present invention.

Therefore, the embodiments described in the present specification and the configurations shown in the drawings are only the most preferred embodiments of the present invention and do not represent all the technical ideas of the present invention. Therefore, It is to be understood that equivalents and modifications are possible.

<1. System configuration>

1 is a schematic configuration diagram of a credit card number recognition system 1 according to an embodiment of the present invention, and FIGS. 2 and 3 illustrate a credit card number recognition system 1 according to an embodiment of the present invention. An example of a credit card to be recognized.

Referring to FIG. 1, the credit card number recognition system 1 according to an embodiment of the present invention includes a computer terminal 2. The computer terminal 2 further includes a card image storage means 21 for storing an image of a credit card, and a print number recognition means 22 for recognizing a print number on which the first four digits of the credit card are printed from the image of the credit card. And an embossed number recognition means 23 for recognizing a 16-digit number embossed from an image of a credit card, and a number comparator 24 for comparing the four-digit printed number with the first four-digit number.

The card image storage means 21 stores an image of a credit card in a DB. The image of the credit card may be generated via a scanner or camera. Here, the card image storing means 21 may process the size conversion into an image of a standard size (eg, a real size) in order to increase the recognition rate of the original image of the generated credit card.

Referring to FIG. 2, the image 100 of the illustrated credit card includes a print number 101 on which the first four digits of the card number are printed, a 16-digit embossed number 102 located above the print number 101, and an IC chip. 105. Here, the print number 101 is "4899", it can be seen that the first four-digit number of the 16-digit number of the embossed number 102 is "4899", the fourth four-digit number is "7004". The second and third four digits of the embossed number 102 are shown with all zero digits "0" for security of the card number.

The print number recognition means 22 recognizes the four digit print number 101 from the image 100 of the credit card. The print number is a unique 4-digit number assigned to each card company such as a VISA card or a master card.

Referring to FIG. 3, all credit cards around the world with IC chips 105 have common font information and layouts in accordance with standard formats. Preferably, the print number recognition means 22 may recognize the 4-digit area of the print number 101 from the card image 100 by using a relative distance recognition method or an absolute position recognition method. In the relative distance recognition method, after the print number recognition unit 22 identifies the rectangular shape of the IC chip 105, the area of the print number 101 is located at a relative distance from the position of the identified IC chip 105. It is recognized as an image area. The absolute position recognition method recognizes an area located in an area defined from the card image 100 as an image area of the print number 101 based on the standard format. The print number recognition means 22 separates the image of the recognized 4-digit print number 101 into the image of the individual number, and then compares the image of the separated individual number with the image of the stored individual number to print number 101. Recognize the four digits of the.

Here, the print number recognition means 22 performs polarization image processing while adjusting reference values of brightness and saturation on the image of the one-digit print number 101. In the polarization image processing, the print number recognition means 22 gives a brightness value higher than a brightness value above a reference value to a brightness value distributed in an image of the one-digit print number 101, and is relative to the brightness value below the reference value. By assigning a low brightness value to binary distribution of the brightness value. In addition, the chroma value is also binarized by giving a higher saturation value to a saturation value of the reference value or more and giving a saturation value of less than the reference value to a saturation value. Of course, if the print number recognition means 22 performs the polarized image processing on the image of the one-digit print number 101, the image is polarized in brightness and saturation, and thus the identification power is increased. When image processing of polarization is performed, the print number recognizing means 22 compares the image of the one-digit print number 101 subjected to the polarization image processing by matching with the image for each number of "0" to "9" previously stored in the DB. Recognize the number as a card number.

For reference, the print number 101 has the smallest font size among the font sizes printed on the card. However, the print number 101 has the highest recognition rate because it is printed on the card surface, so that the error number is hardly recognized. On the other hand, since the embossed number 102 is printed to protrude from the front face of the card, it has a relatively lower recognition rate than the printed number 101 due to light reflection and abrasion.

