EP2300955A1

EP2300955A1 - Method and apparatus for improving biometric identification systems

Info

Publication number: EP2300955A1
Application number: EP09780040A
Authority: EP
Inventors: Thomas Wölfl
Original assignee: Psylock GmbH
Priority date: 2008-07-08
Filing date: 2009-06-30
Publication date: 2011-03-30
Also published as: DE102008040258A1; EA201071382A1; WO2010003849A1; JP2011527475A; CN102089766A

Abstract

The present invention relates to a method and an apparatus for identifying a person, in which biometric data are acquired for the person to be identified, wherein the person to be identified must carry out a recognition action, on the basis of which the biometric data used for identification are determined, wherein at least one step from a group of checking steps is carried out, said step comprising the random selection of the recognition action and the additional solution to a problem, and wherein the problem is a problem which is suited to human capabilities and cannot be solved by automated systems or can only be solved by automated systems with a time requirement which exceeds a particular time requirement needed by an average person.

Description

METHOD AND DEVICE FOR IMPROVING BIOMETRIC IDENTIFICATION SYSTEMS

BACKGROUND OF THE INVENTION

FIELD OF THE INVENTION

The present invention relates to a method and a device for identifying a person, in which biometric data of the person to be identified are detected, wherein the person to be identified must carry out a recognition action on the basis of which the biometric data used for identification are determined.

STATE OF THE ART

Biometric methods are increasingly used to identify people, for example when accessing security areas or when logging into a data processing system such as a personal computer PC. The reason for this is that the widespread password as a means of identification and access control has a variety of security vulnerabilities and on the other hand is uncomfortable for the user, because he is supposed to change the password very often for security reasons, but on the other hand, problems arise with the memory , As a result, users either rely on consistent, insecure passwords or have difficulty identifying access.

Biometric identification systems have the advantage over here of being much safer because access to the biometric data, i. the ability to copy them is much more difficult than with a password, which is easy to use if there is insufficient secrecy. On the other hand, biometric identification systems are more comfortable for the user since they automatically carry the biometric properties with them without any effort, so that they do not have to remember a password.

However, there are certain security gaps even in the case of biometric identification systems, since, for example, it is conceivable that spyware programs or the like could record a data record of biometric access data, which would legitimate access could be deceptively used (so-called replay attack).

DISCLOSURE OF THE INVENTION

OBJECT OF THE INVENTION

It is therefore an object of the present invention to provide a method and a corresponding device for identifying a person based on a biometric method, which increases the security of identification or access control.

In addition, a corresponding method or a corresponding device should continue to be simple to set up and to use, so that the advantages of the biometric method with regard to a high level of comfort for the user are maintained.

TECHNICAL SOLUTION

This object is achieved by a method having the features of claim 1 and a device having the features of claim 12. Advantageous embodiments are the subject of the dependent claims.

The present invention is based on the recognition that the use of an already used record or signal set for unauthorized access or for a false identification of a person (replay attack) already effectively can be avoided if the biometric method at least two, preferably are based on multiple recognition actions, so that in particular by chance one of the two or more recognition actions for the appropriate access control can be selected. For example, a detection action for a biometric method may be to detect a human body part, such as a fingerprint or an eye. Since the human usually has two eyes, there is the possibility of using either the left or the right eye in the detection via the eye. When fingerprint recognition, the ten different fingers can use Find. Since human body parts are usually not 100% identical, it can also be seen which body part is actually used. If the identification system is now designed such that the recognition action is predetermined by the system and is selected from a plurality, that is to say at least two recognition actions, there is a certain probability that in an attempt to avoid identification with stored data, an incorrect recognition action will occur previously recorded without permission, is used so that the delusion attempt can be detected. In this way, it can be assumed with greater certainty that, in fact, a real, living person (life recognition) undergoes the biometric identification and does not use a stored data record which is only recorded for deception by a machine. Such a security function could only be overcome if all possible recognition actions were available as a record for the deception maneuver.

