Method for verifying a user and relative electronic authentication system

Abstract

A method for verifying a user (3) in an electronic authentication system (1) is described, comprising the steps of detecting one or more environmental data, comparing the detected environmental data with expected environmental data, selecting a security level as a function of the comparison, performing, as a function of the selected security level, an authentication operation (5) of the user (3) in order to obtain authentication data (5a), extracting, from the authentication data (5a), a plurality of characteristics (7a), comparing the characteristics (7a) with a predetermined number of environmental verification parameters (8), analysing, during the comparison of the characteristics, the environmental data, the authentication data (5a) and the characteristics (7a) in order to identify an anomaly among the environmental data, the authentication data (5a) and the characteristics (7a) and authenticating the user (3) if from the comparison of the characteristics (7a) they meet said environmental verification parameters (8) and if from said analysis no anomalies have emerged among the environmental data, the authentication data (5a) and the characteristics (7a).

Description

[0001] METHOD FOR VERIFYING A USER AND RELATIVE ELECTRONIC AUTHENTICATION SYSTEM

[0002] DESCRIPTION

[0003] TECHNICAL FIELD OF THE INVENTION

[0004] The present invention relates to a method for verifying a user, an electronic authentication system and a computer-readable storage medium.

[0005] PRIOR ART

[0006] Nowadays, with the rise of increasingly sophisticated artificial intelligence and increasingly aggressive cyber attacks, the need to verify that a user is actually a real person and not an automated system is increasingly evident at the level of cybersecurity. This issue is particularly important for voice authentication systems, as an attacker could record the user's voice and reproduce it to authenticate him- / herself in the place of the user him- / herself.

[0007] An authentication system contains the modules necessary to recognize a user such as Speaker Identification (SID), which is a module that takes care of authorizing the user through voice, and Face Recognition (FR), which is a module that takes care of authorizing the user through the image of his / her face. The authorization modules are not intended to withstand any attacks, as adding this constraint in the training process would reduce their efficiency.

[0008] Possible attacks to which authentication systems may be subjected include:

[0009] - Recording Attack, in which the attacker records the user's voice and reproduces it;

[0010] - Speech Synthesis Attack, in which the attacker uses a speech synthesizer to reproduce the user's voice;

[0011] - MultiModal Synthesis Attack, in which the attacker uses a speech synthesizer to reproduce the user's voice and an image generator to reproduce the user's face;

[0012] - Expert Attack, in which the attacker can interact with the screen of the device to read textual information, generate an audio of the user pronouncing the textual information, generate a video of the user pronouncing the textual information (e.g. DeepFake) and synchronize the audio and video generated;

[0013] - Injection Attack, in which the attacker is able to inject a video / audio directly into the device he / she is using, bypassing the acquisition hardware.

[0014] SUMMARY OF THE INVENTION The technical task of the present invention is therefore to provide a method for verifying a user, an authentication system and a computer-readable storage medium that are able to overcome the drawbacks arising from the prior art.

[0015] The object of the present invention is therefore to provide a method for verifying a user, an authentication system and a computer-readable storage medium that allow effective verification and recognition of an attempt of entry by bots or hackers.

[0016] A further object of the present invention is therefore to provide a method for verifying a user, an authentication system and a computer-readable storage medium that are able to be easily updated so as to be able to recognize new types of attack.

[0017] The defined technical task and the specified objects are substantially achieved by a method for verifying a user, an authentication system and a computer-readable storage medium comprising the technical characteristics set forth in one or more of the appended claims. The dependent claims correspond to possible embodiments of the invention.

[0018] In particular, the specified technical task and the specified objects are substantially achieved by a method for verifying a user in an authentication system.

[0019] The method comprises a step of detecting one or more environmental data associated with the user, with a space in which the user is located and / or with a user device in which the authentication system is installed or installable; it is followed by a step that envisages comparing the detected environmental data with expected environmental data.

[0020] In other words, the system creates an estimate of the environmental and hardware conditions in which the user is located.

[0021] Preferably, the step of detecting and comparing is carried out in such a way as to verify that the detected conditions are consistent with those expected.

[0022] The method further comprises a step of selecting a security level as a function of the comparison. In other words, the method provides for loading a security level relative to the detected environmental and hardware conditions that is compatible with a security level requested by a user or by a service provider.

