Abstract: A traffic signal phase and timing (SPaT) prediction system is disclosed. The system may include a transceiver configured to receive historical traffic signal information associated with a traffic light and real-time traffic signal information. The system may further include a memory configured to store a training data and a trained machine model. The trained machine model may be trained using the training data that includes the historical traffic signal information. The system may further include a processor configured to execute instructions stored in the trained machine model to predict traffic signal information (e.g., SPaT) associated with a future traffic signal cycle based on the real-time traffic signal information. The processor may further output the traffic signal information associated with the future traffic signal cycle to a vehicle.

Abstract: The present teaching relates to method, system, medium, and implementations for biometric-based authentication. When an input image representing biometric information of a person to be authenticated is received, a feature vector of the input image is generated based on the biometric information captured in the input image via a convolution neural network (CNN) that is realized using one or more linear activation functions. The person is authenticated based on the feature vector in accordance with a biometric-based recognition model.

Abstract: This application provides a wearable device-based physiological detection method and a wearable device. The wearable device includes a PPG sensor and at least one motion sensor. After storing M pieces of PPG data in a ring buffer of the PPG sensor, the wearable device may first determine, based on timestamps corresponding to the M pieces of PPG data, a data collection time period corresponding to the M pieces of PPG data, the wearable device separately reads motion data in the data collection time period from a ring buffer of each motion sensor, and if a quantity of pieces of the motion data is greater than a specific numerical value, determines that the motion data is aligned with the M pieces of PPG data, so as to determine multi-source data in a same time period.

Abstract: Embodiments of this application provide a display method and an electronic device. This method can be applied to a terminal device such as a mobile phone or a tablet computer. After detecting an operation of successfully unlocking or returning to a main interface, the terminal device may determine an eyeball gaze position of a user through a camera module, and then determine whether the user is gazing at a notification bar. When recognizing an action of the user gazing at the notification bar, the terminal device may display a notification interface for displaying a notification message.

Abstract: Embodiments of this application provide a photoplethysmography (PPG) based control method and an electronic device and relate to the field of terminal technologies. The method includes: using, by the electronic device, the plurality of LEDs to emit a light signal corresponding to a first current; using, by the electronic device, at least one of the plurality of PDs to obtain a second current returned by the light signal corresponding to the first current via a human tissue; and reducing, by the electronic device, the first current by using a preset first current step size when the electronic device determines that the second current is greater than a target current; or increasing, by the electronic device, the first current by using the preset first current step size when the electronic device determines that the second current is less than the target current.

Abstract: This application relates to the field of fingerprint recognition, and discloses a fingerprint recognition method and an electronic device. An electronic device obtains a fingerprint image of a user by using a fingerprint recognition module in response to a fingerprint recognition operation of the user; and performs a quality evaluation on the obtained fingerprint image, and repairs and recognizes the fingerprint image based on a quality evaluation result.

Abstract: A terminal state identification method and apparatus are provided, and are applied to the field of terminal technologies. The method includes: determining a first probability value based on acceleration data; determining a second probability value based on the acceleration data; determining a state of a terminal user based on the first probability value and the second probability value, where the state of the terminal user includes a first state, a second state, and a third state, when determining that the terminal user is in the first state, where the first interface includes a first window, and the first window is used to display travel recommendation information for a current location to the user.

Abstract: The present teaching relates to method, system, medium, and implementations for biometric authentication. Rotation covariant convolution kernels at multiple convolution layers are obtained with weights learned via machine learning based on rotation invariant (RI) training data. For an input image with fingerprint information captured therein related to a person to be authenticated, an initial feature map is obtained and then at each convolution layer, a feature map is processed based on the RC convolution kernels for the layer to output a rotation covariant (RC) feature map, with the feature map being either the initial feature map and an output RC feature map from a previous convolution layer. The last convolution layer outputs a rotation invariant (RI) feature vector representing fingerprint features of the person in a rotation invariant manner, which is then used to authenticate the person.

Abstract: A fingerprint recognition method and an electronic device. The electronic device includes a touch screen and a fingerprint sensor. The electronic device displays a first interface and captures first fingerprint information that is input by a user at the first interface. The electronic device determines whether the touch screen is in a first state, and the first state is used to indicate that the touch screen is in a screen protector state. If the touch screen is in the first state, the electronic device restores the first fingerprint information by using a first preset artificial intelligence (AI) restoration model to obtain second fingerprint information. The electronic device determines that the second fingerprint information matches preset fingerprint information and displays a second interface. Quality of a fingerprint image in the second fingerprint information is higher than quality of a fingerprint image in the first fingerprint information.

Abstract: This application provides a riding tool identification method and a device. The method includes: obtaining at least one of an acceleration signal acquired by an acceleration sensor and a magnetometer signal acquired by a magnetometer sensor in an electronic device; identifying a riding tool based on at least one of an acceleration feature and a magnetometer feature, to obtain a riding classification result, where the acceleration feature is obtained based on the acceleration signal, and the magnetometer feature is obtained based on the magnetometer signal; obtaining a voice signal acquired by a microphone in the electronic device, and extracting a voice feature based on the voice signal; recognizing a voice broadcast during ride based on the voice feature, to obtain a voice broadcast recognition result; and determining a category of the riding tool based on the riding classification result and the voice broadcast recognition result.

Abstract: This application relates to the field of fingerprint recognition, and discloses a fingerprint recognition method and an electronic device. An electronic device obtains a fingerprint image of a user by using a fingerprint recognition module in response to a fingerprint recognition operation of the user; and performs a quality evaluation on the obtained fingerprint image, and repairs and recognizes the fingerprint image based on a quality evaluation result.