A corresponding safety function can also be used very well, for example, in the biometric method of tipper recognition, in which the recognition of the person to be identified on the operation of a keyboard or a keypad or the like is carried out by analyzing the typing behavior. In this case, the inventive system for identifying a person can specify a specific text for typing on a keyboard or a specific sequence of keys, so that the system can already see whether simple comparison of whether the correct keys or correct key sequence has been pressed Actual acts of recognition have been made for the current access, or whether attempts have been made to gain unauthorized access to previously recorded recognition acts or to pretend a false identity.

Additionally or alternatively, a corresponding biometric identification system can also be improved by the fact that the corresponding identification system provides an additional task for the solution by the person to be identified, the task being designed in such a way that it is tuned to human capabilities, ie that the task of a Can be solved much faster than by a machine or a computing device, or that it can actually only be solved by one person. Examples of such additional tasks may include recognizing structures in an image and rendering the recognized structure and / or answering a question about the structure to be recognized.

For example, in a display device, such as a monitor, the system may display an image comprising a structure of a sequence of numbers and / or letters, wherein the numbers and / or letters may be arranged in arbitrary orientations. With such an arrangement, an automatic system for recognizing the structure is difficult to recognize the corresponding numbers and letters, while a human being can easily recognize the number and / or sequence of letters.

The same applies, for example, to the pictorial representation of a Gegentands, so that as an additional task, the naming of this object can be required. The determination that a person solves the additional task and thus the proof that actually a person performs the identification and not a machine with stored data, is all the clearer if the image includes the structure against a background that makes the recognition of the structure difficult , d. H. more difficult to recognize for an automated recognition system. For example, this is the case when the background before which the structure is shown has a plurality of similarly shaped areas and / or lines with, for example, similar colors and the like as the structure. In such a case, a corresponding automatic recognition system is extremely difficult and takes a huge amount of time if it can even accomplish this task.

Accordingly, as an additional variant, it may be required that the solution of the additional task must at least be started or completed within a specific time requirement, the time requirement being oriented to an average person. If this time requirement is exceeded, the system determines this and aborts the identification with a negative result, because it can not be a human user, but the possibility of an attempt to make an unauthorized identification based on an automated detection.

Especially in the case of biometric recognition by means of the typing behavior, the additional security features can be integrated into the actual recognition by inputting a key operation. Thus, the selection of a specific text to be entered already be associated with the display of an image in which the text is easy to recognize only for a human, so that this additional task for the recognition of a structure is already integrated. At the same time, it is also possible to integrate the time component, ie the check whether the beginning until the user starts typing the string or the conclusion until the user has typed in the correct string, within a predetermined period of time, so that all the additional security features of the Present invention, namely specification of a specific recognition action of a plurality of recognition actions (input of a particular text or a typing sequence), solving an additional task (reading the text or typing sequence to be typed from a non-machine readable image) and time control in the biometric method by means of Tipperken-. tion can be realized in a simple manner.

BRIEF DESCRIPTION OF THE FIGURES

Further advantages, features and features of the present invention will become apparent in the following detailed description of exemplary embodiments with reference to the accompanying drawings, in which the drawings show in a purely schematic manner:

Figure 1 is a view of a data processing system with which the inventive method can be performed, and which is configured to a corresponding device according to the invention;

Figure 2 is an illustration of a structure in an image that may be used in the present invention; and in

Figure 3 is an illustration of another structure in an image for use with the present invention.

PREFERRED EMBODIMENTS

The present invention can be applied to many biometric identification systems in which biometric data for identifying a person is detected due to a plurality of different recognition actions. Accordingly, biometric methods can be used which are based on the recognition of a body part, for example the eyes or the fingerprints, or in which by means of an input into a direction by the person to be identified biometric data can be detected. This may be, for example, a voice recognition in which the user has to speak a certain or arbitrary sentence, so that the frequencies, frequency sequences and the like can be determined from the detected sounds and sounds so as to determine the identity of a user. Another possibility is to capture the typing behavior of the user when operating a keyboard. This can be done, for example, by the user entering any desired text or code or the like in a keyboard. A keyboard is here to be understood very broadly, so that it can be a common keyboard of a personal computer PC or a truncated numeric keypad for access control or a keypad for entering a key sequence in the manner of a Morse code. The keys of a corresponding keyboard can also be configured in a variety of ways. These may be conventional push-button computer keyboard keys, touch-sensitive keypads of a surface such as a screen, or the like.

