Non-invasive biometric systems and methods

Abstract

A non-invasive biometric system includes a processor configured to control a scanner configured to scan and capture one or more anatomical images of a target person's body. The processor is also configured to identify one or more anatomical structures in the captured anatomical images and extract anatomical features from the identified anatomical structures. The processor is further configured to register the extracted anatomical features of the identified anatomical structures to the target person's posture and appearance. The processor is also configured to encode and utilize the extracted anatomical features as biometric data unique to the target person and usable for authentication of the target person.

Description

Technical Field

[0001] This disclosure generally relates to the field of non-invasive biometric systems, and more specifically to non-invasive biometric systems and methods for encoding anatomical features into biometric data. Background Technology

[0002] Typically, biometrics is a method used to uniquely identify a human based on one or more physical or behavioral characteristics. Examples of physical biometrics include fingerprint recognition, iris recognition, and facial recognition. Examples of behavioral biometrics include typing patterns, mouse movement recognition, and gait recognition. The selection of one or more biometrics for an application depends on several factors. Examples of these factors include: (a) universality (i.e., each target person should possess the biometrics being measured); (b) uniqueness (i.e., the measured characteristics should be sufficiently identifiable and distinguishable from each individual in the target group); (c) durability (i.e., the quality of the measured biometrics reasonably remains unchanged over time relative to a particular matching algorithm); (d) measurability (i.e., the ease with which the characteristics are acquired or measured); (e) performance (i.e., the accuracy, speed, and robustness of the technology used); (f) acceptability (i.e., the degree to which the target group accepts the technology); and (g) non-avoidance (i.e., the difficulty in imitating or replacing the biometrics). Existing biometric systems and methods rely on appearance-related features (e.g., fingerprints, iris characteristics, voice, etc.), which can be detected accurately but are also more easily tampered with. For example, in other forgery techniques, plastic surgery has made it relatively easier for malicious individuals to bypass some common biometrics. In another example, some biometrics are robust but more difficult to collect (e.g., DNA, etc.). Therefore, there is a technical problem of how to develop non-invasive biometric systems that satisfy all of the above factors without sacrificing accuracy and ease of use.

[0003] By comparing such a system with some aspects of this disclosure as set forth with reference to the accompanying drawings in the remainder of this application, additional limitations and disadvantages of conventional and traditional methods will become apparent to those skilled in the art. Summary of the Invention

[0004] This disclosure provides a non-invasive biometric system and a method for encoding anatomical features into biometric data, as generally shown and / or described in conjunction with at least one accompanying drawing and set forth more fully in the claims.

[0005] In one aspect, this disclosure provides a non-invasive biometric system including a processor. The processor is configured to control a scanner configured to scan and capture one or more anatomical images of a target person's body. The processor is also configured to identify one or more anatomical structures in the captured anatomical images and extract anatomical features from the identified anatomical structures. The processor is further configured to register the extracted anatomical features of the identified anatomical structures to the target person's posture and appearance. The processor is also configured to encode and utilize the extracted anatomical features as biometric data.

[0006] In a possible implementation, the eigenvector is a discriminative eigenvector.

[0007] A non-invasive biometric system is provided for non-invasive biometric authentication based on imaging of in vivo organs. This non-invasive biometric system includes a processor for extracting and encoding robust and unique anatomical features to obtain and utilize biometric data, which benefits the persistence and non-evasion factors of the non-invasive biometric system. Because the biometric data is based on anatomical features, which are unique to each target individual, the biometric data cannot be tampered with, for example, even through plastic surgery or other means. Furthermore, the disclosed biometric system is non-invasive, accurate, and fail-safe.

[0008] It should be understood that all the above embodiments can be combined. It must be noted that all devices, elements, circuits, units, and apparatuses described in this application can be implemented as software or hardware elements or any combination thereof. All steps performed by the various entities described in this application, and functions described as being performed by the various entities, are intended to mean that the respective entities are suited to or configured to perform the respective steps and functions. Even though specific functions or steps performed by external entities are not reflected in the detailed description of the specific elements of the entity performing such specific steps or functions in the following description of specific embodiments, it will be clear to those skilled in the art that these methods and functions can be implemented in the corresponding software or hardware elements or any combination thereof. It should be understood that the features of this disclosure are readily combined in various ways without departing from the scope of this disclosure as defined by the appended claims.

