KR20040076962A - sampling system for extracting a living body code from features and the iris of the eye - Google Patents

Abstract

PURPOSE: A system for extracting a complex biometric code combining a face and an iris is provided to enhance a speed and the stability for classifying/searching the biometric code needed for identification by offering the encrypted complex biometric code combining the face and the iris extracted from a user. CONSTITUTION: An input part(10) forms a face information input part(11) and an iris information input part(12). A filter(20) corrects a regulated image after a process for optimizing the biometric information inputted from the input part. An information extractor(30) forms a face recognizer(31) and an iris recognizer(32). A classifier(40) generates a classification value for separating the information from a large scale of information fast and precisely based on a common group classification value. A code generator(50) generates the biometric code by encrypting the classification value, the face feature value, and a permanent feature value of the iris. An output part(60) outputs the biometric code generated from the code generator to diverse users needing the security and the certification through the wire/wireless communication network.

Description

Sampling system for extracting a living body code from features and the iris of the eye}

The present invention relates to a biometric code extraction system required for identification, and more particularly to combining the feature values of the user's face and iris extracted from the user's biometric information to improve the search and stability of the biometric code required for identification. The present invention relates to a complex biocode extraction system combining a face and an iris.

Generally, major national facilities such as roads, harbors and airports, or private facilities such as banks and corporate research institutes use identification cards such as social security cards, driver's licenses, passports, student ID cards, ID cards, etc. to manage or control security and access. In particular, cash dispensers, electronic payment systems, online reservation systems such as trains and airplanes, and online shopping systems issue credit cards, certificates, etc. to verify the user's identity unattended and use it only for users who have been given permission to use it. An identity verification system is being used.

However, if you use your ID card, driver's license, passport, or other credit card to verify your identity as described above, you have the risk that your information may be stolen by others. Theft damage is exponential.

Therefore, in order to solve the above problems, conventionally, by using the biometric information such as the user's face or iris to verify the identity to prevent the theft of personal information, in particular, the life of international crime or terror caused by theft of passport, etc. Damage and property damage are prevented beforehand.

However, in the case of identifying the identity using the face information of the user as described above, there is an advantage that can improve the classification and retrieval speed of the identity, while reducing the most important precision and stability in identification.

In addition, when verifying an identity using a user's iris, there is an advantage in that the accuracy and stability of biometric information required for identification is improved, while the classification and retrieval speed of the identity is relatively low.

The present invention has been made in view of the above-described problems, and its object is to provide a complex biometric code that is encrypted by combining the feature values of the user's face and iris extracted from the biometric information of the user, thereby providing the biometric code necessary for identification. This is to improve the classification and retrieval speed of and to improve the stability.

1 is a block diagram illustrating a complex biometric code extraction system combining a face and an iris according to an embodiment of the present invention.

* Description of the symbols for the main parts of the drawings *

10: input unit 11: face information input unit

12: iris information input unit 20: filtering unit

30: information extraction unit 31: face recognizer

32: iris recognizer 40: classification unit

50: code generation unit 60: output unit

The configuration of the present invention for achieving the above object is made as follows.

An input unit including a face information input unit for inputting face information of the user and an iris information input unit for inputting iris information of the user;

A filtering unit for performing correction on normalized images after optimizing biometric information input through the face information input unit and the iris information input unit of the input unit;

Extract the feature values of the face possessed by the user from the biometric information corrected through the filtering unit, classify the common feature series faces from various faces through the face feature values, and classify the common sequence class values for each face of the common feature series And a face recognizer providing a feature value, and an iris recognizer to search for the biometric information corrected through the filtering unit to find an invariant value that is a feature of the iris, and to extract an invariant feature value that may represent the identity of the person. An information extraction unit;

A classification unit for generating a classification value for quickly and accurately separating information from a large amount of information based on the common series classification value extracted through the information extraction unit;

A code generator for generating a biometric code by encrypting a classification value generated by the classification unit, a facial feature value extracted by the face recognizer, and an invariant feature value of the iris extracted by the iris recognizer;

Characterized in that it comprises an output unit for outputting the biometric code generated by the code generator to a variety of users requiring security and authentication through a wired or wireless communication network.

Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the accompanying drawings.

1 is a block diagram illustrating a complex biometric code extraction system combining a face and an iris according to an embodiment of the present invention.

As shown in FIG. 1, the complex biometric code extraction system combining the face and the iris of the present invention includes an input unit 10 for inputting biometric information such as a user's face and iris into an image, and the image information input through the input unit 10. An information extracting unit 30 for extracting feature values such as a face and an iris of a corresponding user from information corrected by the filtering unit 20, the filtering unit 20, and the information extracting unit 30. Generating a biometric code consisting of a classification unit 40 for classifying the extracted information based on the common feature series value, and the classification value and the constant feature values of the face and iris classified through the classification unit 40 The code generation unit 50 and the output unit 60 for outputting the biometric code generated by the code generation unit 50 to the outside through a wired or wireless communication network.

