The technology relates to methods and systems for recognition of conditions from ventilation data. The methods may include acquiring ventilation data for ventilation of a patient during a time period; generating an image based on the acquired ventilation data; providing, as input into a trained machine learning model, the generated image, wherein the trained machine learning model was trained based on images having a same type as the generated image; and based on output from the trained machine learning model, generating a predicted condition of the patient. The image may be generated by storing ventilatory data as pixel channel values to generate a human-indecipherable image.

A method and system for establishing an association between a document and a person or verifying identity comprising capturing images of the person's face and processing the images to generate a face feature vector data. The face feature vector data can not be reverse processed to generate the image of the person's face. The method also captures images of the graphic on the item and processes those images to extract graphic derived feature vector data that is stored inside the graphic code. The graphic extracted face feature vector data cannot be reverse processed to generate the image of the person's face. Then, the face derived feature vector data is compared to the graphic derived feature vector data to develop a similarity value. Then the similarity value is compared to a first threshold value to develop a validation indicator representing the likelihood that the graphic/item are associated with the person.

Methods, systems, and computer-readable storage mediums for detecting facial liveliness are provided. Implementations include actions of processing first and second facial images of a subject to determine first and second corneal reflections of an object, the first and second facial images being captured at first and second sequential time points, determining a corneal reflection change of the object based on the determined first and second corneal reflections, comparing the determined corneal reflection change of the object to a motion associated with the first and second time points, and determining facial liveliness of the subject based on a result of the comparison.

An electronic device for de-identifying dynamic biometric data of a user and an operation method thereof are disclosed. The method includes the operations of obtaining dynamic biometric data regarding a movement of a user; de-identifying the dynamic biometric data by distorting the dynamic biometric data; and transmitting the de-identified dynamic biometric data to a server.

The disclosed embodiments include computerized methods and systems that facilitate two-factor authentication of a user based on a user-defined image and information identifying portions of the image sequentially selected by the user. In one aspect, a communications device presents a first digital image of a first user on a touchscreen display. The communications device may receive, from the first user, information identifying portions of the first digital image selected in accordance with a candidate authentication sequence established by the first user. The selected first image portions may, for example, be associated with corresponding facial features of the first user. The communications device may determine whether the candidate authentication sequence matches a reference authentication sequence associated with the first digital image, and may authenticate an identity of the first user, when the first selection sequence is determined to match the second selection sequence.

Embodiments of the invention provide methods and apparatus for monitoring the routing configuration within an electronic device such that a biometric authentication process can be carried out without interference from other components of the device, such as may occur when the device has become infected with malware for example. The invention may provide a codec or speaker recognition processor, coupled to receive biometric input data, comprising a security module that determines whether a routing configuration complies with one or more rules. The security module may be implemented to prevent genuine biometric data from being output from the speaker recognition processor, and/or to prevent spoof biometric data from being inserted into the authentication module.

A method includes extracting a set of enrollment feature points from an enrollment biometric measurement. The method also includes randomly selecting one or more enrollment code words from an error correction code. The method also includes determining obfuscated enrollment feature point data describing an obfuscated version of the set of feature points that is obfuscated using the one or more enrollment code words. The method also includes determining obfuscated enrollment code word data describing an obfuscated version of the one or more enrollment code words that is obfuscated using a random enrollment polynomial. The method also includes determining an enrollment biometric template including the obfuscated enrollment feature point data and the obfuscated enrollment code word data. The method includes generating a public key based on the random enrollment polynomial that obfuscates the random enrollment polynomial. The method also includes determining enrollment data including the enrollment biometric template.

A face recognition based key generation apparatus controls a key generation model that is formed of a CNN and an RNN to be learned to generate a desired key having a consistent value by using sample facial images of a key owner and a PIN of the key owner as inputs, and the key generation model receives a facial image of the key owner and the PIN of the key owner, as inputs at a desired key generation time, and generates a key.

Aspect may relate to a device that comprises a sensor and a first secure processor. The sensor may receive an input and generate raw data from the input. The first secure processor may control a first execution environment to perform operations including receiving the raw data from the sensor. Further, the device may include a second processor to control a second execution environment to perform operations including: receiving the raw data; performing data processing to determine normalized data from the raw data and additional data; performing feature extraction to the normalized data to determine features; and sending the features to the first execution environment. The first execution environment may use the features to match the features with stored reference features to authenticate a user.

Biometric data are obtained from biometric sensors on a stand-alone computing device, which may contain an ASIC, connected to or incorporated within it. The computing device and ASIC, in combination or individually, capture biometric samples, extract biometric features and match them to one or more locally stored, encrypted templates. The biometric matching may be enhanced by the use of an entered PIN. The biometric templates and other sensitive data at rest are encrypted using hardware elements of the computing device and ASIC, and/or a PIN hash. A stored obfuscated PassWord is de-obfuscated and may be released to the authentication mechanism in response to successfully decrypted templates and matching biometric samples. A different de-obfuscated password may be released to authenticate the user to a remote or local computer and to encrypt data in transit. This eliminates the need for the user to remember and enter complex passwords on the device.