[0009] Additional aspects, advantages, features, and objects of this disclosure will become apparent from the accompanying drawings and the detailed description of exemplary embodiments interpreted in conjunction with the appended claims. Attached Figure Description

[0010] The above-described invention and the following detailed description of exemplary embodiments can be better understood when read in conjunction with the accompanying drawings. Exemplary structures of this disclosure are shown in the drawings for illustrative purposes. However, this disclosure is not limited to the specific methods and apparatus disclosed herein. Furthermore, those skilled in the art will understand that the drawings are not to scale. Wherever possible, the same elements are indicated by the same reference numerals.

[0011] Embodiments of this disclosure will now be described by way of example only with reference to the following figures, in which:

[0012] Figure 1A This is a network environment diagram of an exemplary non-invasive biometric system according to embodiments of the present disclosure;

[0013] Figure 1B This is a block diagram illustrating various exemplary components of a computing device according to embodiments of the present disclosure;

[0014] Figure 2 The diagram illustrates an exemplary scenario of an implementation of an exemplary invasive biometric system according to embodiments of the present disclosure; and

[0015] Figure 3A and Figure 3B Commonly, there are flowcharts illustrating a method for encoding anatomical features into biometric data according to embodiments of the present disclosure.

[0016] In the accompanying drawings, underlined reference numerals are used to indicate items on which the underlined reference numeral is located or items adjacent to it. Ununderlined reference numerals refer to items identified by a line connecting the ununderlined reference numeral to the item. When a reference numeral is ununderlined and accompanied by an associated arrow, the ununderlined reference numeral is used to identify the general item to which the arrow points. Detailed Implementation

[0017] The following detailed description illustrates exemplary embodiments of the present disclosure and how they can be implemented. Although some modes of carrying out the present disclosure have been disclosed, those skilled in the art will recognize that other embodiments for carrying out or practicing the present disclosure are also possible.

[0018] Figure 1A This is a network environment diagram of an exemplary non-invasive biometric system according to embodiments of the present disclosure. (Reference) Figure 1A The diagram illustrates the network environment of a non-invasive biometric system 100A, including a computing device 102, a scanner 104, and a communication network 106. The computing device 102 also includes a processor 102A and a database 102B. A target person 108 is also shown positioned within the scanning area of ​​the scanner 104.

[0019] like Figure 1A As an example, in one embodiment, the processor 102A of the non-invasive biometric system 100A is configured to control the scanner 104 to scan and capture one or more anatomical images of the body of a target person 108. The processor 102A is configured to identify one or more anatomical structures in the captured one or more anatomical images; extract anatomical features from the identified one or more anatomical structures; register the extracted anatomical features of the identified one or more anatomical structures to the pose and appearance of the target person; combine the extracted anatomical features into a single feature vector; and encode and store the single feature vector as biometric data. In one embodiment, the feature vector is an identification feature vector.

[0020] The non-invasive biometric system 100A corresponds to a biometric authentication system, which is based on, for example, scanning of anatomical images and encoding of anatomical features. The non-invasive biometric system 100A is based on encoding discriminative anatomical characteristics from a body scan of a target person 108. The non-invasive biometric system 100A is configured to use a set of non-invasive biometric features based on discriminative in vivo features (such as organs, bones, etc.). In this example, such features can be extracted from a radiograph of the target person 108 for authentication.

[0021] The computing device 102 may include suitable logic, circuitry, interfaces, and / or code configured to communicate with the scanner 104 via a communication network 106 (e.g., a propagation channel). The computing device 102 includes a processor 102A and a database 102B. Examples of the various computing devices 102 may include, but are not limited to, computer systems, personal digital assistants, portable computing devices, electronic devices, storage servers, cloud-based servers, web servers, application servers, or combinations thereof.

[0022] Processor 102A is configured to process input provided by scanner 104. Processor 102A is also configured to control scanner 104 to scan and capture one or more anatomical images of the body of target person 108. Examples of processor 102A may include, but are not limited to, processors, digital signal processors (DSPs), microprocessors, microcontrollers, complex instruction set computing (CISC) processors, application-specific integrated circuit (ASIC) processors, reduced instruction set computing (RISC) processors, very long instruction word (VLIW) processors, state machines, data processing units, graphics processing units (GPUs), and other processors or control circuits.

