Access authentication using obfuscated biometrics

Abstract

A method for authorizing access to one or more secure computer resources includes the step of obfuscating a reference biometric vector into an obfuscated reference biometric vector using similarity-preserving obfuscation. An authentication biometric vector is obfuscated into an authentication biometric vector obfuscated using similarity-preserving obfuscation. The similarity between the obfuscated authentication biometric vector and the obfuscated reference biometric vector is tested. Based on the similarity being within an authentication threshold, access to one or more secure computer resources is authorized.

Description

Background Technology

[0001] Biometrics is commonly used as an authentication method for accessing secure computer resources. Examples of biometric authentication methods include facial recognition, voice analysis, fingerprint recognition, and iris recognition. means of solving the problem

[0002] The means for solving the problem is intended to introduce, in a brief form, one of the concepts described in the following detailed description. The means for solving the problem does not represent a primary or essential function of the claimed subject matter, nor does it limit the scope of the claimed subject matter. Furthermore, the claimed subject matter is not limited to embodiments that resolve any or all disadvantages mentioned in any part of this disclosure.

[0003] A method for authorizing access to one or more secure computer resources includes the step of obfuscating a reference biometric vector into an obfuscated reference biometric vector using similarity-preserving obfuscation. An authentication biometric vector is obfuscated into an authentication biometric vector obfuscated using similarity-preserving obfuscation. The similarity between the obfuscated authentication biometric vector and the obfuscated reference biometric vector is tested. Based on the similarity being within an authentication threshold, access to one or more secure computer resources is authorized. Brief explanation of the drawing

[0004] Figure 1 schematically illustrates the authorization of access to secure computer resources using biometrics. FIGS. 2a through 2c illustrate exemplary methods for authorizing access to secure computer resources. FIG. 3 schematically illustrates converting a biometric identifier into a biometric vector and obfuscating the biometric vector into an obfuscated biometric vector. FIG. 4 schematically illustrates testing the similarity between an obfuscated reference biometric vector received from a client computing device and an obfuscated authentication biometric vector. Figure 5 schematically illustrates an exemplary computing system. Specific details for implementing the invention

[0005] As described above, biometric authentication is often used to authorize access to secure computer resources. An authentication biometric identifier (e.g., fingerprint or facial scan) detected by a client computing device is transmitted to an authentication computing device and compared with a previously measured reference biometric identifier—e.g., various fingerprints or facial scans provided by the user during account creation. For example, the authentication computing device may be a network-accessible server that restricts access to computer resources (e.g., local machine login, local machine file access, local machine peripheral access, or cloud-based service access) and allows access to the restricted computer resources only after the client computing device has successfully authenticated biometrically.

[0006] This scenario is schematically illustrated in FIG. 1, which shows an exemplary client computing device (100) equipped with a biometric sensor (102). The biometric sensor is configured to detect a biometric identifier from a human user (104). In this example, the biometric sensor is a camera configured to capture an image of the user's face. In other examples, other appropriate biometric identifiers of the human user may be measured, which are described in more detail below.

[0007] After the biometric device (102) captures an authentication biometric vector (in this case, an image of the user's face), the client computing device (100) transmits the authentication biometric identifier to the authentication computing device (108) over the network (106) (e.g., the Internet). Upon receiving the authentication biometric identifier, the authentication computing device compares the authentication biometric identifier with a stored reference biometric identifier known to correspond to an authorized user. If the authentication biometric identifier matches the reference biometric identifier within an authentication threshold, the authentication computing device communicates the authentication authority to the client computing device to authorize access to one or more secure computer resources.

[0008] However, the secure transmission and storage of biometric information are of the utmost importance. Recognizing that a malicious user may attempt to intercept transmitted biometric information or improperly access stored biometric information, the present disclosure describes a technique for authorizing access to one or more secure computer resources based on obfuscated biometric vectors. Specifically, a client computing device may measure an authenticating biometric identifier, convert the biometric identifier into a biometric vector, and obfuscate the biometric vector using similarity-preserving obfuscation before transmitting it to an authenticating computing device. After receiving the obfuscated authenticating biometric vector, the authenticating computing device may test for similarity between the obfuscated authenticating biometric vector and a previously received obfuscated reference biometric vector. Similarity-preserving obfuscation enables the authenticating computing device to authorize access to secure computer resources by testing for similarity in the obfuscated vector, while mitigating the risk that the obfuscated vector could be inverted to restore the original biometric identifier. In this way, biometric identifiers can be used to safely and securely authorize access to computer resources without compromising user privacy.

[0009] FIGS. 2a through 2c illustrate an exemplary method (200) for authorizing access to one or more security resources. The method (200) may be implemented through any suitable computing device having any suitable form factor and hardware configuration. As an example, but not a limitation, one or both of the client computing device and the authentication computing device may be a desktop, laptop, smartphone, server, tablet, video game console, media center, fitness device, vehicle computer system, or virtual / augmented reality device. In some examples, one or both of the client computing device and the authentication computing device may be implemented as a computing system (500) described below in relation to FIG. 5.