[0023] Preferably, in the event that it is not possible to select a compatible security level, the verification method further provides for the step of detecting one or more environmental data so as to improve the detected environmental conditions.

[0024] Preferably, if the environmental conditions are known a priori, it is possible to define a more accurate and stringent security level. The method also provides for carrying out, as a function of the selected security level, an authentication operation of the user in order to obtain authentication data associated with the user, with a space in which the user is located and / or with a user device in which the authentication system is installed or installable.

[0025] The method also provides for extracting, from the authentication data, a plurality of characteristics.

[0026] Preferably, the step of performing the authentication operation is carried out by proposing to the user one or more actions to be performed in order to obtain specific authentication data and / or specific characteristics.

[0027] The characteristics are correlated to actions performed by the user, to details of the space in which the user is located and how the authentication data are obtained from the authentication system.

[0028] The method also provides for comparing the characteristics with a predetermined number of environmental verification parameters. The predetermined number of environmental verification parameters is selected as a function of the security level. In other words, the method provides for loading a number of queries necessary to respond to the required security level.

[0029] The method further provides for analysing, during the comparison of the characteristics, the environmental data, the authentication data and the characteristics in order to identify an anomaly among the environmental data, the authentication data and the characteristics. In other words, the method defines “liveness modules” to be performed during the performance of the authentication operation (or challenge) or during the comparison of the characteristics to ensure that the data obtained are authentic and not generated specifically to respond to the environmental verification parameters.

[0030] Preferably, the step of extracting the characteristics is carried out by combining two or more characteristics together in order to obtain complex characteristics with which to perform the step of comparison with the environmental verification parameters.

[0031] Preferably, the combination of the characteristics also allows to reduce the number of environmental verification parameters to be compared and / or the number of actions to be carried out.

[0032] Preferably, the comparison step is carried out by assigning a weight to the environmental verification parameters and by selecting specific environmental verification parameters as a function of an overall weight required by the selected security level. Preferably, the step of assigning the weight of the verification parameters allows to define a minimum threshold level for each environmental verification parameter (or query) in order to determine a minimum security level to be guaranteed.

[0033] Finally, the method provides for authenticating the user if from the comparison the characteristics meet the environmental verification parameters and if no anomalies have emerged among the environmental data, the authentication data and the characteristics from the analysis.

[0034] The specified technical task and the specified objects are further achieved by an authentication system comprising means for obtaining environmental data and authentication data and a processing unit configured to perform one or more of the above method steps.

[0035] In general, it should be noted that in the present context the processing unit is presented as subdivided into distinct functional modules (storage modules or operating modules) for the sole purpose of describing the functions thereof in a clear and complete manner.

[0036] In reality, this processing unit can consist of a single electronic device, suitably programmed to perform the functions described, and the different modules can correspond to hardware entities and / or routine software forming part of the programmed device.

[0037] Alternatively, or in addition, said functions can be performed by a plurality of electronic devices over which the aforesaid functional modules can be distributed.

[0038] The processing unit can further rely on one or more processors to execute the instructions contained in the memory modules.

[0039] The specified technical task and the specified objects are further achieved by a computer-readable storage medium comprising instructions which, when executed by a computer, induce the computer to perform the above method for verifying a user.

[0040] Preferably, a computer program (stored in the storage medium) comprises instructions so that it is launched on the system including one or more of the characteristics described in the present description.

[0041] In one embodiment the computer program is stored in a memory of the user interface device.

[0042] In one embodiment, the computer program is stored remotely. In this embodiment, it is a “SAAS (Software As A Service)” computer program, that is, configured to be implemented only in the presence of an internet connection. BRIEF DESCRIPTION OF THE DRAWINGS

[0043] Further characteristics and advantages of the present invention will become clearer from the indicative, and therefore non-limiting, description of an embodiment of a verification method, an authentication system and a computer-readable storage medium.

[0044] Such a description will be set out below with reference to the accompanying drawings, which are provided solely for illustrative and therefore non-limiting purposes, in which: figure 1 is a schematic representation of the method subject-matter of the present invention.

[0045] DETAILED DESCRIPTION OF THE INVENTION

[0046] The present invention relates to a method for verifying a user in an authentication system 1.