[0023] Database 102B can store biometric data. Furthermore, scanner 104 may include suitable logic, circuitry, interfaces, or code configured to scan and capture one or more anatomical images of the body of target person 108. In embodiments, scanner 104 corresponds to an X-ray scanner used for anatomical feature encoding. Examples of scanner 104 may include, but are not limited to, computed tomography (CT) scanners, magnetic resonance imaging (MRI) scanners, positron emission tomography (PET) scanners, ultrasound scanners, single-photon emission computed tomography (SPECT) scanners, or other medical imaging modalities. However, other scanners may be used without limiting the scope of the invention, provided that such scanners are also configured to scan and capture one or more anatomical images of the body of target person 108.

[0024] Communication network 106 includes a medium (e.g., a communication channel) through which computing device 102 communicates with scanner 104. Examples of communication network 106 may include, but are not limited to, cellular networks, wireless sensor networks (WSNs), cloud networks, local area networks (LANs), metropolitan area networks (MANs), and / or the Internet.

[0025] Beneficially, the non-invasive biometric system 100A is used for non-invasive biometric authentication based on imaging of in vivo organs. The non-invasive biometric system 100A satisfies various factors such as universality, uniqueness, durability, measurability, performance, acceptability, and non-circumvention without sacrificing accuracy and ease of use.

[0026] Figure 1B This is a block diagram of a computing device according to embodiments of the present disclosure. Reference Figure 1B A block diagram 100B of a computing device 102 is shown, which includes a memory 110, a network interface 112, and a display unit 118. The memory 110 also includes... Figure 1A Database 102B. Memory 110 is configured to store biometric data, such as first biometric data 114A, second biometric data 114B, up to Nth biometric data 114N. Multiple unique tags, such as first unique tag 116A, second unique tag 116B, up to Nth unique tag 116N, are also shown. Processor 102A and database 102B are also shown.

[0027] The memory 110 may include suitable logic, circuitry, interfaces, or code configured to store anatomical features as biometric data (such as storing first biometric data 114A, second biometric data 114B, up to Nth biometric data 114N). In embodiments, the memory 110 corresponds to local memory, such as electrically erasable programmable read-only memory (EEPROM), random access memory (RAM), read-only memory (ROM), central processing unit (CPU) cache memory, etc.

[0028] Network interface 112 includes interfaces configured to transmit data via network interface 112. Figure 1A The communication network 106 is the hardware or software that establishes communication between the computing device 102 and the scanner 104. Examples of the network interface 112 may include, but are not limited to, a computer port, a network socket, a network interface controller (NIC), and any other network interface device. Furthermore, the display unit 118 is used to display a subset of visual and location features belonging to one or more anatomical structures of the target person 108.

[0029] A non-invasive biometric system 100A is provided, including a processor 102A. The processor 102A can be configured to control a scanner 104 for scanning and capturing (…). Figure 1A The processor 102A can be configured to control the scanner 104 to capture one or more anatomical images of the body of the target person 108. In an embodiment, the one or more anatomical images of the body of the target person 108 captured by the scanner 104 can be visible on the display unit 118 of the computing device. In an embodiment, the scanner 104 corresponds to an X-ray scanner configured to allow X-rays to pass through the body of the target person 108 and then detect the passing X-rays. According to an embodiment, the scanner 104 can be at least one of a computed tomography (CT) scanner, a magnetic resonance imaging (MRI) scanner, a positron emission tomography (PET) scanner, an ultrasound scanner, a single-photon emission computed tomography (SPECT) scanner, or other medical imaging modalities. Therefore, the scanner 104 can be configured to capture one or more anatomical images of the body of the target person 108 based on the type of scanner 104. In an embodiment, one or more anatomical images allow medical personnel to view the interior of the body of the target person 108 without any risk of exploratory surgery. In one example, scanner 104 may be configured to capture cross-sectional images (or tomographic images) of the body of the target person 108. In another example, scanner 104 may be configured to capture three-dimensional images of the body of the target person 108.