[0010] Additionally, the steps of the method (200) are generally divided between two different computing devices—a client computing device and an authentication computing device—but are not limited thereto. In other examples, the steps of the method (200) may be performed by more than two different computing devices. Alternatively, the steps of the method (200) may be performed entirely by a single computing device, for example, by a client computing device. In other words, a single computing device may perform the roles of both the client computing device and the authentication computing device.

[0011] Referring to FIG. 2a, in 202, the method (200) includes the step of securing one or more computer resources. Any suitable hardware or software resources of any suitable computing device may be secured through biometric authentication as described herein. Such resources may include physical devices (e.g., processors, sensors, storage devices), computer data (e.g., documents, databases, computer code), user accounts / profiles, software applications and / or other security components, content, or functions of the computing device. In other words, resources of the client computing device, resources of the authenticating computing device, and / or resources of any other suitable computing device accessible through a computer network may be secured until successful biometric authentication is achieved. This is reflected in FIG. 2a, where step 202 is illustrated between the client and the authenticating computing device to indicate that resources of one of the devices may be secured, resources of both may be secured, or resources of neither device may not be secured. Prior to authorization, the computing device may be configured to block user access to secure computer resources (e.g., denying login and / or data decryption).

[0012] In 204, the method (200) includes the step of measuring a reference biometric identifier on a client computing device. In some examples, this may be performed during the setup or registration phase, during which a new user is authorized to access the computer resource until the computer resource is first secured and / or subsequent biometric authentication. However, the reference biometric identifier may also be collected at any appropriate time. Additionally, the reference biometric identifier may be periodically refreshed or replaced in some cases—for example, to account for changes in the user's appearance or other biometric features over time.

[0013] An exemplary biometric identifier (300) is schematically illustrated in FIG. 3. In particular, while FIG. 3 is described in relation to a reference biometric identifier, it will be understood that the biometric identifier (300) may correspond to a reference biometric identifier or to an authenticated biometric identifier. Similarly, the transformation and obfuscation described below in relation to the biometric identifier (300) may be applied to one or both of the reference biometric identifier and the authenticated biometric identifier.

[0014] In this example, the biometric identifier is a fingerprint. However, both the reference biometric identifier and the authentication biometric identifier may take on appropriate forms and, generally, will be any type of measurable or quantifiable information related to the user that is maintained relatively fixed and can subsequently be used to verify the user's identity. As a non-limiting example, appropriate biometric identifiers may include a detected user face, iris, retina, fingerprint, palm print, voice, DNA sequence, or handwriting sample (e.g., signature). Such biometric identifiers may be "detected" by any appropriate biosensor in any appropriate manner. For example, such sensors may include cameras, scanners, microphones, pressure / touch sensors, or chemical analyzers.

[0015] Referring to FIG. 2a, the method (200) in 206 includes the step of converting a reference biometric identifier into a reference biometric vector on a client computing device. This is schematically illustrated in FIG. 3, whereby the biometric identifier (300) is converted into a biometric vector (302). In the illustrated example, the biometric vector (302) is a three-dimensional vector defined by three values ​​of magnitude (r), polar angle (Θ), and azimuth (φ). However, this is for the sake of simplicity. In practice, the biometric vector may include any number of dimensions, such as hundreds, thousands, or millions of dimensions. For the purposes of this disclosure, the biometric vector will be described as having "N" dimensions, where N can be any suitable positive integer.

[0016] The way a biometric identifier is converted into a biometric vector can vary significantly depending on the implementation and type of the biometric identifier being measured. Any vector representation may be compatible with the present disclosure. If the biometric identifier is an image (e.g., a human face, fingerprint, or iris), other dimensions of the vector may optionally correspond to image pixel values ​​or image features (e.g., edges, corners, blobs) recognized or extracted from the image. Generally, any suitable technique for feature extraction may be used, including edge / corner detection, blob detection / extraction, template matching, etc. In some cases, the vector transformation function may be tuned to a specific type of biometric data (e.g., a vector transformation function specifically trained to convert an image of a human face into a 128-dimensional feature vector). In other cases, more general feature extraction functions such as Speed-Up Robust Features (SURF) or Scale-Invariant Feature Transform (SIFT) may be used.

[0017] For non-visual biometric identifiers, other appropriate feature extraction techniques may be used. For example, speech samples can be converted into feature vectors by sampling the speech samples in the time domain and providing amplitudes for the audio data at regular intervals. Each individual sample of the audio data, or 'slice,' can be used to generate values ​​for different dimensions of the feature vector. Alternatively, a frequency domain representation of the speech samples may be used, where values ​​for the vector dimensions can be assigned based on the frequencies and magnitudes of different frequency components present in the audio data.

