KR20200021848A - System and Method for Judging Biometric Recognition and Fake Biometric Detection of Mobile Device - Google Patents

Abstract

Disclosed are biometric recognition performance and forgery defense capability analysis system of a mobile terminal and a method thereof. According to the present invention, forgery defense capability analysis system can evaluate biometric recognition performance of a biometric recognition sensor and forgery defense capabilities for distinguishing a forged living body. To this end, a mobile terminal acquires a sample image from a sample for examination in an examination mode and provides the sample image to an analysis system only when the sample for examination is a living body.

Description

System and Method for Judging Biometric Recognition and Fake Biometric Detection of Mobile Device}

The present invention relates to an analysis system capable of evaluating biometric performance of a biometric sensor such as a fingerprint, a face, and an iris mounted on a mobile device, and analyzing a counterfeit defense force capable of detecting a forgery.

As society has all its functional elements mechanized and networked, authentication systems are developing to restrict people's access. Since passwords and magnetic cards, which are widely used in the related art, are vulnerable to security vulnerabilities and loss, biometrics using biometric features are being developed.

Biometric authentication is unique to each individual and is not only an excellent means of authentication, but also does not cause a problem of portability or loss, it is relatively good security. To date, fingerprints are widely used for biometric authentication, but recently, face recognition and iris recognition have also been developed.

The fingerprint remains the same from birth to death, and the probability of having the same fingerprint as another person is almost zero. In addition, the fingerprint sensor has a feature that can be made smaller than other biometric sensors. Facial recognition is very convenient because the user only needs to stare at the sensor, that is, the camera without any other motion, and does not need any other physical contact. However, there is a drawback that the face is susceptible to surrounding conditions such as lighting and the face itself may change with time. Iris recognition is known to be the most accurate means of authentication because it is not changed for life after being completed after approximately 18 months, and is almost impossible to replicate. However, since the iris must be close to the camera and the light, the user may feel a sense of rejection, and the product size is large.

Personal authentication based on recent biometrics goes beyond simply controlling access to open doors, accessing data online, personal authentication such as banks, online payments, and data security on personal mobile devices such as mobile phones or tablets. The range of use is getting wider. With this trend, biometric means are being installed in personal mobile devices. However, since the mobile device is limited in size, the biometric means for mobile device is also limited in size. Therefore, it is a new task to improve the recognition performance with miniaturization.

On the other hand, one of the vulnerabilities of biometric authentication is related to fake biometrics. Normal personal authentication is mainly used in areas where security is important, such as access control, e-commerce, financial transactions, security of personal computers (PCs), and office payment systems. The most important thing is to effectively distinguish between living organisms (hereinafter referred to as 'false living organisms'). For example, when attempting fingerprint authentication using printed paper or silicon, the fingerprint pattern should be able to distinguish whether the fingerprint is a fake fingerprint. The ability of biometric sensors to distinguish counterfeit organisms is called counterfeit defense.

Therefore, when evaluating the quality of the biometric sensor, it is important to evaluate the counterfeit defense along with the biometrics.

Detection algorithms for biometric authentication or counterfeit defense are not standardized, and thus are areas that need to be developed by manufacturers. Not all manufacturers are superior in technology, and as they are miniaturized specifically for mobile, some manufacturers produce certified sensors without basic counterfeit protection.

Therefore, it is necessary to check whether the biometric device has basic anti-counterfeit protection as a national institution that performs approval according to various laws with respect to the biometric device.

[Related Prior Art]

Republic of Korea Patent Application Publication No. 10-2008-0062217 (Invention name: Method for measuring the quality of the fingerprint image using the block-level quality measurement value and its recording medium)

This invention uses a method of checking the quality of a fingerprint area (ridge) itself acquired as a fingerprint image using pixel values. To this end, the present invention measures the quality of the fingerprint image by dividing the fingerprint image into a plurality of blocks and obtaining the average of the pixel values of each block. Above all, this invention relates to the measurement of the quality of the fingerprint image itself, not to the method for determining the quality of the fingerprint sensor.