The embossed number recognition means 23 recognizes the 16-digit embossed number 102 from the image 100 of the credit card. First, the embossed number recognition means 23 identifies the image of the first four digit number 103 of the embossed number at a position defined based on the standard format. Next, the embossed number recognition means 23 separates the image of the identified four digit number 103 into the image of the individual number, and then performs the polarization image processing on the image of the separated individual number. The embossed number recognition means 23 divides the image of the one-digit embossed number with the polarized image processed into a plurality of unit areas, and then judges whether the divided unit areas are blank or filled and identifies them with the corresponding numbers.

Here, in the process of determining whether the unit area is empty, the embossed number recognizing means 23 obtains a degree of coincidence with respect to individual numbers from 0 to 9 for each one-digit image of the embossed number 102, and matches the most. The high number is recognized as the card number.

The number comparison unit 24 compares the recognized four-digit number of the printed number 101 with the first four-digit number 103 of the recognized embossed number to determine whether they match. Of course, the print number 101 and the first four digit number 103 must be the same number according to the standard format. Therefore, when it is determined that they do not match, the embossed number recognition means 23 can find the starting position while changing the position of the first four digit number 103 of the embossed number little by little. Incidentally, the embossed number recognizing means 23 performs the polarization image processing of changing the reference value of brightness and saturation until the first four digit number 103 of the embossed number matches the print number 101, while performing the polarization image processing. The recognizing process of 103 is repeated.

If the recognized print number 101 and the first four digit number 103 of the recognized embossed number are determined to be the same number, the embossed number recognizing means 23 performs a check for each of the remaining 12 digits of the embossed number 102. The polarization image process and the determination process of the unit area are performed to recognize the remaining 12 digit numbers. Here, the embossed number recognition means 23 recognizes the remaining 12 digit numbers using the reference values of brightness and saturation used when the first four digit number 103 coincides with the print number 101. Then, since the recognition process for the remaining 12 digit numbers is performed using the reference value when the first 4 digit number 103 is recognized, the recognition process can be faster. Of course, the embossed number recognition means 23 can change the reference value as necessary while processing the recognition of the remaining 12 digits.

Furthermore, the computer terminal 2 further includes a card number validity checker 25. The card number validity checking unit 25 checks the validity of the recognized 16-digit embossed number 102. The card number validity checking unit 25 may be provided with a test result after requesting a validity check by transmitting the embossed number 102 recognized as an external card company or an authentication authority.

Furthermore, the computer terminal 2 further includes a recognition number substituting unit 26 for performing a re-recognition process by replacing the card number until the validity check succeeds when the validity check of the card number fails. The recognition number substituting unit 26 performs the validity check again by substituting the number of the lowest matching number with the next matching degree number in the determination process of the unit area among the recognized 16 digit card numbers.

Meanwhile, the computer terminal 2 may be implemented as a client terminal in a local environment or as a server in a network environment. For example, when implemented as a client terminal, the client terminal reads the stored card number image one by one, recognizes the 16-digit card number from the read card image, and stores the recognition result including the recognized card number 16 digits. do. In addition, for example, when a server is implemented, a user takes a card image using a mobile phone terminal and transmits the card image to the server through a network, and the server recognizes a 16-digit card number from the received card image and receives a recognition result. The mobile phone terminal responds.

4A to 4J are exemplary views of analyzing a unit area of an embossed number in an image of a credit card according to an embodiment of the present invention.

The computer terminal 2 performs a recognition process on the 16-digit embossed number to recognize the 16-digit card number. Taking the " 0 " shown in Fig. 4A as an example, the embossed number recognition means 23 identifies the image of the 16-digit embossed number 102 from the card image, and identifies the 4-digit number from the 16-digit identified image. Identifies four images. Next, the embossed number recognition means 23 identifies an image of one digit number from the image of the four digit number. Then, the embossed number recognition means 23 identifies 13 unit area images from 1 to ① which are predefined for the one-digit image. Here, the unit regions may overlap each other. For example, the unit region ① includes the unit region ⑧ and overlaps with some regions of the unit regions ⑥ and ⑬. If the identified 13 unit region images match the pre-stored determination condition of "0", the degree of coincidence in which the one-digit image matches the number "0" is 100%. Here, the judgment condition of "0" is that ① is not blank, ⑤, ⑥, and 공백 are blank, and ⑦, ⑧, ⑨, ⑩, ⑪, and ⑬ are filled. The embossed number recognition means 23 can calculate the degree of agreement with the total points added by assigning points for each of the ten unit areas to be compared under the determination condition of the number " 0 ". Alternatively, it is also possible for the relief number recognition means 23 to calculate the ratio of the areas corresponding to the unit areas of the corresponding one digit number for each of the ten unit areas defined by the determination condition of " 0 ", to obtain a degree of agreement. In this way, the embossed number recognition means 23 obtains nine correspondences for each of the images of the one-digit number of the embossed number 102 from "0" to "9", and finds the number of the highest coincidence among the nine correspondences. Judging by the recognized one-digit card number.