[0018] Regardless of the type of the initial biometric identifier and the nature of the vector transformation, the reference biometric vector will take the form of multiple individual values ​​corresponding to the dimensions of the individual vectors. These values ​​may have any appropriate range, such as, for example, between 0 and 1, between -1 and 1, or between 0 and 100. Depending on the implementation, the values ​​of the feature vector may be normalized, compressed, or modified in some cases, for example, to conserve storage space / network bandwidth or to mitigate the effects of outliers or signal noise.

[0019] Referring to FIG. 2a, in 208, the method (200) includes the step of obfuscating a reference biometric vector into an obfuscated reference biometric vector using similarity-preserving obfuscation. This is also schematically illustrated in FIG. 3, where a biometric vector (302) is obfuscated through similarity-preserving obfuscation (305) to produce an obfuscated biometric vector (306). As illustrated, during similarity-preserving obfuscation, the deflection (Θ) and azimuth (φ) of the biometric vector are rotated by different values ​​to produce obfuscated deflection Θo and azimuth φo. The rotation (or other transformation) applied during similarity-preserving obfuscation may be determined based on an authentication credential (304), which is described in more detail below.

[0020] 'Similarity-preserving obfuscation' can take the form of any appropriate function that can be applied to a biometric vector to modify the information encoded by the biometric vector. In particular, similarity-preserving obfuscation modifies the biometric vector in a way that mitigates or eliminates the possibility for a malicious user to reconstruct the original biometric identifier from the obfuscated biometric vector. It is noted that obfuscation maintains the similarity between two independent vectors before and after obfuscation.

[0021] Similarity-preserving obfuscation can be deterministic obfuscation in some cases, meaning that similarity-producing obfuscation will always produce exactly the same output for a given input. In other words, for a similarity-preserving obfuscation function E, the similarity S1 between two initial biometric vectors A and B is identical to the similarity S2 between two obfuscated biometric vectors Ao and Bo.

[0022] S1=A, B

[0023] S2=E(A), E(B)=Ao, Bo=S1

[0024] Depending on the implementation, similarity-preserving obfuscation can take various forms. As one example, similarity-preserving obfuscation can be a deterministic vector rotation. For instance, if two biometric vectors are each located on an N-dimensional unit sphere, the similarity between these vectors can be compared by calculating the cosine angle distance. Thus, the similarity-preserving obfuscation function can be any appropriate N-dimensional rotation, which preserves the angle and thus maintains the similarity between the obfuscated vectors. Alternatively, the similarity between two vectors can be determined by calculating an L2 norm. Thus, similarity-preserving obfuscation can take the form of an N-dimensional Euclidean transformation in which angles and distances are preserved. Therefore, the L2 norm calculated for input vectors A and B is identical to the L2 norm calculated for obfuscated vectors Ao and Bo.

[0025] Alternatively, similarity-preserving obfuscation is not entirely deterministic and can preserve the similarity of independent vectors before and after obfuscation. In other words, if the similarity between two input vectors A and B is substantially equal to the possible output vectors Ao and Bo, then for a given input vector A, similarity-preserving obfuscation can generate a range of possible output vectors Ao.

[0026] In any case, the exact transformation applied as part of the similarity-preserving obfuscation can be generated in any appropriate manner. In some cases, in the similarity-preserving obfuscation, the received authentication credentials may be used as a seed. This is illustrated in FIG. 3, where the similarity-preserving obfuscation uses the authentication credentials (304) as a seed. For example, the authentication credentials may be input as a seed to a random number generator to generate a sequence of rotations or other transformations to be applied to the input biometric vector. By subsequently providing the same (or substantially similar) authentication credentials during the obfuscation of the subsequent vector (e.g., the authentication biometric vector), the same (or substantially similar) sequence of rotations or other transformations is generated to preserve vector similarity.

[0027] Authentication credentials can be consistently reproduced by taking the form of any suitable information or data that is maintained relatively statically. For example, authentication credentials may take the form of a user-provided password or PIN, a device identifier (e.g., MAC address), or a numerical representation of a biometric identifier such as a fingerprint. If a password, PIN, or other credentials that may change over time are used, changing these credentials may trigger a new obfuscation of the reference biometric vector using the new password / PIN, etc., as a new seed (i.e., repeat step 208 and optionally repeat steps 204 and 206 of FIG. 2a).

[0028] Returning to FIG. 2a, at 210, the method (200) includes the step of transmitting an obfuscated reference biometric vector from a client computing device to an authenticating computing device. At 212, the method (200) includes the step of storing the previously obfuscated reference biometric vector from the reference biometric vector at the client computing device using similarity-preserving obfuscation at the authenticating computing device. For example, the obfuscated reference biometric vector may be stored as part of user access credentials and / or a profile, and the authenticating computing device may use it to authorize user access to any restricted computer resources on any cooperative computing device.