[0047] The present invention also relates to an authentication system 1 capable of carrying out the aforementioned method.

[0048] The authentication system 1 comprises environmental data and authentication data acquisition means 2a and comprises a processing unit 2b (e.g. a microprocessor) configured to perform one or more of the above method steps.

[0049] In general, it should be noted that in the present context and in the subsequent claims, the processing unit 2b is presented as subdivided into distinct functional modules (storage modules or operating modules) for the sole purpose of describing the functions thereof in a clear and complete manner.

[0050] In reality, the processing unit 2b can consist of a single electronic device, suitably programmed to perform the functions described, and the various modules can correspond to hardware entities and / or routine software belonging to the programmed device.

[0051] Alternatively, or in addition, said functions can be performed by a plurality of electronic devices over which the aforesaid functional modules can be distributed.

[0052] The processing unit 2b can further rely on one or more processors to execute the instructions contained in the memory modules.

[0053] By authentication system 1 is therefore meant the set of hardware and software components that make up a system regulating the accesses by means of a user device 2 by a user 3.

[0054] By user device 2 can be intended to mean smartphones, personal computers, tablets, smart TVs and other similar devices provided with audio, video and / or sensor devices of different types (such as, for example and not limited to, accelerometers, gyroscopes and the like).

[0055] The method provides for detecting one or more environmental data associated with the user 3, with a space 4 in which the user 3 is located and / or with the user device 2 in which the authentication system 1 is installed or installable.

[0056] The method also provides for comparing the detected environmental data with expected environmental data. In other words, the method (i.e. the authentication system 1) creates an estimate of the environmental and hardware conditions in which the user 3 is located. In particular cases, the environmental conditions are known a priori (for example the user device 2 is a fixed terminal within a bank or a station) and it is therefore possible to verify that the detected conditions are consistent with the expected environmental conditions.

[0057] The method also provides for selecting a security level as a function of the comparison. In other words, as a function of the environmental conditions, the method determines a set of rules determining a minimum security level that the authentication system 1 can guarantee, as will become clearer in the continuation of the present description. In other words, the method provides for identifying essential characteristics that must be met by the environmental data in order to verify the user 3.

[0058] The method further provides for performing, as a function of the selected security level, an authentication operation 5 of the user 3.

[0059] The authentication operation 5 allows obtaining authentication data 5a associated with the user 3, with the space 4 in which the user 3 is located and with the user device 2 in which the authentication system 1 is installed or installable.

[0060] Preferably, the step of detecting the environmental data and the step of performing the authentication operation 5 are carried out by means of audio 6a and / or video 6b acquisition and / or geolocation data and / or movement data.

[0061] In other words, the user device 2 is provided with environmental data acquisition means 2a. Such acquisition means 2a may comprise a camera (for obtaining a video clip and / or an image), a microphone (for obtaining an audio) or motion sensors (for obtaining geographical or motor positions of the user 3).

[0062] Preferably, the step of performing the authentication operation 5 is carried out by proposing to the user one or more actions 5b to be performed in order to obtain specific authentication data.

[0063] Therefore, the authentication operation 5 is a challenge that the method (i.e. the authentication system 1 ) proposes to the user 3 with the purpose of collecting the aforementioned authentication data in input. For example, the user 3 may be asked to slightly rotate the user device 2 so as to allow the acquisition of geometric and gyroscopic data relative to the space 4.

[0064] The method further provides for extracting, from the authentication data, a plurality of characteristics 7a. The characteristics 7a are correlated to actions performed by the user 3, to details of the space 4 in which the user 3 is located and / or how the authentication data are obtained from the authentication system 1 .

[0065] Preferably, the step of extracting the characteristics 7a is carried out by extracting one or more details from an audio, a video clip, an image and / or from position / movement data of the user 3 or of the user device 2.

[0066] Preferably, the step of performing the authentication operation 5 is carried out by proposing to the user one or more actions 5b to be performed in order to obtain specific characteristics 7a.