As another example, using " 9 " shown in FIG. 4J as an example, the embossed number recognition means 23 analyzes an image of a single digit number to obtain nine degrees of correspondence for each number from " 0 " to " 9 ". If the coincidence of the number " 9 " is the highest among the nine degrees of coincidence, the embossed number recognition means 23 judges the image of the one digit number as " 9 ". Here, the judgment condition of "9" is that ① is blank, ②, ③, ④, ⑦, ⑨ and ⑩ are filled, and ⑤ and ⑥ are blank. For reference, when the embossed number of the credit card is 9 or 7, the numbers "9" and "7" are recognized as high match due to light reflection and abrasion. For example, when an image of one digit number is recognized as "9" because it has the highest degree of agreement with the number "9", the next higher number is often "7". Similarly, when an image of one digit number is recognized as "7", the number with the highest degree of agreement after "7" is often "9".

For the determination processing of the unit area, the computer terminal 2 stores area characteristic information indicating whether or not there is a space for the 13 unit areas by numbers "0" to "9" as a table.

FIG. 5 is an exemplary diagram showing a table indicating the degree of correspondence for each number of the analyzed image of relief numbers according to an embodiment of the present invention.

In the table shown, the rows represent the numbers "0" to "9", and the columns represent the 16-digit embossed numbers. For example, in the case of the third embossed number, the correspondence with the number "1" is the highest at 82%, followed by the number "0" at 76%. Herein, it is assumed that among the 16 degree embossed numbers, the degree of coincidence of the third embossed number is the lowest as 82%. It is assumed that the card number validity checking unit 25 has failed the validity check of the card number. Then, the recognition number replacement unit 26 selects the third number having the lowest match as the replacement target, and then repeats the validity check using the second highest matching "0" as the recognized card number instead of the number "1". do.

6 is an exemplary view of recognizing the embossed number of the credit card using the reflection of light according to another embodiment of the present invention.

In another embodiment of the present invention, the computer terminal 2 of FIG. 1 further includes reflective portion recognition means 27. The reflection part recognition means 27 adjusts a threshold value for the image of the one-digit embossed number to shade the remaining parts except for the bright part by being reflected by the light.

For example, in the case of "4899" embossed into the first four digits of the image of the credit card, the number '4' is recognized as two bright dots on the top and two bright dots on the middle. In addition, the number '9' is recognized by two bright dots on each side and one bright dot on the middle left. As such, the numbers "0" to "9" are different from each other in the position distribution of the bright spots due to the reflection of light, so that the embossed numbers may be identified using the pattern of the position distribution.

Therefore, the computer terminal 2 is located at the position of the bright point recognized by the reflection of the light by the reflective part recognition means 27 instead of the embodiment of determining whether the unit area is empty by the embossed number recognition means 23. The embossed number may be identified using an embodiment for determining. Alternatively, the computer terminal 2 performs the embodiment using the unit area by the embossed number recognition means 23 and the embodiment using the reflection of the light by the reflection part recognition means 27. In comparison, it is possible to increase the verification rate of the card number. Alternatively, the computer terminal 2 may perform either embodiment first, and if the card number validity check fails more than a predetermined number of times, the other embodiment may be performed to compensate for errors in the recognition process. Do.

<2. Method composition>

The credit card number recognition method for recognizing the credit card number according to an embodiment of the present invention can be preferably realized through the construction of the credit card number recognition system 1 described above.

7 is a schematic flowchart of a credit card recognition method for recognizing a credit card number according to an embodiment of the present invention.