[0030] The processor 102A can also be configured to identify one or more anatomical structures in one or more captured anatomical images. First, the processor 102A can be configured to receive one or more captured anatomical images, and then the processor 102A can be configured to identify one or more anatomical structures in the captured anatomical images. In this example, the identified one or more anatomical structures are also visible on the display unit 118 of the computing device 102, such as... Figure 2 Further illustrated and described herein. According to embodiments, the identified one or more anatomical structures may be at least one of a single organ, a group of organs, a single bone, a group of bones, the liver, the spleen, or a combination of body organs and bones. Moreover, the processor 102A may be configured to extract anatomical features of the identified one or more anatomical structures. In the example, the extracted anatomical features correspond to heartbeat, liver condition (e.g., stretching / compression or affected), shape and size of bone marrow, etc. However, the extracted anatomical features may correspond to other features without limiting the scope of the invention.

[0031] Processor 102A can also be configured to register extracted anatomical features of one or more identified anatomical structures to the posture and appearance of target person 108. In embodiments, processor 102A can be configured to register the extracted anatomical features in a robust manner (i.e., robust to organ stretching or organ compression). Moreover, the extracted anatomical features can be registered to the posture and appearance of target person 108, which is unique to target person 108. In an example, the posture of target person 108 may correspond to the position of target person 108 standing, sitting, or lying on the floor or bed. In another example, the appearance of target person 108 may correspond to the external phenotype or facial features of target person 108. According to embodiments, processor 102A can be configured to combine the extracted anatomical features into one or more discriminative feature vectors. In embodiments, the extracted anatomical features are obtained from multiple scans of target person 108 by scanner 104. Moreover, processor 102A can be configured to detect and extract any prominent anatomical structures from the multiple scans. Subsequently, processor 102A can be configured to combine all extracted anatomical features into one or more discriminative feature vectors for target person 108. Therefore, one or more discriminative feature vectors are unique to target person 108 and are also different from the discriminative feature vectors of other target persons.

[0032] Processor 102A can also encode and utilize the extracted anatomical features as biometric data. In one embodiment, processor 102A can be configured to use an encoding algorithm to encode the extracted anatomical features as biometric data. Furthermore, processor 102A can be configured to utilize the extracted anatomical features as biometric data. Subsequently, processor 102A can be configured to store the extracted anatomical features as biometric data in a database 102B of the memory 110 of computing device 102. According to an embodiment, processor 102A can also be configured to combine extracted anatomical features from one or more identified anatomical structures into biometric data. In such an embodiment, processor 102A can be configured to encode one or more identified anatomical structures and the position of each anatomical structure relative to other anatomical structures derived from the identified one or more anatomical structures.

[0033] According to an embodiment, processor 102A may also be configured to assign a unique tag to biometric data, and the unique tag indicates a target person 108. In an implementation, first biometric data 114A may include one or more identifying feature vectors of one or more anatomical structures of the target person 108. Moreover, processor 102A may also be configured to assign a first unique tag 116A to first biometric data 114A, such as the first unique tag 116A indicating the target person 108. In addition, the first unique tag 116A also serves as the identity of the target person 108, because the first unique tag 116A is unique to the target person 108. The first unique tag 116A assigned to first biometric data 114A can be used for authentication of the target person 108. Similarly, a second unique tag 116B may be assigned to second biometric data 114B corresponding to a second target person, and an Nth unique tag 116N may be assigned to Nth biometric data 114N corresponding to an Nth target person. According to an embodiment, processor 102A may also be configured to store biometric data together with unique tags in memory 110. Therefore, the biometric data stored in memory 110, along with a unique tag, can be used for later comparisons, such as during the registration phase (or authentication phase) of target person 108. According to an embodiment, processor 102A can also be configured to combine the extracted anatomical features with other biometric data of target person 108. In an example, other biometric data of target person 108 may include a color image of target person 108, which may be captured and processed along with a scan to extract facial features. Therefore, the extracted anatomical features and other biometric data of target person 108 can be used together for target person authentication. In an implementation, processor 102A can be configured to identify the target person using each individual anatomical feature separately.