Disclosure of Invention An object of the present invention is to provide an analysis system capable of evaluating biometric performance in a biometric sensor such as a fingerprint, a face, and an iris mounted on a mobile device, and analyzing a counterfeit defense force capable of detecting a forged biometric. In providing.

Analysis system according to the present invention for achieving the above object can analyze the biometric performance and counterfeit defense of the mobile device. The analysis system of the present invention includes a terminal connection unit, feature point extraction unit and counterfeit defense force analysis unit. The terminal connection unit is connected to the mobile terminal to be inspected and receives a sample image from the mobile terminal. In this case, the mobile terminal acquires a sample image from the test sample to perform bio-authentication, and when it is determined that the test sample is a living body, provides the sample image to the terminal connection unit.

The feature point extracting unit extracts feature point data from the sample image received by the terminal connection unit, and the counterfeit defense force analysis unit queries the feature point data extracted by the feature point extracting unit to a database to determine whether the sample image is an image of a living body or a fake animal. Analyze the counterfeit defense of the mobile terminal.

According to an embodiment, the mobile terminal may include a biometric sensor for obtaining a sample image from the test sample by fingerprint recognition, face recognition, or iris recognition method, and biometric recognition based on the sample image acquired by the biometric sensor. And a recognition control unit for performing a forgery determination and a system connection unit for proceeding to a test mode when connected to the terminal connection unit. The system connection unit provides the sample image to the terminal connection unit when the recognition control unit judges that the test sample is not a forgery during the inspection mode, and the sample when the recognition control unit judges the test sample as a forgery. The image is not provided to the terminal connection unit.

According to another embodiment, the database registers and manages sample images for the test sample. The registered sample image includes a sample image obtained from a living body and a sample image obtained from a fake organism. .

According to another embodiment, the analysis system of the present invention finally outputs the quality measuring unit for evaluating the quality of the sample image received by the terminal connection unit, the quality score of the quality measuring unit and the counterfeit defense force score of the counterfeit defense analysis unit It may further include a reporting unit.

The present invention also extends to the analysis method of the analysis system. Analysis method of the present invention comprises the steps of connecting to the mobile terminal to be inspected through the terminal connection unit; Receiving a sample image from the mobile terminal; Extracting feature point data from a sample image received from the mobile terminal; And querying the extracted feature point data in a database to determine whether the sample image is an image of a living body or an image of a counterfeit organism.

The biometric sensor analysis system according to the present invention may determine biometric capability and counterfeit defense in a mobile device equipped with a biometric sensor as an inspection object.

Basically, the present invention passes the test by analyzing and determining the performance of the biometric sensor of the mobile device unit and at the same time, by providing a software module that can distinguish basic counterfeit organisms in the absence of a defense capable of distinguishing counterfeit organisms. A biometric sensor mounted on a mobile device ensures basic counterfeit protection.

1 is a conceptual diagram of a biometric sensor analysis system for a mobile device of the present invention;
2 is a block diagram of a mobile terminal to which the present invention is applied;
3 is a block diagram of an analysis system of the present invention, and
Figure 4 is a flow chart provided in the description of the biometric performance and counterfeit defense analysis method for the mobile device of the analysis system of the present invention.

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

1 to 3, the biometric sensor analysis system 110 for a mobile device of the present invention is connected to a mobile terminal 130 equipped with a biometric sensor module and mounted on a mobile terminal 130. Analyze the sensor module's ability to recognize and counterfeit. In the database 111 provided in the biometric sensor analysis system 110, information about various test samples X is recorded and managed, and the test sample x is input to the biometric sensor of the mobile terminal to perform the test. To perform.

Mobile devices

Referring to FIG. 2, the mobile terminal 130 is a device which is portable by a person and corresponds to devices of relatively small size. For example, a mobile phone, a tablet, a personal digital appliance (PDA), and the like correspond to the mobile terminal 130 of the present invention.