[0029] The obfuscated reference biometric vector may be transmitted in any appropriate manner through any suitable computer network. In some examples, the obfuscated reference biometric vector may be transmitted over the Internet. In some embodiments, the obfuscated reference biometric vector may be further encrypted for network transmission. Alternatively, as discussed above, the steps of the method (200) may be implemented by a single device in some cases, in which case the obfuscated reference biometric vector may be stored by a client computing device for authentication purposes without needing to be transmitted to an authentication computing device.

[0030] The method (200) continues in FIG. 2b. In 214, the method (200) includes the step of measuring an authentication biometric vector on a client computing device. This occurs after an obfuscated reference biometric vector has previously been transmitted to the authentication computing device. Typically, the authentication biometric vector is measured when a user attempts to access one or more secure computing resources, at which time the user must submit biometric authentication.

[0031] The authentication biometric vector may be measured in substantially the same manner as described above for the reference biometric identifier. In other words, the authentication biometric vector may be, as a non-limiting example, a detected user face, iris, retina, fingerprint, palm print, voice, DNA sequence, or handwriting sample (e.g., signature), and may be detected by any suitable biometric sensor. In any case, the authentication and reference biometric identifiers correspond to the same biometric attribute of the user; that is, if the reference biometric identifier is a detected fingerprint, the authentication biometric identifier will be a new measurement of the same fingerprint.

[0032] Referring further to FIG. 2b, the method (200) includes the step of converting an authenticated biometric identifier into an authenticated biometric vector in 216. In 218, the method (200) includes the step of obfuscating the authenticated biometric vector into an obfuscated authenticated biometric vector. In 220, the method (200) includes the step of transmitting the obfuscated authenticated biometric vector to an authenticated computing device. In 222, the method (200) includes the step of storing the obfuscated authenticated biometric vector at least temporarily in an authenticated computing device, wherein the obfuscated authenticated biometric vector is one that has been previously obfuscated from the authenticated biometric vector using similarity-preserving obfuscation in a client computing device. Each of these steps may be performed substantially as described above with respect to the reference biometric identifier, the reference biometric vector, and the obfuscated reference biometric identifier.

[0033] In particular, the similarity between the reference biometric vector and the authentication biometric vector is preserved during similarity-preserving obfuscation only when the same authentication credential is provided for both vectors. Therefore, even if a malicious user can provide an authentication biometric identifier that is similar to or identical to the authorized user's actual biometric identifier, the malicious user cannot access the secure computer resources unless this authentication credential is known.

[0034] Method (200) continues in FIG. 2c. In FIG. 224, method (200) includes the step of testing similarity between an obfuscated reference biometric vector and an obfuscated authentication biometric vector. This is schematically illustrated in FIG. 4. As illustrated, a client computing device (100) transmits an obfuscated reference biometric vector (400) and an obfuscated authentication biometric vector (402) to an authentication computing device (108). In particular, any time may elapse between the transmission of the two obfuscated vectors, that is, the obfuscated authentication biometric vector may be transmitted after the obfuscated reference biometric vector in seconds, minutes, hours, days, weeks, or years. When the two obfuscated vectors are received by the authentication computing device, the authentication computing device tests similarity between the two vectors.

[0035] As discussed above, the similarity between two obfuscated vectors can be tested in any appropriate manner. As one unlimited example, if the similarity-preserving obfuscation is a vector rotation, testing the similarity of the obfuscated vectors may involve calculating the cosine angle distance between the two vectors. Alternatively, if the similarity-preserving obfuscation is a Euclidean transformation, testing the similarity between the obfuscated vectors may involve calculating the L2 norm between the two vectors. In general, an appropriate method may be used to compare the similarity between two vectors.

[0036] Referring to FIG. 2c, in 226, the method (200) includes the step of transmitting an authentication approval to a client computing device based on the similarity being within an authentication threshold. This is also illustrated in FIG. 4, where, after the authentication computing device (108) performs a similarity test, the authentication approval (404) is transmitted to the client computing device.

[0037] Any appropriate authentication threshold may be used to determine whether access to a secure computer resource should be authorized. As a non-limiting example, a 99% confidence interval may be used. Generally, the higher the authentication threshold, the stronger the security of the computer resource, but at the same time, the risk of potential false negatives increases. Depending on the implementation, system security can reach a desirable threshold while balancing usability. Various authentication thresholds may be described depending on the type of biometric identifier collected. For example, images of a user's face may exhibit noise due to lighting conditions, glasses, beards, etc., and therefore have a relatively lower authentication threshold compared to other biometric identifiers that are less sensitive to noise, such as fingerprints. Furthermore, in some examples, various authentication thresholds may be set for each security resource—for example, a relatively higher authentication threshold may be used for relatively more sensitive resources.