[0034] According to an embodiment, processor 102A may also be configured to perform a query for one or more pre-stored discriminative feature vectors against a database 102B comprising a plurality of pre-stored discriminative feature vectors for authentication of target person 108. Processor 102A acquires and stores the plurality of pre-stored discriminative feature vectors. In an example, the plurality of pre-stored discriminative feature vectors correspond to a plurality of discriminative feature vectors obtained after scanning a plurality of target persons. In an implementation, the plurality of pre-stored discriminative feature vectors may include subsets of visual and location features belonging to one or more anatomical structures encoded as one or more discriminative feature vectors. In such an embodiment, processor 102A may also be configured to match at least two discriminative feature vectors in database 102B by comparing subsets of visual and location features belonging to one or more anatomical structures encoded as biometric data for authentication of target person 108. Processor 102A may be configured to execute an algorithm that compares a subset of the visual and location features of one or more discriminative feature vectors with a subset of the visual and location features of another discriminative feature vector. In the example, if a subset of the visual and location features of one or more discriminative feature vectors is identical to a subset of the visual and location features of another discriminative feature vector, then at least two discriminative feature vectors match; otherwise, they do not match. Therefore, the processor 102A of the non-invasive biometric system 100A can use the matching of at least two discriminative feature vectors from the database 102B to authenticate and verify the target person 108 with improved accuracy, speed, and robustness at various security checkpoints (e.g., in airports, office buildings).

[0035] The non-invasive biometric system 100A can be used for non-invasive biometric authentication based on imaging of in vivo organs. The processor 102A of the non-invasive biometric system 100A can be used to extract and encode robust and unique anatomical features to obtain and utilize biometric data, which benefits the persistence and non-circumvention factors of the non-invasive biometric system 100A. Because the biometric data is based on anatomical features, which are unique to each target individual, the biometric data cannot be altered, for example, not even through plastic surgery or other means. Furthermore, the non-invasive biometric system 100A is accurate and fail-safe. The non-invasive biometric system 100A also meets various factors such as universality, uniqueness, persistence, measurability, performance, acceptability, and non-circumvention without sacrificing accuracy and ease of use.

[0036] Figure 2 This is a diagram illustrating an exemplary scenario of an implementation of an exemplary invasive biometric system according to embodiments of the present disclosure. Combined with Figure 1A and Figure 1B Component description Figure 2 . refer to Figure 2It shows ( Figure 1A Figure 200 illustrates an exemplary scenario of an implementation of a non-invasive biometric system 100A. The non-invasive biometric system 100A includes a computing device 102 and a scanner 104 connected via a communication network 106. One or more anatomical structures, such as a first anatomical structure 202, a second anatomical structure 204, and a third anatomical structure 206, identified in an anatomical image of a target person 108 are also shown.

[0037] First, processor 102A can be configured to control scanner 104 via communication network 106 to scan and capture anatomical images of target person 108, such as the anatomical image of target person 108 shown on display unit 118 of computing device 102. Then, processor 102A can be configured to identify one or more anatomical structures in the anatomical image of target person 108 shown on display unit 118 of computing device 102, such as a first anatomical structure 202, a second anatomical structure 204, and a third anatomical structure 206. In an example, the first anatomical structure 202 corresponds to the kidney of target person 108, the second anatomical structure 204 corresponds to the heart of target person 108, and the third anatomical structure 206 corresponds to the liver of target person 108.

[0038] According to an embodiment, processor 102A can also be configured to process one or more identified anatomical structures to extract visual features, which are further registered together in database 102B to generate one or more discriminative feature vectors. In the example, after processing a first anatomical structure 202, a second anatomical structure 204, and a third anatomical structure 206 from one or more anatomical structures, processor 102A extracts a subset of the visual features. In the example, processor 102A can be configured to execute an algorithm for extracting a subset of visual and localization features. Subsequently, processor 102A can be configured to register the visual features together with the first anatomical structure 202, the second anatomical structure 204, and the third anatomical structure 206 to generate one or more discriminative feature vectors. In this embodiment, processor 102A can also be configured to process the visual features into one or more discriminative feature vectors based on the pose and appearance of the target person 108. Moreover, the visual features are processed to be deformation-agnostic in order to form one or more discriminative feature vectors. In other words, processor 102A can be configured to execute algorithms for extracting visual features about the first anatomical structure 202, the second anatomical structure 204, and the third anatomical structure 206. Processor 102A also processes the visual features into one or more discriminative feature vectors robust to the external body shape and posture of the target person 108. In embodiments, processor 102A can also be configured to process the visual features so that they are deformation-agnostic, thereby extracting discriminative features (e.g., features encoding specific organ microstructures, locations). In possible embodiments, the algorithm executed by processor 102A can be a data-driven (e.g., machine learning) model that learns possible deformations of one or more anatomical structures (e.g., organs) to compensate for them, thereby obtaining deformation-agnostic anatomical representations to form one or more discriminative feature vectors.