The computer terminal 2 stores the image of the credit card in the DB (S21). Here, the computer terminal 2 may set the standard size of the image and adjust the size of the image to fit the standard size. Images of credit cards can be generated by cameras or scanners, and are not limited to any particular method of creation.

The computer terminal 2 reads the card image from the DB, and recognizes four digits of the print number 101 from the card image (S22).

Here, the computer terminal 2 may first recognize the rectangular shape of the IC chip 105, and then extract a four-digit image of the print number 101 at a relative position from the recognized IC chip 105. It is also possible to extract the area image of the print number 101 using the absolute position defined in the standard format of the credit card. When the four digit image is obtained, the computer terminal 2 generates an image of the one digit print number 101 by dividing into four. When the one-digit print number 101 is generated, the computer terminal 2 performs polarization image processing on the image of the one-digit print number 101 while adjusting the reference values of brightness and saturation. Due to the polarization image processing, the background area and the number area of the one-digit image are clearly distinguished and the discriminating power is maximized. Thereafter, the computer terminal 2 compares the polarized image-processed one-digit image with the image of the one-digit print number stored in the DB and recognizes the matching number as the card number. If the computer terminal 2 repeats the above recognition process four times, it can recognize the four-digit card number. According to the standard format of the credit card, the four digits of the card number recognized from the print number 101 necessarily match the first four digits of the embossed character.

When the card number is recognized, the computer terminal 2 recognizes the first four digit number 103 from the image of the 16 digit embossed number 102 (S23).

Here, the computer terminal 2 extracts the area image of the 16-digit embossed number 102 using the absolute position of the embossed character 102 promised by the standard format. The computer terminal 2 generates an image of the first four digit number 103 from the sixteen digit image of the embossed number 102. The computer terminal 2 generates an image of the one-digit embossed number by dividing the image of the first four-digit number 103 into four equal parts. When the image of the one digit embossed number is generated, the computer terminal 2 performs the polarization image processing on the generated one digit image. The computer terminal 3 divides the polarized image-processed one-digit image into 13 unit regions as illustrated in FIGS. 4A to 4J. The computer terminal 3 determines whether the space is empty by referring to the determination information of the unit area for each number of "0" to "9" stored in the DB, for the divided unit image of the unit area, and the degree of agreement according to the number according to the determination result. Calculate and determine the card with the highest matched number. In the present invention, there may be a variety of methods such as the point calculation method for each unit area and the area calculation method for each unit area in the method of calculating the degree of agreement, and there is no limitation to a specific calculation method. If the computer terminal 2 repeats the determination process of the unit area four times with respect to the first four digit number 103 of the embossed character, the first four digit embossed number can be recognized.

When the first four digit number 103 of the embossed number is recognized, the computer terminal 2 checks whether the four digit card number recognized from the print number 101 and the four digit card number recognized from the first four digit number 103 match. Comparison is made (S24). If the comparison result does not match, the computer terminal 2 adjusts the start position of the embossed number 102 or repeats the determination process of the unit area until the comparison result matches after performing the polarization image process. do.

If the comparison result matches, the computer terminal 2 recognizes the card number by performing the polarization image process and the determination process of the unit area on the remaining 12 digits among the 16 digits of the embossed number 102. (S231).

Here, the computer terminal 2 performs polarization image processing on the 12 digit number using the reference values of brightness and saturation used in the step S23, and adjusts the reference value as necessary while performing the polarization image processing. Can be.

When the 16-digit card number is recognized from the embossed number 102, the computer terminal 2 performs the validity check of the recognized card number (S25). The validity check may be performed via a local method or a network method.

Thereafter, the computer terminal 2 determines the test result of the validity test (S251). If the validity check is successful, the process of card recognition is terminated.

If the validation fails, the computer terminal 2 substitutes the next matched number for the lowest matched card number among the remaining 12 digits except the first 4 digits, and succeeds in validating the card. Repeat until. In some cases, a validity check may be performed for a number of two or more digits while the numbers are replaced according to the degree of agreement. For example, the validity check may be performed while replacing the numbers in the order of the degree of agreement for the 5th digit number and the 16th digit number.