[0038] Furthermore, the present disclosure has so far assumed that an obfuscated authentication biometric vector would be compared to a single obfuscated reference biometric vector stored in an authentication computing device. However, in practice, an authentication computing device may store any number of various obfuscated reference biometric vectors, and an obfuscated authentication biometric vector may be compared to some or all of the stored reference vectors. For example, one obfuscated reference biometric vector may correspond to an image of a user's face wearing glasses, and another obfuscated reference biometric vector may correspond to a different image of a user not wearing glasses. The authentication computing device may optionally receive and store multiple obfuscated reference biometric vectors corresponding to multiple different types of biometric identifiers, for example, one or more vectors may correspond to an image of a user's face, while one or more other vectors may correspond to the user's fingerprint(s).

[0039] If multiple obfuscated reference biometric vectors are stored, authentication authorization may be transmitted if an obfuscated authentication biometric vector passes a similarity test with any of the stored obfuscated reference biometric vectors. In other words, an obfuscated authentication biometric vector can be tested individually against each stored obfuscated reference biometric vector, and access to a secure resource may be authorized if at least one pair of vectors matches within the authentication threshold. Alternatively, access to a secure computer resource may be authorized only if the obfuscated authentication biometric vector matches one or more stored obfuscated reference biometric vectors within the authentication threshold.

[0040] Authentication authorization may take any appropriate form. For example, authentication authorization may take the form of a decryption key. Additionally, in some cases, authentication authorization may include other appropriate information, for example, a trust interval in which an obfuscated authentication biometric vector matches an obfuscated reference biometric vector, or a manifest indicating which specific security resource has been made available.

[0041] Referring to FIG. 2c, the method (200) in 228 includes the step of authorizing access to a secure computer resource based on receiving an authentication approval from a client computing device. Conversely, if the authentication computing device reports that access is not authorized, the client computing device may prompt the user to provide a new authentication biometric identifier, verify the user's identity using other methods, ask the owner or administrator of the secure computer for assistance, or simply deny authorization of access to the secure computer resource.

[0042] In some cases, the authentication authorization may eventually expire, and access to the secure computer resource may be limited to a single session or a fixed time limit. Therefore, when the authentication authorization expires, the computer resource may be protected again until the biometric authentication process is successfully repeated. Thus, the method (200) may return to step (214) or another appropriate step before the secure computer resource can be unlocked again.

[0043] In addition to this, or alternatively, the obfuscated reference biometric vector stored by the authentication computing device may expire, and a new reference biometric identifier may need to be collected before the secure computing resource can be accessed. Accordingly, the method (200) may return to step (202) or another appropriate step to submit a new obfuscated reference biometric vector to the authentication computing device.

[0044] The methods and processes described herein may be associated with a computing system of one or more computing devices. In particular, such methods and processes may be implemented as an executable computer application program, a network-accessible computing service, an application-programming interface (API), a library, or a combination of computing resources and / or other computing resources.

[0045] FIG. 5 schematically illustrates a simplified representation of a computing system (500) configured to provide any of the computing functions described herein. The computing system (500) may take the form of one or more personal computers, network-accessible server computers, tablet computers, home entertainment computers, gaming devices, mobile computing devices, mobile communication devices (e.g., smartphones), virtual / augmented / mixed reality computing devices, wearable computing devices, Internet of Things (IoT) devices, embedded computing devices, and / or other computing devices.

[0046] The computing system (500) includes a logic subsystem (502) and a storage subsystem (504). The computing system (500) may optionally include a display subsystem (506), an input subsystem (508), a communication subsystem (510), and / or other subsystems not shown in FIG. 5.

[0047] A logical subsystem (502) includes one or more physical devices configured to execute instructions. For example, the logical subsystem may be configured to execute instructions that are part of one or more applications, services, or other logical configurations. The logical subsystem may include one or more hardware processors configured to execute software instructions. In addition to or otherwise, the logical subsystem may include one or more hardware or firmware devices configured to execute hardware or firmware instructions. The processors of the logical subsystem may be single-core or multi-core, and the instructions executed therein may be configured for sequential processing, parallel processing, and / or distributed processing. Optionally, individual components of the logical subsystem may be distributed across two or more individual devices, which may be remotely located and / or configured for coordinated processing. Aspects of the logical subsystem may be virtualized and executed by a remotely accessible network computing device configured in a cloud computing configuration.

[0048] The storage subsystem (504) includes one or more physical devices configured to temporarily and / or permanently hold computer information, such as data and instructions, executable by the logic subsystem. If the storage subsystem includes two or more devices, the devices may be located together and / or remotely. The storage subsystem (504) may include volatile, non-volatile, dynamic, static, read / write, read-only, random-access, sequential-access, location-addressable, file-addressable, and / or content-addressable devices. The storage subsystem (504) may include removable devices and / or embedded devices. When the logic subsystem executes instructions, the state of the storage subsystem (504) may be changed to hold different data, for example.

[0049] The aspects of the logic subsystem (502) and the storage subsystem (504) may be integrated together into one or more hardware logic components. These hardware logic components may include PASIC / ASIC (program- and application-specific integrated circuit), PSSP / ASSP (program- and application-specific standard product), SOC (system-on-a-chip), and CPLD (complex programmable logic device).