[0067] The characteristics 7a (or features) are extracted using Video Analysis, Image Analysis and Audio Analysis components. These components can exploit a combination of Machine Learning and Computer Vision, Image Processing and Audio Processing techniques. In particular, the modules may interact with each other to reduce processing time and provide more detailed characteristics 7a. For example, the Video Analysis module may provide an estimate of the environmental geometry by exploiting the Image Analysis modules for each frame. The Audio Analysis module can in turn extract information on the audio background starting from the provided estimate of the environmental geometry.

[0068] Preferably, the step of extracting the characteristics 7a is carried out by combining two or more characteristics 7a together in order to obtain complex characteristics 7b.

[0069] In other words, the extracted characteristics 7a can be combined together in order to obtain higher level characteristics. For example, a segmentation map produced by the Video / lmage Analysis modules may be combined with the image textures and the estimate of the pose of the user 3 in order to obtain a complex characteristic 7b describing the geometry and the appearance of a part of the body of the user 3 (e.g. the face, a hand, an eye, the mouth etc.).

[0070] By the term “segmentation map” is meant an image obtained by means of an image segmentation process from a starting image, i.e. an image composed by a set of segments, level curves or aggregates of pixels adapted to fully describe each area of the original image.

[0071] The objective of a segmentation map is therefore to classify each pixel of an image according to several criteria, including:

[0072] - the semantic segmentation defining the class to which the represented object belongs --e.g. the pixels belonging to different people in a photograph all have the same value, but different from the background one and other objects;

[0073] - the segmentation of instances defining, for each pixel, the object to which this pixel belongs --e.g. the pixels belonging to different people in a photograph all have different values;

[0074] - the “panoptic” segmentation combining the semantic and instance segmentation.

[0075] The method also provides for the step of comparing the characteristics 7a with a predetermined number of environmental verification parameters 8. The predetermined number of environmental verification parameters 8 is selected as a function of the security level and determine the essential characteristics that must be met in order to verify the user.

[0076] The environmental verification parameters 8 (or queries) are a well-defined request about a characteristic 7a to be verified in one or more input data and provides a probabilistic response determined by a mathematical model.

[0077] For example, an environmental verification parameter 8 may be to evaluate whether the input image is static or not, whether the space 4 is closed, or whether the user 3 is located in a context consistent with the time and place where the user 3 is located, or whether the user 3 has spoken by moving his / her lips.

[0078] Other environmental verification parameters 8 can be relative to whether the user 3 has moved his / her lips, whether the geometries in the background are consistent, whether the user 3 has blinked his / her eyes, whether the user 3 has moved his / her head, whether the image during the acquisition is too still and other similar parameters.

[0079] The introduction of additional environmental verification parameters 8 offers the possibility of increasing the security level of the authentication system 1 against technologically more advanced attacks. Therefore, as a function of the detected environmental data and of the comparison, a suitable security level is selected that could require, as a function of the context, more complex environmental verification parameters 8 and therefore the need to obtain the complex characteristics 7b.

[0080] For example, if the authentication system 1 is installed and functioning in a portable electronic device (e.g., a smartphone), the user 3 has the physical ability to hold the device, and the user 3 is not located in front of a flat background, the security level will require checking whether the user 3 is located in an enclosed space and whether the context is consistent with the time and place where the user 3 is located.

[0081] Preferably, the comparison step is carried out by assigning a weight to said environmental verification parameters 8 and by selecting specific environmental verification parameters 8 as a function of an overall weight required by the selected security level. In other words, it is possible to define a minimum threshold level for each environmental verification parameter 8 in order to determine the minimum security level that must be guaranteed.

[0082] Therefore the responses (deriving from the comparison) to the environmental verification parameters 8 are provided in the form of calculated probability given the input characteristics 7a (or complex characteristics 7b). The method therefore provides for assigning a minimum response weight or threshold to each environmental verification parameter 8 on the basis of the security level. The minimum threshold is indispensable, for example, for the environmental verification parameters 8 relative to the presence of a face. The responses of the characteristics 7a are combined according to a mathematical model and a final threshold determining a final decision regarding the authentication of the user 3.

[0083] The method further provides for the step of analysing, during the comparison of the characteristics 7a (i.e. the complex characteristics 7b), the environmental data, the authentication data and the characteristics 7b in order to identify an anomaly among the environmental data, the authentication data and the characteristics 7a. In other words, the method provides for verifying that the data in input to the authentication system 1 are real by means of liveness modules.