FIG. 1 shows a normal personal computer PC 1 with a screen 2 as an output device and a conventional keyboard 3 as an input device. Such a personal computer system may be configured as an identification device according to the present invention, so that a method for identifying a person according to the present invention proceeds thereon.

In the exemplary embodiments described below, the biometric method is based on the tipper identification, as already described in various documents. By way of example, reference should be made to the publication WO 98/06020 A2, in which the present invention can be used. However, the present invention is not limited to biometric data acquisition by means of tipper identification, nor to the embodiment of a tipper identification identification described in the above-cited PCT application.

The identification of a user of the PC 1 can be done, for example, in such a way that the user is prompted when starting the system to enter a specific or arbitrary text via the keyboard 3.

In unspecified storage units of the personal computer, a data processing program for identifying the user is stored, which runs accordingly and controls the identification process. In particular, the data processing program forms a detection device which detects values and / or electrical signals, which are triggered by the actuation of the keyboard 3, from sensors and the keyboard 3 and evaluates them such that information about the typing behavior is generated by a user corresponding evaluation unit, which is also realized for the most part by the data processing program to be compared with stored biometric data of a person to be identified, so that the identity of the user can be determined with a corresponding match.

The corresponding software-equipped PC 1 further realizes a module for determining the authenticity of the biometric data according to the present invention, in order to rule out that fake biometric data are used which in a previous identification z. B. have been detected and stored by spyware or the like and are now linden in the stored form automatically for access to the PC system use.

The module for determining the authenticity of the biometric data generated randomly from a variety of different ways a specific text that is used for the determination of typing behavior. Unlike systems of the prior art, the user can thus not always enter the same text or freely choose what text he wants to enter, but he must at least partially use the specified random text by the random generator, which then for the Determination of typing behavior characteristics is used. In this way it is ruled out that a data record stored at any time can be used as an assumed keyboard input in an identification or access control since there is very little chance that the stored data with regard to the key operation will coincide randomly with the required inputs.

Additionally or alternatively, the module for determining the genuineness of the biometric data is provided with a task generator which generates an additional task, which would like to have access to the PC system 1 from the person to be identified or the user, preferably within one for a human usual time frame must be solved. This task is designed in such a way that a machine or an automated system can not solve such a task at all or only with a considerably higher expenditure of time. For these tasks, so-called CAPTCHA (Completely Automated Public Turing Test to keep Computers and Humans apart (full automatic public Turing test to distinguish computers and humans)) can be used. Such tests include, for example, the recognition of a letter and / or number sequence, wherein the letters and / or numbers are arranged in any orientation, in particular against a background that makes the automatic pattern recognition for a machine or a computer extremely difficult.

Likewise, structures such as simple objects can be displayed in a manner that also makes it difficult for an automatic recognition system to recognize the corresponding object. This can be done, for example, that the object consists of a plurality of same-colored or differently colored areas, the background is also composed of similarly shaped or similar colored areas, so that for a trained human eye, the object is very quickly recognizable while a automatic recognition system must perform a lengthy pattern matching, if any pattern recognition is possible.

Examples of this are shown in FIGS. 2 and 3, in which an image with a letter number sequence AB 12 is provided in FIG. 2 against a background with further lines and the like, wherein the letters and numbers are provided in different orientations. The letter-number sequence of the image 4 can be displayed on the monitor 2, the user then having to enter the corresponding letter or digit sequence in an input mask 5 via the keyboard 3.