On the other hand, the computer terminal 2 is an embodiment of determining the position of the bright portion by using the above-described reflection of light instead of the embodiment of determining whether the unit area is empty in the step (S22) and the step (S231). It may be replaced. In addition, the computer terminal 2 may compare each recognized card number after implementing each embodiment.

In the above-described embodiment, the term "~ part" is not used to mean a hardware classification of the credit card number recognition system. Accordingly, a plurality of constituent parts may be integrated into one constituent part, and one constituent part may be divided into a plurality of constituent parts. Also, the constituent unit may mean a hardware constituent unit, but may mean a constituent unit of software. Therefore, it should be understood that the present invention is not particularly limited by the term "part ".

While the present invention has been particularly shown and described with reference to exemplary embodiments thereof, it is to be understood that the invention is not limited to the disclosed exemplary embodiments. It goes without saying that various modifications and variations are possible within the scope of equivalence of the scope.

1: credit card number recognition system 2: computer terminal
21: card image storage means 22: print number recognition means
23: embossed number recognition means 24: number comparison unit

Claims (25)

Card image storage means for storing a card image photographed from a credit card;
Print number recognizing means for recognizing a print number on which a first four digit number is printed from a card number of 16 digits from the card image;
Recognize the first four-digit embossed number located above the print number among the sixteen-digit embossed numbers of the card image until it matches the recognized print number, and if the print number and the first four-digit embossed number match, the 16 Embossed number recognizing means for recognizing the remaining 12-digit embossed numbers among the digit embossed numbers; And
A number comparison unit for comparing the recognized print number and the first four-digit embossed number to determine whether they match.
And a computer terminal comprising a credit card number recognition system. The method of claim 1,
The computer terminal,
Performing image processing to polarize the image of the individual number identified from the card image while adjusting the reference values of brightness and saturation, and then recognize the card number from the image of the individual number which has been processed. Number recognition system. 3. The method according to claim 1 or 2,
The print number recognition means,
A relative distance recognition method for recognizing an area of the print number located at a relative distance from an IC (chip) location of the card image; And
Absolute position recognition method for recognizing the area of the print number located in the predefined area in the card image
After recognizing the area of the print number by using any one of the recognition method of the image, and comparing the image of the individual number identified from the recognized area of the print number and the image of the pre-stored number to recognize the matching print number Number recognition system of a credit card. The method of claim 1,
The computer terminal,
And dividing an image of an individual embossed number into a plurality of unit areas, and determining whether each of the divided unit areas is blank or filled, and recognizing the number as a number from 0 to 9. 5. The method of claim 4,
The computer terminal,
And a table in which information on whether each unit area is blank or filled is stored for a plurality of unit areas corresponding to each of the numbers 0 to 9. 5. The method of claim 4,
The computer terminal,
A number recognition system of a credit card, characterized in that each unit of the embossed number is compared with each other by comparing the space of the unit area to obtain a degree of coincidence with respect to numbers 0 to 9, and that the number with the highest coincidence is recognized as an embossed number. The method according to claim 1 or 4,
The embossed number recognition means,
Recognizing a credit card number, the image processing of polarization is repeated for each individual digit of the first four digit embossed number while adjusting the reference value of brightness and saturation until the first four digit embossed number matches the print number. system. The method according to claim 1 or 4,
The embossed number recognition means,
A number of a credit card characterized in that polarization image processing is performed for each individual digit of the 12-digit embossed number by using reference values of brightness and saturation used when the first four-digit embossed number corresponding to the print number is recognized. Recognition system. The method according to claim 1 or 4,
The computer terminal,
If the 16-digit embossed number is recognized as a card number, further comprising a card number validity checker for validating the card number. The method of claim 9,
The computer terminal,
If the validation fails, further comprising a recognition number replacement unit for repeating the validation by replacing the number with the next matching degree for the embossed number with the lowest match among the 12-digit embossed number A credit card number recognition system. The method according to claim 1 or 4,
The computer terminal,
For each image of each embossed number, shade the remaining portions except for the bright portion above the threshold due to the reflection of light, and then use the position information of the bright portion to obtain the degree of coincidence of the numbers 0 to 9, respectively. And crediting part recognition means for recognizing the highest number as an embossed number. 12. The method of claim 11,