[0084] Preferably, the step of analysing and identifying an anomaly is carried out by performing a further authentication operation 5 in order to obtain further authentication data, i.e. further characteristics 7a (or 7b) to be compared with the previously obtained authentication data, i.e. with the previously obtained characteristics 7a (or 7b). In other words, this module or Content Verification provides for requesting the user 3 to repropose a specific content proposed by the authentication system 1 during the authentication step. This category of modules is designed to protect the system from “recording” type attacks, since it requires re-proposing either dynamic content or solving a simple challenge requiring an interaction on the part of the user 3 that is always different. Preferably, the step of analysing and identifying an anomaly is carried out by verifying a synchronization between the acquired audio and the acquired video. For example, the method provides for verifying the synchrony between the lips of the user 3 with the words spoken, or the position of the device 2 in the surrounding space 4 obtained by crossing the estimate of the environmental geometry obtained from the video clip with position data obtained by means of the user device 2 coming, for example, from the gyroscope of the device 2.

[0085] Preferably, the step of analysing and identifying an anomaly is carried out by verifying that the environmental data and the authentication data are not generated by means of generative Artificial Intelligence models. In other words, the method verifies that the acquired video clip is effectively captured by a camera and is not a video generated by the aforementioned generative models. The modules therefore verify the realism of the image in input.

[0086] Preferably, the step of analysing and identifying an anomaly is carried out by comparing the frames of the video clip with each other, analysing a movement pattern of the user 3 and / or the textures of the image.

[0087] Preferably, the step of analysing and identifying an anomaly is carried out by verifying that the audio is consistent with the video clip by analysing background noise and by obtaining data from the user device 2.

[0088] Preferably, the step of analysing and identifying an anomaly is carried out by verifying that the environmental data and / or the authentication data have not been injected into the user device 2.

[0089] The method finally provides for the step of authenticating the user 3 if from the comparison the characteristics 7a meet the environmental verification parameters 8 and if no anomalies have emerged among the environmental data, the authentication data and the characteristics 7a (or 7b) from the analysis. In other words, the method returns a result 9 relative to a successful authentication or of a rejected authentication.

[0090] In the event of successful authentication, it means that a predetermined number of environmental verification parameters 8 have been confirmed and that no anomalies have emerged from the analysis.

[0091] In case of reject, it is possible that environmental parameters have not been confirmed and / or that anomalies considered critical and significant of an attack on the authentication system 1 have emerged from the analysis.

[0092] The present invention further relates to a computer-readable storage medium comprising instructions which, when executed by an electronic computer, induce the computer to perform the above method for verifying a user.

[0093] The present invention further relates to a computer program comprising instructions for executing the above method for verifying a user when the program is executed by a computer or processing unit 2b.

[0094] Advantageously, the present invention is capable of overcoming the drawbacks which have emerged from the prior art.

[0095] Advantageously, the present invention has high robustness and security.

[0096] Advantageously, the present invention provides a modular system that can be configured according to the needs of the user, according to the context of use and of the available computational power.

[0097] Advantageously, the environmental verification parameters 8 and the modules for identifying an anomaly can be updated as the technology progresses.

[0098] Advantageously, the method is able to keep track of the failed environmental verification parameters 8 in order to verify which modules for identifying an anomaly are most discriminating for a certain type of attack.

Claims

CLAIMS1. Method for verifying a user (3) in an electronic authentication system (1 ), the method comprising the steps of:- detecting one or more environmental data associated with said user (3), with a space(4) in which said user (3) is located and / or with a user device (2) in which said electronic authentication system (1 ) is installed or installable;- comparing said detected environmental data with expected environmental data;- selecting a security level as a function of said comparison;- performing, as a function of said selected security level, an authentication operation(5) of the user (3) in order to obtain authentication data (5a) associated with said user (3), with a space (4) in which said user (3) is located and / or with a user device (2) in which said authentication system (1 ) is installed or installable;- extracting, from said authentication data (5a), a plurality of characteristics (7a), said characteristics (7a) being correlated to actions performed by the user (3), to details of the space (4) in which the user (3) is located and / or how said authentication data (5a) are obtained from the authentication system (1 );- comparing said characteristics (7a) with a predetermined number of environmental verification parameters (8), said predetermined number of environmental verification parameters (8) being selected as a function of said security level;- analysing, during said comparison of the characteristics, the environmental data, the authentication data (5a) and said characteristics (7a) in order to identify an anomaly among said environmental data, said authentication data (5a) and said characteristics (7a);- authenticating said user (3) if from said comparison said characteristics (7a) meet said environmental verification parameters (8) and if from said analysis no anomalies have emerged among said environmental data, said authentication data (5a) and said characteristics (7a).