However, the mobile terminal 130 connected to the analysis system 110 of the present invention includes a biometric sensor 201, a recognition control unit 203, a system connection unit 205, and a communication interface 207. The biometric sensor 201 and the recognition controller 203 perform a biometric process inherent in the mobile terminal 130, and are a target of actual analysis by the analysis system 110 of the present invention.

The biometric sensor 201 is for fingerprint recognition, iris recognition, or face recognition, and generates a sample image (a biological image or a fake biological image) from an input test sample (biological or forged biometric). The sensor 201 for fingerprint recognition is a fingerprint sensor for obtaining a fingerprint image, and the sensor 201 for iris or face recognition is a camera for photographing an iris or face. In the case of the mobile terminal 130, it is difficult to have a separate camera for capturing an iris or a face, and it is common to use a built-in camera mounted as a basic specification in a mobile device.

The test sample may be a living organism or a fake living organism. For example, fake fingerprints are made by molding or printing a fingerprint pattern on silicone, gelatin, rubber, film, paper, and the like. For face recognition. The counterfeit sample may use a face photograph printed in color, black and white or infrared (IR), and the face photograph is input to the camera. For iris recognition. The counterfeit sample can use a model made of fake eye (bill) material, and input the iris model into the camera.

The recognition controller 203 performs biometric recognition (fingerprint recognition, face recognition or iris recognition) from the sample image acquired by the biometric sensor 201 and determines whether or not a fake living body. Therefore, the recognition control unit 203 is implemented as 'bio recognition software' for bio-recognition and 'counterfeit discrimination software' for forgery discrimination, and may be mounted in an operating system program of the mobile terminal 130. In some cases, the forgery discrimination may be performed by a forgery discrimination software of the recognition controller 203 and a hardware configuration of the biometric sensor 201.

On the other hand, when the recognition control unit 203 is not equipped with the 'counterfeit defense force' by providing the basic 'counterfeit discrimination software' held by the inspector in advance before the test, so that the mobile terminal 130 completed the final inspection to have a basic counterfeit defense force Can be.

The biometric sensor 201 and the recognition controller 203 perform a biometric process unique to the mobile terminal 130. However, when the test system is connected to the analysis system 110 and proceeds to the test mode, the recognition controller 203 receives a test sample according to a biorecognition algorithm, generates a sample image, and checks the quality of the input sample and the counterfeit organism. It determines whether or not, and provides the recognition result to the system connection unit 205. Recognition results include (1) 'recognition success' which is determined to be not a counterfeit organism, and (2) 'perception failure by forgery' which is determined to be a counterfeit organism.

The system connection unit 205 processes a software protocol for accessing the analysis system 110, and is typically implemented as an API (Application Programming Interface) provided by the analysis system 110. Therefore, when the API provided by the analysis system 110 of the present invention is mounted on the conventional mobile device, the system connection unit 205 is installed.

The system connection unit 205 does not participate in the biometric process unique to the mobile terminal 130 and operates when connected to the analysis system 110. When connected to the analysis system 110 and proceeds to the 'test mode' controlled by the analysis system 110, the system connection unit 205 receives the biometric result from the recognition control unit 203. If the recognition result provided from the recognition control unit 203 is 'recognition success', the system connection unit 205 provides the analysis image 110 through the communication interface 207 the sample image of the inspection object. When providing the sample image to the analysis system 110, the system connection unit 205 provides the sample image in the form of a stream, and usually at a high speed (e.g., 16 frames per second) to smoothly advance the inspection mode. It is desirable to send.

If the recognition result provided from the recognition control unit 203 is a 'failure recognition failure', the system connection unit 205 notifies the analysis system 110 without providing the sample image to the analysis system 110, and then the next inspection sample. Wait for the recognition result.

The communication interface 207 includes a physical interface that can connect to the analysis system 110. As the physical interface, for example, a wireless interface capable of connecting to a wireless network such as a wireless LAN or a 4G / 5G mobile communication network is possible, and a wired interface such as USB can be used, and a mobile device ( The communication interface built in 130 can be used as it is. The system connection unit 205 provides the image and / or recognition result obtained by the biometric sensor 201 to the analysis system 110 using the communication interface 207.