[0039] Figure 3A and Figure 3B Commonly, these are flowcharts illustrating embodiments of a method for encoding anatomical features into biometric data according to this disclosure. Figure 1A , Figure 1B and Figure 2 Component description Figure 3A and Figure 3B . refer to Figure 3A and Figure 3B The diagram illustrates a flowchart of a method 300 for encoding anatomical features into biometric data. Method 300 includes steps 302 through 330.

[0040] A method 300 is provided for encoding and utilizing anatomical features as biometric data. Method 300 is based on identifying and distinguishing in vivo features, such as organs and bones. In an example, such features can be extracted from radiographic images of a target person 108 for authentication.

[0041] At 302, the processor 102A controls the scanner 104 to scan and capture one or more anatomical images of the body of the target person 108. In an embodiment, one or more anatomical images of the body of the target person 108 captured by the scanner 104 can be viewed on the display unit 118 of the computing device.

[0042] At 304, the control scanner 104 also includes control of at least one of a computed tomography (CT) scanner, magnetic resonance imaging (MRI) scanner, positron emission tomography (PET) scanner, ultrasound scanner, single-photon emission computed tomography (SPECT) scanner, or X-ray or other medical imaging modalities. Therefore, the scanner 104 can be configured to capture one or more anatomical images of the body of the target person 108 based on the type of scanner 104. In implementations, one or more anatomical images allow medical personnel to view the interior of the target person 108's body without any risk of exploratory surgery.

[0043] At 306, processor 102A identifies one or more anatomical structures in one or more captured anatomical images. In the example, the identified one or more anatomical structures may be visible on display unit 118 of computing device 102. According to an embodiment, the identified one or more anatomical structures may be at least one of a single organ, a group of organs, a single bone, a group of bones, the liver, the spleen, or a combination of body organs and bones. Moreover, processor 102A is configured to extract anatomical features from the identified one or more anatomical structures.

[0044] At 308, the processor 102A extracts anatomical features of each identified anatomical structure from a plurality of anatomical structures. In the example, the extracted anatomical features correspond to heartbeat, liver condition (e.g., stretching / compression or affected), bone marrow shape and size, etc. However, the extracted anatomical features may correspond to other features without limiting the scope of the invention.

[0045] At 310, the processor 102A registers the extracted anatomical features of one or more identified anatomical structures to the pose and appearance of the target person 108. Moreover, the extracted anatomical features can be registered to the pose and appearance of the target person 108 in a way that is unique to the target person 108.

[0046] At 312, the processor 102A combines the extracted anatomical features into one or more discriminative feature vectors. In this embodiment, the extracted anatomical features are obtained from multiple scans of the target person 108. Furthermore, the processor 102A can be configured to detect and extract any prominent anatomical structures from the multiple scans. Subsequently, the processor 102A can be configured to combine all the extracted anatomical features into one or more discriminative feature vectors of the target person 108.

[0047] At 314, processor 102A encodes and utilizes one or more discriminative feature vectors as biometric data. In an embodiment, processor 102A may be configured to use an encoding algorithm to encode one or more discriminative feature vectors as biometric data.

[0048] In 316, method 300 further includes combining extracted anatomical features from one or more identified anatomical structures into biometric data. Processor 102A can be configured to combine extracted anatomical features from one or more identified anatomical structures into biometric data.

[0049] In step 318, method 300 further includes assigning a unique label to the biometric data, and the unique label indicating a target person 108. Processor 102A can be configured to assign a unique label to the biometric data.

[0050] In method 300, method 320 further includes storing the biometric data together with a unique tag in memory 110. Processor 102A can be configured to store the biometric data together with the unique tag in memory 110.

[0051] In step 322, method 300 further includes combining the extracted anatomical features with other biometric data of the target person 108. Processor 102A can be used to combine the extracted anatomical features with other biometric data of the target person 108.

[0052] In step 324, method 300 further includes performing a query for one or more discriminative feature vectors against a database 102B that includes multiple pre-stored discriminative feature vectors for authentication of the target person 108. Processor 102A can be configured to perform a query for one or more discriminative feature vectors against database 102B.