2. Verification method according to claim 1 , wherein the step of detecting the environmental data and the step of performing said authentication operation (5) are carried out by means of audio (6a) and / or video (6b) acquisition, and wherein said step of extracting the characteristics (7a) is performed by extracting one or more details from an audio, a video clip and / or an image.

3. Verification method according to claim 1 or 2, wherein said step of comparing the environmental data is carried out by assigning a weight to said environmentalverification parameters (8) and by selecting specific environmental verification parameters (8) as a function of an overall weight required by the selected security level.

4. Verification method according to one or more of the preceding claims, wherein said step of extracting the characteristics (7a) is carried out by combining two or more characteristics (7a) together in order to obtain complex characteristics (7b) with which to perform said step of comparison with the environmental verification parameters (8).

5. Verification method according to one or more of the preceding claims, wherein said step of performing the authentication operation (5) is carried out by proposing to the user one or more actions (5b) to be performed for obtaining specific authentication data (5a) and / or specific characteristics (7a).

6. Method according to one or more of the preceding claims, wherein said step of analysing and identifying an anomaly is carried out by performing a further authentication operation (5) in order to obtain further authentication data (5a), i.e. further characteristics (7a) to be compared with the previously obtained authentication data (5a) or with the previously obtained characteristics (7a).

7. Method according to claim 2, wherein said step of analysing and identifying an anomaly is carried out by means of at least one of the following:- verifying a synchrony between the acquired audio and the acquired video;- estimating an environmental geometry obtained from the video clip and from position data obtained by means of the user device (2);- verifying that said environmental data and said authentication data (5a) are not generated by means of generative models.

8. Method according to claim 7 when dependent on 2, wherein said step of analysing and identifying an anomaly is carried out by comparing the frames of said video clip with each other, analysing a motion pattern of the user (3) and / or the textures of the image.

9. Method according to claim 2, wherein said step of analysing and identifying an anomaly is carried out by verifying that the audio is consistent with the video clip by analysing background noises and by obtaining position data from the user device (2).

10. Method according to one or more of the preceding claims, wherein said step of analysing and identifying an anomaly is carried out by verifying that the environmental data and / or the authentication data (5a) have not been injected into the user device (2).11 . Authentication system (1 ) for a user, the system comprising:- acquisition means (2a) configured to detect environmental data associated with the user (3), with a space (4) in which the user (3) is located and / or with a user device (2) inwhich the electronic authentication system (1 ) is installed or installable;- a processing unit (2b) configured to:■ compare said detected environmental data with expected environmental data;■ select a security level as a function said comparison;■ perform, as a function of said selected security level, an authentication operation (5) of the user (3) in order to obtain authentication data (5a) associated with said user (3), with a space (4) in which said user (3) is located and / or with a user device (2) in which said authentication system (1 ) is installed or installable;■ extract, from said authentication data (5a), a plurality of characteristics (7a), said characteristics (7a) being correlated to actions performed by the user (3), to details of the space (4) in which the user (3) is located and / or how said authentication data (5a) are obtained from the authentication system (1 );■ compare said characteristics (7a) with a predetermined number of environmental verification parameters (8), said predetermined number of environmental verification parameters (8) being selected as a function of said safety level;■ analyse, during said comparison of the characteristics, the environmental data, the authentication data (5a) and said characteristics (7a) in order to identify an anomaly between said environmental data, said authentication data (5a) and said characteristics (7a);■ authenticate said user (3) if from said comparison said characteristics (7a) meet said environmental verification parameters (8) and if from said analysis no anomalies have emerged between said environmental data, said authentication data (5a) and said characteristics (7a).

12. Computer-readable storage medium comprising instructions which, when executed by a computer, induce the computer to perform the method according to one or more of claims 1 to 10.

Images