Analysis system

Analysis system 110 of the present invention analyzes the biometric capability and counterfeit defense of the mobile terminal 130 connected for the test. Referring to FIG. 3, the analysis system 110 includes a database 111, a network interface 301, a terminal connection unit 303, a quality measuring unit 305, a feature point extraction unit 307, and a counterfeit defense force analysis unit 309. And a reporting unit 311.

In the database 111, information on a test sample is registered. Here, the information about the test sample includes those obtained from living organisms and those obtained from fake organisms. The sample information includes a sample image, feature point data extracted from the sample image, a registration code of each sample image, and the like. The feature point data is the same algorithm as the algorithm in which the feature point extractor 307 described below extracts the feature point from the sample image. It is extracted.

Registering sample images extracted from various counterfeit samples in the database 111 is advantageous for improving the quality of inspection. On the other hand, the database 111 does not simultaneously register the biological image and the fake biological image from which the same feature point is extracted.

The network interface 301 is a component for connecting the mobile terminal 130 to be examined to the analysis system 110 of the present invention, and is connected to the communication interface 207 of the mobile terminal 130. Accordingly, the network interface 301 may be a wired interface such as a USB network or may be a wireless interface that can be connected to a wireless LAN or a mobile communication network.

The terminal connection unit 303 receives a recognition result from the system connection unit 205 of the mobile terminal 130. If the recognition result is 'recognition success' as the mobile terminal 130 determines that the sample is not a forgery, the recognition result provided by the mobile terminal 130 includes a sample image of the sample for inspection. The terminal connection unit 303 provides the sample image received from the mobile terminal 130 to the quality measuring unit 305 and the feature point extraction unit 307.

The quality measuring unit 305 calculates a quality score by measuring the quality of the sample image provided by the terminal connection unit 303. The quality measuring unit 305 may not be essential in the analysis system 110 or the analysis method of the present invention.

The quality measurement method of the sample image of the quality measuring unit 305 is evaluated using a sample image quality evaluation algorithm. The sample image quality evaluation algorithm may use a conventionally known method for evaluating the quality of a fingerprint image, a face image, or an iris image, and the quality of the sample image may be changed in a plurality of steps, for example, It is evaluated in five levels: Grade, Good (Level 3), Fair (Level 4), and Poor (Level 5).

Even if the biometric sensor 201 of the mobile terminal 130 determines that the recognition is successful, the quality of the sample image may not reach a certain level. Therefore, it is preferable to perform the quality evaluation of the quality measuring unit 305.

The feature point extractor 307 extracts the feature point from the sample image provided by the terminal connection unit 303. As described above, the algorithm used by the feature extraction unit 307 to extract the feature point is the same as the algorithm used to register the sample in the database 111, and the algorithm known in the art regarding fingerprint recognition, face recognition or iris recognition. Can be used.

In this case, the feature point extractor 307 does not extract the feature points for all the sample images provided by the terminal connection unit 303, but instead selects the feature points for the sample images whose quality measured by the quality measuring unit 305 is equal to or greater than a predetermined reference value. Extraction can increase the accuracy of 'counterfeit defense evaluation'.

The feature point extractor 307 provides the feature defense data extracted from the sample image to the counterfeit defense force analysis unit 309.

The anti-counterfeiting force analysis unit 309 determines whether the sample image provided by the terminal connection unit 303 is obtained from a living body or a forgery organism using the test sample information registered in the database 111. Conduct a 'counterfeit defense assessment'. Counterfeit defense force analysis unit 309 provides the analysis results to the reporting unit 311.

For the 'counterfeit defense force evaluation', the counterfeit defense force analysis unit 309 compares the feature points extracted by the feature point extraction unit 307 with the feature points of the samples registered in the database 111, the sample registered in the database 111 Search for the same sample. If, as a result of querying the database 111, if the feature point extracted by the feature point extraction unit 307 is determined to be a forgery living body, the mobile terminal 130 erroneously judges the test forgery sample as a living body to the analysis system 110 It is provided.