The input unit 10 forms a face information input unit 11 for inputting face information of the user and an iris information input unit 12 for inputting iris information of the user.

For example, the face information input unit 11 preferably uses a CCD camera, and in particular, the iris information input unit 12 preferably uses a low light high recognition CCD camera capable of inputting the pupil of the eye and the iris. Do.

The filtering unit 20 is responsible for preprocessing and filtering biometric information, such as a face and an iris, input through the input unit 10. The filtering unit 20 performs a correction operation on a normalized image after optimizing the biometric information.

The information extracting unit 30 extracts a feature value of a face possessed by the user from the biometric information corrected by the filtering unit 20, and classifies common feature series faces from various faces through the face feature value. The face recognizer 31 which provides a common sequence classification value and a feature value for each face of the common feature series and the biometric information corrected through the filtering unit 20 are searched to find an invariant value that is a characteristic of the iris, An iris recognizer 32 is formed to extract an invariant feature value that may represent the identity of the person.

The classification unit 40 generates a classification value having a header function for quickly and accurately separating information from a large amount of information based on the common series classification value extracted through the information extraction unit 30.

The code generator 50 may generate a classification value generated by the classifier 40, a facial feature value extracted by the face recognizer 31, and an invariant feature value of the iris extracted by the iris recognizer 31. Generate the biometric code by encryption.

The output unit 60 outputs the biometric code generated by the code generation unit 50 to various users who need security and authentication through a wired or wireless communication network.

Referring to the operation of the biometric code extraction system combining the face and the iris of the present invention as described above are as follows.

First, when a user inputs his or her biometric information to the input unit 10 in order to extract the biometric code required for security and authentication of the user, the face information input unit 11 and the iris information input unit 12 of the input unit 10 The image of the user's face and the iris is output to the filtering unit 20, respectively.

In this case, the filtering unit 20 performs an optimization process after optimizing an image of a face and an iris of the user input through the input unit 10, and applies corresponding biometric information to the information extracting unit 30. Output

Subsequently, the face recognizer 31 of the information extractor 30 retrieves face information from the biometric information passed through the filtering unit 20, extracts feature values of the face possessed by the user, and then features the face. Comparing / analyzing the values to classify the common features in the various faces of the users to generate a common feature sequence value and outputs to the classifier 40.

At the same time, the iris recognizer 32 of the information extraction unit 30 searches for the iris information from the biometric information passed through the filtering unit 20 to find an invariant value that is characteristic of having the iris, and then identifies the user. Extracts an invariant feature value that may represent and outputs it to the classification unit 40.

Subsequently, the classification unit 40 analyzes the information applied from the face recognizer 31 and the iris recognizer 32 of the information extractor 30 to quickly generate a large amount of information based on a common sequence classification value. A classification value having a header function that accurately separates information is generated and output to the code generation unit 50.

In this case, the code generator 50 may include a classification value generated through the classifier 40, a facial feature value extracted through the face recognizer 31, and an iris extracted through the iris recognizer 31. The biometric code encrypted by the invariant feature value using Triple DES 128 bit or SEED is generated and output to the output unit 60.

Subsequently, the output unit 60 outputs the biometric code encrypted through the code generation unit 50 to various users who need security and authentication through a wired / wireless communication network, that is, a public institution or a financial institution.

As described above, the present invention extracts feature values such as the user's face and iris from the user's biometric information, classifies the extracted facial feature values based on the common feature series, and then classifies the face and the classification values. Encoding complex feature codes of face and iris by encoding feature values and invariant feature values of iris improves the retrieval and stability of biometric codes required for identification of high complexity large unit environments.

In addition, by using the biometric code to verify the identity, it is possible not only to prevent damage caused by counterfeiting of resident registration card, passport, financial card (credit card, transportation card, cash card), etc. And property damage can be prevented.

Claims (1)

An input unit 10 including a face information input unit 11 for inputting face information of the user and an iris information input unit 12 for inputting iris information of the user; A filtering unit 20 for correcting the normalized image after optimizing the biometric information input through the face information input unit 11 and the iris information input unit 12 of the input unit 10; The feature value of the face possessed by the user is extracted from the biometric information corrected by the filtering unit 20, and the common feature sequence faces are classified from various faces through the face feature value, and for each face of the common feature sequence The face recognizer 31 which provides a common sequence classification value and a feature value, and the biometric information corrected through the filtering unit 20 may be searched to find an invariant value that is a characteristic of an iris, and may indicate an identity of a corresponding person. An information extracting unit 30 forming an iris recognizer 32 for extracting an invariant feature value; A classification unit 40 for generating a classification value for quickly and accurately separating information from a large amount of information based on the common series classification value extracted through the information extraction unit 30; Generating biometric code by encrypting the classification value generated by the classifier 40, the facial feature value extracted by the face recognizer 31, and the constant feature value of the iris extracted by the iris recognizer 32. A code generator 50; Complex biometric code combining the face and iris, characterized in that the output unit 60 for outputting the biometric code generated by the code generation unit 50 to a variety of users requiring security and authentication through a wired or wireless communication network Extraction system.