[0050] The logic subsystem and the storage subsystem may collaborate to instantiate one or more logic machines. As used herein, the term “machine” is used to collectively refer to a combination of hardware, firmware, software, instructions, and / or other components that collaborate to provide computer functions. In other words, “machine” is not an abstract idea and always takes a tangible form. A machine may be instantiated by a single computing device, or a machine may include two or more subcomponents instantiated by two or more different computing devices. In some embodiments, a machine includes a local component (e.g., a software application executed by a computer processor) that collaborates with a remote component (e.g., a cloud computing service provided by a network of server computers). Software and / or other instructions that provide functions to a particular machine may optionally be stored as one or more unexecuted modules in one or more suitable storage devices.

[0051] If a display subsystem (506) is included, it may be used to present a visual representation of the data held by the storage subsystem (504). This visual representation may take the form of a graphical user interface (GUI). As the method and process described herein change the data held by the storage machine and thereby change the state of the storage machine, the state of the display subsystem (506) may likewise be changed to visually represent the change in the underlying data. The display subsystem (506) may include one or more display devices utilizing virtually any type of technology. Such display devices may be coupled with the logic machine (502) and / or the storage machine (504) of the shared enclosure, or such display devices may be peripheral display devices.

[0052] If an input subsystem (508) is included, it may include or interface with one or more input devices, such as a keyboard, mouse, touch screen, or game controller. In some embodiments, the input subsystem may include or interface with selected natural user input (NUI) components. These components may be integrated or peripheral devices, and the conversion and / or processing of input operations may be handled onboard or offboard. Exemplary NUI components may include a microphone for speech and / or voice recognition; an infrared, color, stereoscopic, and / or depth camera for machine vision and / or gesture recognition; a head tracker, eye tracker, accelerometer and / or gyroscope for motion detection and / or intent recognition, as well as an electric field sensing component for evaluating brain activity. Additionally, the input subsystem (508) may include any sensor capable of collecting biometric identifiers as discussed above, such as, for example, a camera, scanner, pressure sensor, or chemical analyzer.

[0053] If a communication subsystem (610) is included, it may be configured to enable the computing system (600) to communicate with one or more other computing devices. The communication subsystem (610) may include wired and / or wireless communication devices compatible with one or more different communication protocols. As an example without limitation, the communication subsystem may be configured for communication over a wireless telephone network, or a wired or wireless local area or wide area network. In some embodiments, the communication subsystem may enable the computing system (500) to send and / or receive messages from and to other devices over a network such as the Internet.

[0054] The methods and processes disclosed herein may be configured to provide the user and / or any other human being with control over any personal data and / or potentially sensitive data. Whenever data is stored, accessed, and / or processed, the data may be handled in accordance with privacy and / or security standards. When user data is collected, the user or other relevant parties may specify how the data is used and / or stored. Whenever user data is collected for any purpose, the user data may be collected only with the utmost respect for user privacy (e.g., user data may be collected only where the user who owns the data provides positive consent and / or the user who owns the data is notified whenever user data is collected). If data is disclosed for access by a human being other than the user or used in a decision-making process, the user's consent may be collected before the data is used and / or disclosed. The user may opt in and / or opt out of data collection at any time. After data has been collected, the user may issue a command to delete the data and / or restrict access to the data. All potentially sensitive data may be optionally encrypted and / or anonymized where possible to provide additional protection for user privacy. Users may specify, for example, data, metadata, or parts of the statistics / results of processed data to be disclosed to other parties for further processing. Personal data and / or confidential data may be kept completely private, for example, being decrypted only temporarily for processing or only for processing on the user's device, otherwise stored in an encrypted form. Users may retain and control the encryption keys for the encrypted data.Alternatively, or in addition to this, the user may, for example, designate a trusted third party that holds and controls the encryption key for the encrypted data to provide the user with access to the data according to an appropriate authentication protocol.

[0055] Where the methods and processes disclosed herein include machine learning (ML) and / or artificial intelligence (AI) components, the ML and / or AI components may make decisions based at least partially on the training of the components regarding training data. Accordingly, the ML and / or AI components may be trained on diverse and representative datasets containing sufficient relevant data regarding various users and / or user groups. In particular, since the training datasets may include data regarding various human individuals and groups, performance is improved in relation to the user experience of the users and / or user groups as the ML and / or AI components are trained.

[0056] ML and / or AI components can be further trained to make decisions that minimize potential bias toward human individuals and / or groups. For example, when an AI system is used to evaluate qualitative and / or quantitative information about human individuals or groups, it can be trained not to change based on differences between individuals or groups that are not intended to be measured by the qualitative and / or quantitative information, thereby ensuring that decision-making is not influenced in unintended ways, for example, by differences between individuals and groups.