Since the counterfeit defense analysis is usually performed in a series of continuous processes using a plurality of preset test samples (biological and counterfeit living) as a set (Set), the examiner sequentially performs a set of test samples for the mobile terminal 130 Input to the biometric sensor 201, the anti-counterfeiting force analysis unit 309 comprehensively analyzes the anti-counterfeiting force against a set of test samples. If the mobile terminal 130 performs accurate and complete forgery determination, the sample image for the counterfeit sample should not be classified and provided to the analysis system 110, and all sample images for the biological sample will be provided to the analysis system 110. Should be. Therefore, the analysis items of the anti-counterfeiting force analysis unit 309 includes the following.

(1) Counterfeit detection rate or counterfeit tolerance rate determined by the number of fake organisms in the sample image provided by the mobile terminal 130.

(2) As the mobile terminal 130 determines that the biological sample included in the set of test samples is forged, the mobile terminal 130 comprehensively determines whether the biological sample is not provided with the sample image.

The reporting unit 311 stores the quality measurement result of the quality measuring unit 305 and the analysis result of the counterfeit defense force analysis unit 309 together with the sample image in the database 111 and outputs them to the inspector.

Hereinafter, with reference to Figure 4 describes a mobile-based recognition performance and counterfeit defense analysis method of the present invention.

<Enter inspection mode: S401>

For the test, when the test target mobile terminal 130 and the analysis system 110 are communicatively connected, and the terminal connection unit 303 is connected to the system connection unit 205 of the mobile terminal 130 according to a predetermined protocol, the system connection unit 205 enters a test mode.

<Input of test sample: S403>

During the test mode, the inspector sequentially inputs a set of test samples to the biometric sensor 201 of the mobile terminal 130 under the control of the analysis system 110. As described above, a set of test samples includes biological samples and counterfeit samples.

According to an embodiment, the terminal connection unit 303 may confirm that the system connection unit 205 has entered the test mode, and then guide the examiner to input the test sample through the display unit or the speaker.

<Receive sample image: S405>

In accordance with the input of the test sample, the biometric sensor 201 of the mobile terminal 130 obtains a sample image from the test sample, the recognition control unit 203 essentially determines whether the sample is a living or counterfeit organisms. . If the recognition control unit 203 determines that the input test sample is not a fake sample, the system connection unit 205 provides a recognition success message to the terminal connection unit 303 of the analysis system 110 together with the sample image. do. Therefore, only the sample image of the sample determined to be a living body among the set of test samples is provided to the analysis system 110.

The terminal connection unit 303 provides the sample image received from the mobile terminal 130 to the quality measuring unit 305 and the feature point extraction unit 307.

<Measurement of Sample Image Quality: S407>

The quality measuring unit 305 calculates a quality score or grade by performing quality inspection on the sample image provided from the terminal connection unit 303.

<Feature Point Extraction of Sample Image: S409>

The feature point extractor 307 extracts the feature point from the sample image provided from the terminal connection unit 303.

According to the embodiment, the feature point may not be extracted for the sample image having the quality score calculated in step S407 below the reference value, but such processing is not essential. On the other hand, step S409 may be performed in parallel with step S407.

<Forgery Determination: S411>

The counterfeit defense force analysis unit 309 searches for the same sample among the samples registered in the database 111 by comparing the feature points extracted by the feature point extractor 307 with the feature points of the samples registered in the database 111. If, as a result of querying the database 111, if the feature point extracted by the feature point extraction unit 307 is determined to be a forgery living body, the mobile terminal 130 erroneously judges the test forgery sample as a living body to the analysis system 110 It is provided.

Counterfeit defense analysis and reporting: S413 to S417

Steps S403 to S411 are sequentially performed on a set of test samples (S413).

When the input for a set of test samples is completed, the anti-counterfeiting force analysis unit 309 performs the anti-counterfeiting force analysis of the mobile terminal 130. If the mobile terminal 130 performs accurate and complete forgery determination, the sample image for the counterfeit sample should not be classified and provided to the analysis system 110, and all sample images for the biological sample will be provided to the analysis system 110. Should be (S415).