[0053] At 326, at least two discriminative feature vectors in database 102B are matched by comparing subsets of visual and location features belonging to one or more anatomical structures encoded as biometric data for authentication of target person 108. Processor 102A can be configured to match at least two discriminative feature vectors in database 102B.

[0054] At 328, the identified one or more anatomical structures are processed to extract visual features, which are further registered together in a database to generate one or more discriminative feature vectors. Processor 102A can be configured to process the identified one or more anatomical structures to extract visual features.

[0055] At 330, visual features are processed into one or more discriminative feature vectors based on the pose and appearance of the target person 108. Furthermore, the visual features are processed to be deformation-agnostic in order to form one or more discriminative feature vectors. Processor 102A can be configured to process visual features into one or more discriminative feature vectors based on the pose and appearance of the target person 108.

[0056] Method 300 can be used for non-invasive biometric authentication based on imaging of in vivo organs. Method 300 can be used to extract and encode robust and unique anatomical features to obtain and utilize biometric data, which is beneficial for improving the persistence and non-evasion factors of the non-invasive biometric system 100A. Because the biometric data is based on anatomical features, which are unique to each target individual, the biometric data cannot be tampered with, for example, even through plastic surgery or other means.

[0057] Steps 302 and 330 are merely illustrative, and other alternatives may be provided without departing from the scope of the claims herein, wherein one or more steps are added, one or more steps are removed, or one or more steps are provided in a different order.

[0058] Modifications to the embodiments of the present disclosure described above are possible without departing from the scope of the disclosure as defined by the appended claims. Expressions such as “comprising,” “including,” “incorporated,” “having,” and “are” used to describe and claim this disclosure are intended to be interpreted in a non-exclusive manner, allowing for the presence of additional items, parts, or elements not explicitly described. Singular references are also interpreted to refer to the plural. The word “exemplary” is used herein to mean “serving as an example, instance, or illustration.” Any embodiment described as “exemplary” is not necessarily to be construed as preferred or advantageous over other embodiments and / or to exclude the incorporation of features from other embodiments. The word “optionally” is used herein to mean “provided in some embodiments and not in others.” It should be understood that certain features of the present disclosure described in the context of a single embodiment for clarity may also be provided in combination in a single embodiment. Conversely, various features of the present disclosure described in the context of a single embodiment for brevity may also be provided individually or in any suitable combination or as suitably as in any other embodiment described in this disclosure.

Claims

1. A method for non-invasive biometric identification, comprising: The processor controls the scanner to scan and capture one or more anatomical images of the target person's body; The processor identifies one or more anatomical structures in the captured one or more anatomical images; The processor extracts the anatomical features of each identified anatomical structure from the plurality of anatomical structures; The processor registers the extracted anatomical features of one or more identified anatomical structures to the pose and appearance of the target person; as well as The processor encodes and utilizes the extracted anatomical features as biometric data; The extracted anatomical features are combined into one or more discriminative feature vectors, wherein the identified one or more anatomical structures are processed to extract visual features, and based on the target person’s posture and appearance, possible deformations of one or more anatomical structures are determined to compensate for the visual features through the possible deformations, thereby obtaining the one or more discriminative feature vectors with unknown deformations. At least two discriminative feature vectors in a database are matched by comparing subsets of visual and location features belonging to one or more anatomical structures encoded into the biometric data for the authentication of the target person.

2. The method according to claim 1, wherein, Controlling the scanner also includes controlling at least one of a computed tomography (CT) scanner, a magnetic resonance imaging (MRI) scanner, a positron emission tomography (PET) scanner, an ultrasound scanner, a single-photon emission computed tomography (SPECT) scanner, or an X-ray or other medical imaging modality.

3. The method according to claim 1, further comprising: Query one or more of the authentication feature vectors against a database containing multiple pre-stored authentication feature vectors for the authentication of the target person.

4. The method according to claim 3, further comprising: The visual features are further registered together in the database to generate the one or more discriminative feature vectors.

5. The method according to claim 1, further comprising: The extracted anatomical features of one or more identified anatomical structures are combined to form the biometric data.

6. The method according to claim 1, further comprising: A unique label is assigned to the biometric data, wherein the unique label indicates the target person.

7. The method according to claim 6, further comprising: The biometric data, along with the unique tag, is stored in the memory.

8. A computer program product comprising instructions that, when run on a computer, cause the computer to perform the method as described in any one of claims 1-7.

Images