[0057] ML and / or AI components are designed to provide context regarding their behavior, allowing the humans implementing the ML and / or AI systems to take responsibility for decisions or evaluations made by the systems. For example, ML and / or AI systems may be configured for reproducible behavior; for instance, when making pseudo-random decisions, a random seed may be used and recorded so that the decision can be replicated later. As another example, data used to train and / or test ML and / or AI systems may be curated and retained to facilitate future investigations into the behavior of the ML and / or AI systems related to this data. Additionally, ML and / or AI systems may be continuously monitored to identify potential biases, errors, and / or unintended consequences.

[0058] The present disclosure is presented by way of example and with reference to the associated drawings. Components, processing steps, and other elements that may be substantially identical in one or more drawings are considered equivalent and are described in repetition as minimal. However, it should be noted that elements considered equivalent may differ to some extent. Further note is that some drawings are schematic and may not be drawn to actual scale. Various drawing scales, aspect ratios, and number of components shown in the drawings may be intentionally distorted to make certain features or relationships more easily recognizable.

[0059] For example, as a method for authorizing access to one or more secure computer resources, the method comprises the steps of: storing an obfuscated reference biometric vector on an authenticating computing device—this obfuscated reference biometric vector is previously obfuscated from the reference biometric vector using similarity-preserving obfuscation on a client computing device—and receiving an obfuscated authentication biometric vector—the obfuscated authentication biometric vector is previously obfuscated from the authentication biometric vector using similarity-preserving obfuscation on a client computing device—and testing the similarity between the obfuscated reference biometric vector and the obfuscated authentication biometric vector on the authenticating computing device; and, based on the similarity being within an authentication threshold, communicating an authentication authorization from the authenticating computing device to the client computing device to authorize access to one or more secure computer resources. For example, or in another example, similarity-preserving obfuscation is deterministic obfuscation. For example or in another example, similarity-preserving obfuscation is a deterministic vector rotation. For example or in another example, the step of testing the similarity between an obfuscated reference biometric vector and an obfuscated authenticated biometric vector includes the step of calculating the cosine angle distance between the obfuscated reference biometric vector and the obfuscated authenticated biometric vector. For example or in another example, similarity-preserving obfuscation is a deterministic Euclidean transformation. For example or in another example, the step of testing the similarity between an obfuscated reference biometric vector and an obfuscated authenticated biometric vector includes the step of calculating the L2 norm between the obfuscated reference biometric vector and the obfuscated authenticated biometric vector.For example or another example, in similarity-preserving obfuscation, the received authentication credential is used as a seed. For example or another example, the received authentication credential is a password. For example or another example, the received authentication credential is a device identifier. For example or another example, the received authentication credential is a biometric identifier.

[0060] For example, a method for allowing access to one or more secure computer resources comprises the steps of: obfuscating a reference biometric vector into an obfuscated reference biometric vector using similarity-preserving obfuscation on a client computing device; transmitting the obfuscated reference biometric vector from the client computing device to an authenticating computing device; measuring an authenticating biometric identifier on the client computing device; converting the authenticating biometric identifier into an authenticating biometric vector on the client computing device; obfuscating the authenticating biometric vector into an obfuscated authenticating biometric vector using similarity-preserving obfuscation on the client computing device; transmitting the obfuscated authenticating biometric vector from the client computing device to the authenticating computing device; and allowing access to one or more secure computer resources based on receiving an authenticating approval from the authenticating computing device, wherein the authenticating approval is predetermined on the authenticating computing device based on the similarity between the obfuscated reference biometric vector and the obfuscated authenticating biometric vector being within an approval threshold. For example or in another example, the reference biometric vector is transformed from the reference biometric identifier measured by the client computing device. For example or in another example, similarity-preserving obfuscation is a deterministic vector rotation. For example or in another example, similarity-preserving obfuscation is a deterministic Euclidean transformation. For example or in another example, in similarity-preserving obfuscation, the received authentication credential is used as a seed. For example or in another example, the authentication biometric identifier is a detected human fingerprint. For example or in another example, the authentication biometric identifier is a detected human iris. For example or in another example, the authentication biometric identifier is a detected human face. For example or in another example, the authentication biometric identifier is a detected human voice.

[0061] For example, a method for authorizing access to one or more secure computer resources comprises the steps of obfuscating a reference biometric vector into an obfuscated reference biometric vector using similarity-preserving obfuscation, obfuscating an authentication biometric vector into an obfuscated authentication biometric vector using similarity-preserving obfuscation, testing the similarity between the obfuscated authentication biometric vector and the obfuscated reference biometric vector, and authorizing access to one or more secure computer resources based on the similarity being within an authentication threshold.

[0062] It will be understood that the configurations and / or methods described herein are exemplary by nature, and that these specific embodiments or examples should not be interpreted restrictively, as numerous variations are possible. Specific routines or methods described herein may represent one or more of any number of processing strategies. Accordingly, the various operations illustrated may be performed in the order illustrated, in a different order, in parallel, or omitted in some cases. Likewise, the order of the processes described above may be changed.