The reporting unit 311 ends the analysis by finally outputting the anti-counterfeiting force performance of the mobile terminal 130 unit analyzed by the anti-counterfeiting force analysis unit 309 to the examiner. When the analysis ends, the terminal connection unit 303 ends the test mode, and the connection between the system connection unit 205 of the mobile terminal 130 is terminated (S417).

In the above method, the analysis method of the analysis system 110 of the present invention is performed.

While the above has been shown and described with respect to preferred embodiments of the invention, the invention is not limited to the specific embodiments described above, it is usually in the technical field to which the invention belongs without departing from the spirit of the invention claimed in the claims. Various modifications can be made by those skilled in the art, and these modifications should not be individually understood from the technical spirit or prospect of the present invention.

Claims (8)

In the biometric performance and counterfeit defense analysis system of a mobile device,
A terminal connection unit connected to a mobile terminal to be inspected and receiving a sample image from the mobile terminal (the mobile terminal acquires a sample image from a test sample to perform bio certification and determines that the test sample is a living body). The sample image is provided to the terminal connection unit);
A feature point extraction unit for extracting feature point data from the sample image received by the terminal connection unit;
And a counterfeit defense force analysis unit for analyzing the counterfeit defense force of the mobile terminal by checking a feature point data extracted by the feature point extractor to a database to determine whether the sample image is an image of a living body or an image of a counterfeit organism. Recognition performance and counterfeit defense analysis system.
The method of claim 1,
The mobile terminal,
A biometric sensor for obtaining a sample image from the test sample by fingerprint recognition, face recognition, or iris recognition method;
A recognition controller configured to perform bio recognition based on the sample image acquired by the bio recognition sensor and determine whether or not to forgery; And
In case of being connected with the terminal connection unit, the process proceeds to the test mode, and when the recognition control unit determines that the test sample is not a forgery in the test mode, the sample image is provided to the terminal connection unit, and the recognition control unit provides the test. And a system connection unit which does not provide the sample image to the terminal connection unit when the for-use sample is determined to be a forgery.
The method of claim 1,
The database registers and manages sample images for the test sample. The registered sample image biometric performance and counterfeit defense analysis system of a mobile device, characterized in that the sample image obtained from the living body and the sample image obtained from the fake organisms.
The method of claim 1,
A quality measuring unit evaluating the quality of the sample image received by the terminal connection unit; And
And a reporting unit for finally outputting a quality score of the quality measuring unit and a counterfeit defense force score of the counterfeit defense force analysis unit.
In the analysis method of the analysis system for the bio-recognition performance and counterfeit defense of the mobile device,
Connecting to the mobile terminal to be inspected through the terminal connection unit;
Receiving a sample image from the mobile terminal; (The mobile terminal acquires a sample image from the test sample and performs bio certification, and when the test sample is determined to be a living body, provides the sample image to the terminal connection unit.)
Extracting feature point data from a sample image received from the mobile terminal; And
And querying the extracted feature point data in a database to determine whether the sample image is an image of a living body or an image of a counterfeit organism.
The method of claim 5,
The mobile terminal,
When the system connection unit is connected to the terminal connection unit, proceeding to a test mode, and then obtaining a sample image from the test sample by a biometric sensor by fingerprint recognition, face recognition, or iris recognition method;
Performing bio-recognition based on a sample image acquired by the biometric sensor and determining whether or not to counterfeit; And
If it is determined that the test sample is not a forgery in the test mode, the system connection unit provides the sample image to the terminal connection unit. Analysis method of the analysis system, characterized in that it comprises a step not provided.
The method of claim 5,
The database registers and manages sample images for the test sample. The registered sample image analysis method of the analysis system, characterized in that the sample image obtained from the living body and the sample image obtained from the fake organisms.
The method of claim 5,
Evaluating the quality of the sample image received by the terminal connection unit; And
And finally outputting a quality score according to the quality evaluation and a counterfeit defense strength score according to the counterfeit defense force analysis.

Images