The computer terminal,
12. A credit card number recognition system, comprising: determining whether a unit area is empty and determining a recognized 12-digit number and location information of a bright part by comparing the recognized 12-digit number. The method according to claim 1 or 4,
The computer terminal,
Can be built as a client terminal or network server,
The credit card number recognition system of claim 1, wherein the card server receives a card image through a network and responds to a card number recognition result. A credit card number recognition method processed by a computer terminal that recognizes a card number from a card image of a credit card,
(a) a card image storing step of storing a card image photographed from a credit card;
(b) a printing number recognition step of recognizing a printing number on which a first four digit number is printed among 16 digit card numbers from the card image;
(c) a four-digit embossed number recognizing step of recognizing a first four-digit embossed number located above the printed number among the sixteen-digit embossed numbers of the card image until it matches the recognized print number;
(d) a number comparison step of comparing the recognized print number with the first four digit embossed number to determine whether they match; And
(e) a 12-digit embossed number recognizing step of recognizing the remaining 12-digit embossed number of the 16-digit embossed number if the print number and the first four-digit embossed number match
The credit card number recognition method comprising a. The method of claim 14,
Step (b), step (c) and step (e) above,
The computer terminal performing image processing to polarize the image of the individual number identified from the card image while adjusting reference values of brightness and saturation, and then recognizing the card number from the image of the individual number processed. Characterized by the credit card number recognition method. 16. The method according to claim 14 or 15,
The step (b)
A relative distance recognition step of recognizing an area of the print number located at a relative distance from an IC (chip) location of the card image; or
An absolute position recognition step of recognizing an area of the print number located in a predefined area of the card image
After recognizing the area of the print number by using a credit card, characterized in that by comparing the image of the individual number identified from the recognized area of the print number and the image of the pre-stored number to recognize the matching print number Number recognition method. 15. The method of claim 14,
Step (c) and step (e) above,
Wherein the computer terminal divides the image of the individual embossed number into a plurality of unit areas, and determines whether each of the divided unit areas is blank or filled and recognizes the number as 0 to 9; Number recognition method. 18. The method of claim 17,
The computer terminal,
And a table in which information on whether each unit area is blank or filled is queried for a plurality of unit areas corresponding to each of the numbers 0 to 9. The method according to claim 14 or 17,
The computer terminal,
A method of recognizing a number of credit cards, comprising: comparing each space of an individual embossed number to obtain a degree of coincidence with respect to numbers 0 to 9, and recognizing a number having the highest coincidence as an embossed number. . The method according to claim 14 or 17,
The step (c)
And repeating, by the computer terminal, polarizing image processing for each individual digit of the first four digit embossed number while adjusting the reference value of brightness and saturation until the first four digit embossed number matches the print number. How to recognize the number of your credit card. The method according to claim 14 or 17,
The step (e)
A number of a credit card characterized in that polarization image processing is performed for each individual digit of the 12-digit embossed number by using reference values of brightness and saturation used when the first four-digit embossed number corresponding to the print number is recognized. Recognition method. The method according to claim 14 or 17,
and (f) the computer terminal further comprising a card number validity checking step of validating the card number when the 16-digit embossed number is recognized as the card number. 23. The method of claim 22,
(g) If the validity check fails, the computer terminal repeatedly performs the validity check by replacing the embossed number with the lowest match among the 12-digit embossed numbers with a number having the next highest degree of agreement. The method further comprises the step of recognizing a credit card number. The method according to claim 14 or 17,
After the step (b)
The computer terminal shades the remaining portions of the images of the individual embossed numbers except for the bright portion above the threshold due to the reflection of light, and then uses the position information of the bright portion to determine the degree of correspondence of the numbers 0 to 9. And a reflection position recognizing step of recognizing a number having the highest match as an embossed number. 25. The method of claim 24,
The computer terminal may further include determining whether the unit area is empty and determining the recognized 12-digit number and the location information of the bright part to compare the recognized 12-digit number to determine whether the credit card is identical. Number recognition method.

Images