[0063] The subject matter of the claims of the present disclosure includes various processes, systems and configurations, and all novel and non-obvious combinations and sub-combinations of other features, functions, operations and / or characteristics disclosed in this specification and all equivalents thereof.

Claims

Claim 1 A method for authorizing access to one or more secure computer resources, comprising the steps of: storing an obfuscated reference biometric vector on an authentication computing device—wherein the obfuscated reference biometric vector is previously obfuscated from a reference biometric vector using similarity-preserving obfuscation on a client computing device, wherein the reference biometric vector comprises N dimensions, and the similarity-preserving obfuscation is a deterministic vector rotation in which a rotation sequence is applied to one or more of the N dimensions, and wherein the rotation sequence is generated based on a seed value—and receiving an obfuscated authentication biometric vector—wherein the obfuscated authentication biometric vector is previously obfuscated from an authentication biometric vector using similarity-preserving obfuscation on the client computing device—and, on the authentication computing device, testing the similarity between the obfuscated reference biometric vector and the obfuscated authentication biometric vector. A method comprising the step of communicating an authentication approval from the authentication computing device to the client computing device based on the similarity being within an authentication threshold to authorize access to the one or more secure computer resources, wherein in the similarity-preserving obfuscation, the received authentication credential is used as the seed value for the random number generator to generate the rotation sequence. Claim 2 delete Claim 3 delete Claim 4 A method according to claim 1, wherein the step of testing the similarity between the obfuscated reference biometric vector and the obfuscated authenticated biometric vector comprises the step of calculating the cosine angle distance between the obfuscated reference biometric vector and the obfuscated authenticated biometric vector. Claim 5 delete Claim 6 A method according to claim 1, wherein the step of testing the similarity between the obfuscated reference biometric vector and the obfuscated authenticated biometric vector comprises the step of calculating the L2 norm between the obfuscated reference biometric vector and the obfuscated authenticated biometric vector. Claim 7 In claim 1, the received authentication credential is a password. Claim 8 In claim 1, the method wherein the received authentication credential is a device identifier. Claim 9 A method according to claim 1, wherein the received authentication credential is a biometric identifier. Claim 10 A method for authorizing access to one or more secure computer resources, comprising the steps of: obfuscating a reference biometric vector into an obfuscated reference biometric vector using similarity-preserving obfuscation on a client computing device—wherein the reference biometric vector comprises N dimensions, and the similarity-preserving obfuscation is a deterministic vector rotation in which a rotation sequence is applied to one or more of the N dimensions, and the rotation sequence is generated based on a seed value—; transmitting the obfuscated reference biometric vector from the client computing device to an authentication computing device; measuring an authentication biometric identifier on the client computing device; converting the authentication biometric identifier into an authentication biometric vector on the client computing device; obfuscating the authentication biometric vector into an obfuscated authentication biometric vector using the similarity-preserving obfuscation on the client computing device; transmitting the obfuscated authentication biometric vector from the client computing device to the authentication computing device; and receiving an authentication acknowledgment from the authentication computing device, wherein one A method comprising the step of authorizing access to the above-mentioned secure computer resources—wherein the authentication authorization is previously determined in the authentication computing device based on the similarity between the obfuscated reference biometric vector and the obfuscated authentication biometric vector being within an authentication threshold—wherein the similarity-preserving obfuscation, the received authentication credential is used as the seed value for the random number generator to generate the rotation sequence. Claim 11 In claim 10, the method wherein the reference biometric vector is converted from a reference biometric identifier measured at the client computing device. Claim 12 delete Claim 13 delete Claim 14 The method of claim 10 or 11, wherein the authentication biometric identifier is a detected human fingerprint. Claim 15 The method according to claim 10 or 11, wherein the authentication biometric identifier is a detected human iris. Claim 16 The method of claim 10 or 11, wherein the authentication biometric identifier is a detected human face. Claim 17 The method of claim 10 or 11, wherein the authentication biometric identifier is a detected human voice. Claim 18 A method for authorizing access to one or more secure computer resources, performed by a computing system comprising a client computing device and an authentication computing device, comprising: a step by the client computing device of obfuscating a reference biometric vector into an obfuscated reference biometric vector using similarity-preserving obfuscation, wherein the reference biometric vector comprises N dimensions, the similarity-preserving obfuscation is a deterministic vector rotation in which a rotation sequence is applied to one or more of the N dimensions, and the rotation sequence is generated based on a seed value; a step by the client computing device of obfuscating an authentication biometric vector into an obfuscated authentication biometric vector using the similarity-preserving obfuscation; a step by the authentication computing device of testing the similarity between the obfuscated authentication biometric vector and the obfuscated reference biometric vector; and a step by the authentication computing device of authorizing access to the one or more secure computer resources based on the similarity being within an authentication threshold, wherein the similarity-preserving obfuscation is a received authentication A method for generating the rotation sequence using the credential as the seed value for the random number generator.

Images