FIG. 3 shows a further variant of such a task matched to human capabilities, in which a simple structure, such as the table 6, has to be recognized against a background with a large number of similar surfaces 7. If the image 4 of FIG. 3 is displayed on the monitor 2 in the access control in the PC system of FIG. 1, the question as to which object can be recognized in the image 4 can be asked as an additional task, whereupon the user or the person to be identified must enter the term "table" in the input field 5 via the keyboard. In this way, in addition to the underlying recognition of the typing behavior, it is initially excluded by randomly selected but system-dictated text that a recorded actuation of the keyboard 3 is used for access control. Subsequently, on the additional task in which, for example, the letter number sequence of Figure 2 or the object of Figure 3 must be recognized in the image 4, further determined whether a person actually sits in front of the PC 1 and the keyboard 3. Only one person can solve the additional task posed by the identification system to the user at all or in a given time, the time being oriented to the time requirement of an average user, which is far below what an automatic system, for example, is Picture recognition system, for the task asked would need.

On the one hand, the time component can be designed such that a correct input, that is to say the solution of the additional task, must take place within a specific time frame. Additionally or alternatively, this can also be designed so that at least with the input, z. B. the password, must be started.

Although the present invention has been described in detail with reference to the accompanying embodiments, it will be understood by those skilled in the art that the invention is not limited to these embodiments, but that the invention also includes modifications and changes, in particular the omission of individual features presented as well as the different combination of features presented, as long as the scope of the appended claims is not abandoned. In particular, the present invention includes the combination of all features presented.

Claims

claims

1. A method for identifying a person in which biometric data of the person to be identified is detected, wherein the person to be identified must perform a recognition act on the basis of which the biometric data used for identification are determined, characterized in that at least one step is performed from a set of verification steps involving the random selection of the recognition action and the additional solution of a task, the task being a human-adapted task which does not exist on automated systems, or only with a certain one of an average one People's required time required can be solved by exceeding the time required.

2. The method according to claim 1, characterized in that the solution of the task must be started or completed within a certain time required for the time interval.

3. The method according to any one of the preceding claims, characterized in that the recognition action is selected from the group comprising the typing on a keyboard, a vocal playback and the detection of a human body part.

4. The method according to any one of the preceding claims, characterized in that the biometric data include the typing behavior, the voice or speech recognition, the recognition of fingerprints and / or the detection of eyes.

5. The method according to any one of the preceding claims, characterized in that the additional task of recognizing structures in an image and the reproduction of the recognized structure and / or the answering of a question to be recognized

Structure includes.

6. The method according to claim 5, characterized in that the image as a structure comprises at least one number and / or letter sequence, wherein the numbers and / or letters are arranged in arbitrary orientations.

7. The method according to any one of claims 5 or 6, characterized in that the image as a structure comprises at least one pictorial representation of an object and the reproduction of the structure comprises the naming of the object.

8. The method according to any one of claims 5 to 7, characterized in that the image comprises the structure with a background that makes it difficult to recognize the structure.

A method according to claim 8, characterized in that the structure and the background comprise a plurality of similarly shaped surfaces and / or lines and / or similar colors.

10. The method according to any one of the preceding claims, characterized in that the detection action and the additional solution of a task carried out by a single action.

11. The method according to any one of the preceding claims, characterized in that the additional task is selected from a variety of different tasks.

12. A device for identifying a person with a detection device for acquiring biometric data of a person to be identified and an evaluation unit for comparing the acquired biometric data with stored data of a person to be identified and determining the identity of the person to be identified, characterized in that the device comprises a module for determining the genuineness of the biometric data, which comprises a random number generator which randomly determines which biometric data are acquired, and / or a task generator which generates an additional task matched to human capabilities which does not or only with automated systems a time requirement exceeding a certain amount of time can be solved.

13. The apparatus according to claim 12, characterized in that

Module for determining the authenticity of the biometric data comprises a memory for storing additional tasks and associated solutions.

14. The apparatus according to claim 12 or 13, characterized in that

Module for determining the authenticity of the biometric data comprises an input unit for inputting the solution and a comparison unit for determining the correctness of the solution.

15. Device according to one of claims 12 to 14, characterized in that

Module for determining the authenticity of the biometric data comprises a time recording for determining the time between the task and the beginning and / or completion of the solution input.

16. Device according to one of claims 12 to 15, characterized in that the device is prepared so that the method according to one of claims 1 to 11 is realized by the expiration of a data processing program.