Various example embodiments are directed towards systems, apparatuses, and/or methods of obtaining a reference image for optical object recognition, the method including driving a subset of light sources of a plurality of light sources, receiving, using an object recognition sensor, light reflected off a first target object, obtaining a first reference image based on the reflected light, obtaining a first target image associated with the first target object based on the reflected light, obtaining at least one first environment information using at least one environmental sensor while driving the subset of light sources, the first environment information associated with a surrounding environment, storing the first reference image and the first environment information together, and obtaining a first effective image for the first target object based on the first target image and the first reference image.

A processor-implemented method includes: obtaining an input embedding vector corresponding to an input fingerprint image for authentication; determining a confidence value of the input embedding vector based on fingerprint data of an initial model including either one or both of a trained real fingerprint determination model and a trained fake fingerprint determination model that are provided in advance; and updating the initial model based on the input embedding vector, in response to the confidence value being greater than or equal to a first threshold.

A method, system, and computer program product for identifying a malicious user obtain a plurality of service requests for a service provided by a processing system, each service request of the plurality of service requests being associated with a requesting user and a requesting system, and a plurality of service responses associated with the plurality of service requests, each service response of the plurality of service responses being associated with the processing system; and identify the requesting user as malicious based on the plurality of service requests and the plurality of service responses.

In one aspect, a method is performed by a computing system that is in communication with a content-presentation device, where the content-presentation device is scheduled to perform, at a modification start-time, a content-modification operation that comprises modifying a modifiable content-segment that the content-presentation device is scheduled to receive. The method includes determining that the content-presentation device has detected, within a predefined time period after the modification start-time and during performance of the content-modification operation, a mismatch between reference fingerprint data representing the modifiable content-segment and query fingerprint data representing content received by the content-presentation device from a tuner device connected to the content-presentation device. The method also includes determining that the mismatch is due to the tuner device having modified the modifiable content-segment before the modification start-time and, in response to determining that the mismatch is due to the tuner device having modified the modifiable content-segment before the modification start-time, performing an action.

Systems and methods for detecting data acquisition conditions using color-based penalties can include a computing device obtaining a sequence of images acquired by a photodetector. The computing device can determine, for each pixel position of a plurality of pixel positions associated with the sequence of images, a respective penalty score indicative of a similarity between a color value of a pixel of the pixel position and a desired color value. The desired color value can represent a color property of light emitted from body parts of users when placed opposite to the photodetector. The computing device can determine, using penalty scores of the plurality of pixel positions, a relative position of a body part of a user with respect to a desired position.

An optical fingerprint sensor with spoof detection includes a plurality of lenses; a pixel array including a plurality of first photodiodes, a line between a center of each first photodiode and an optical center of each lens forms an optical axis; at least one apertured baffle-layer positioned between the image sensor and the plurality of lenses, each having a respective plurality of aperture stops, each aperture stop being center-aligned with the optical axis; and a plurality of second photodiodes intercalated with the plurality of first photodiodes; and a color filter layer between the pixel array and the plurality of lenses, said color filter layer includes a plurality of color filters positioned such that each second photodiode is configured to detect electromagnetic energy having passed through lens, a color filter, and at least one aperture stop not aligned along the optical axis.

Systems and methods for blood pressure estimation using smart offset calibration can include a computing device associating a calibration photoplethysmographic (PPG) signal generated from a first sequence of image frames obtained from a photodetector of the computing device with one or more measurement values generated by a blood pressure measurement device different from the computing device. The computing device can obtain a recording PPG signal generated from a second sequence of image frames obtained from the photodetector, and identify a calibration model from a plurality of blood pressure calibration models based on the calibration PPG signal and the recording PPG signal. The computing device can generate a calibrated blood pressure value using the recording PPG signal, features associated with the calibration PPG signal and the identified calibration model.

A display device includes: a display unit configured to display an image; a fingerprint sensing unit on one surface of the display unit and comprising fingerprint sensing pixels configured to sense a fingerprint; and a readout circuit configured to receive a fingerprint sensing signal from the fingerprint sensing pixels, wherein the readout circuit is configured to generate and output a normal fingerprint image and a dry fingerprint image, which correspond to the fingerprint sensing signal, during a fingerprint registration mode.

Examples of electronic devices are disclosed. An example electronic device includes a first housing including a first inner face, a second housing including a second inner face, and a hinge pivotably connecting the first housing to the second housing for rotation between at least a folded configuration and an open configuration, wherein the first inner face is adjacent the second inner face in the folded configuration. The example electronic device further includes a camera extending from the first inner face, and a cap on the second inner face and in alignment with the camera in the folded configuration, the cap configured to retract from the second inner face in the folded configuration to accommodate extension of the camera from the first inner face.

Methods, systems, and media for secure authentication of users using one or more biometric recognition systems are provided. In some embodiments, the method comprises: receiving an indication that a biometric identifier is to be used to authenticate a user to a service; receiving (i) the biometric identifier of the user from a capture device and (ii) knowledge-based secondary information associated with the user from an input device; determining a Voronoi cell identifier that corresponds to the biometric identifier; calculating a hash of the Voronoi cell identifier and the knowledge-based secondary information; transmitting the hash to a server device for verification; in response to transmitting the hash to the server device, receiving a response indicating whether the hash matches a previously stored hash that was stored in the server device; and determining whether to automatically authenticate the user to the service based on the response from the server device.