Abstract: Disclosed are a method and an apparatus for preventing an accidental touch, applied to the field of terminal technologies. The method includes: obtaining a quaternion of a terminal; determining, based on the quaternion of the terminal, first information and attitude angle information of the terminal; detecting motion information of the terminal; detecting ambient light information of the terminal; determining, based on the first information, the attitude angle information, the motion information, and the ambient light information, that the terminal is in a first mode; and entering, by the terminal, a screen-off state to reduce power consumption. In addition, the terminal can be prevented from being awakened accidentally in always on display and lock screen-on states, thereby effectively preventing an accidental touch.

Abstract: Embodiments of this application disclose a heart rate monitoring method and apparatus. A harmonic weight value is calculated according to a PPG signal, the harmonic weight value is then used to determine a confidence of a heart rate, and a heart rate value is output only when the confidence is high. The method includes: determining that a wearer of the electronic device is a living body; obtaining a first PPG signal; determining a first heart rate and a confidence of the first heart rate according to the first PPG signal, where the confidence of the first heart rate is determined by weight coefficients of M harmonic signals; outputting the first heart rate in a case that the confidence of the first heart rate meets a preset confidence condition; and displaying the first heart rate.

Abstract: This application provides a heart rate detection method and an electronic device. Through the solution, when it is detected that a user wearing a smart wearable device is making a first workout, a PPG signal is acquired through a PPG sensor in the smart wearable device; first heart rate data is obtained based on the PPG signal and a first deep sequence neural network model; second heart rate data is obtained based on the PPG signal and a first frequency tracking algorithm model; and the first heart rate data and the second heart rate data are fused to obtain a target heart rate of the user. Because a frequency tracking algorithm can quickly track a heart rate change, thus the solution can compensate for a scenario in which a deep sequence neural network cannot implement timely tracking in the case of a sudden change of heart rate.

Abstract: This application discloses a vital sign detection method and an electronic device. The method includes: obtaining an acceleration signal and PPG signals of a plurality of channels within a preset time period; calculating, in a case that there is at least one correlation coefficient in the correlation coefficient matrix less than or equal to a first threshold, a variance of the PPG signals of each channel in the plurality of channels; performing frequency domain fusion after performing a FFT on the PPG signals of the plurality of channels, to obtain a frequency domain signal of the fused PPG signal; performing the FFT on the acceleration signal to obtain a frequency domain signal of the acceleration signal; inputting the frequency domain signal of the fused PPG signal and the frequency domain signal of the acceleration signal into a deep neural network to obtain a value of a vital sign.

Abstract: A heart rate detection method and an electronic device are provided. The method includes: acquiring a first PPG signal with a first preset duration; determining a plurality of heart rate values based on the first PPG signal; determining a difference of every two adjacent heart rates based on the plurality of heart rate values, the difference of every two adjacent heart rates being used for generating a heart rate difference trajectory graph; acquiring a parameter of a heart rate difference trajectory based on the heart rate difference trajectory graph, the parameter of the heart rate difference trajectory including one or more of the following parameters: an angle, a quantity, an angle standard deviation, and a distance standard deviation; determining a heart rhythm type of the first PPG signal according to the parameter of the heart rate difference trajectory; and displaying the heart rhythm type of the first PPG signal.

Abstract: This application provides a physiological detection signal quality evaluation method, an electronic device, and a storage medium. The method includes: obtaining a data sample set collected by a detection device in different scenarios; selecting sample data of a scenario from the sample data set, and dividing the sample data into physiological data and PPG data; calculating an energy ratio of the physiological data based on the physiological data and the PPG data; separately calculating quality evaluation parameters of sample data of a plurality of scenarios based on energy ratios of the physiological data corresponding to different scenarios; and evaluating, based on the quality evaluation parameters, quality of a physiological detection signal collected by the detection device.

Abstract: Provided is a biometric information imaging module, used for imaging palm veins, and method of using same. The imaging module includes a housing with a cover attached to a top of the housing; a lens assembly with a lens for taking images provided in the housing; a light shielding structure provided around at least part of the lens of the lens assembly for shielding the lens from light reflected from the cover; and an image sensor to convert light focused by the lens assembly to photoelectric conversion to an electrical signal. The lens assembly is attached to a first circuit board and the light shielding structure is connected to a second circuit board in the housing.

Abstract: This application provides a deep learning-based heart rate detection method and a wearable device, and relates to the field of communications technologies. According to this solution, a deep Attention network is trained by using PPG and ACC sensor signals as inputs and using heart rate information and exercise scenario information as model outputs, and heart rate detection is performed by using a model obtained through training. Because motion artifact noise is distributed differently in different scenarios, a nonlinear relationship between a scenario, noise, and a heart rate may be fitted by using a deep Attention network learning mechanism, so that data denoising processing and feature extraction can be adaptively implemented in different scenarios, to achieve effects of signal denoising, signal fusion, and complex scenario identification in a plurality of complex scenarios, thereby canceling interference of motion artifact noise to a PPG signal, and improving precision of heart rate detection.

Abstract: Embodiments of this application provide a method for displaying a two-dimensional code and an electronic device. The method is performed by an electronic device, and includes: obtaining first sensor data, where the first sensor data is collected at a first moment; if it is determined, based on the first sensor data, that a first action of a user corresponding to the first moment is a preset action, determining whether the user performs a wrist turning action within a preset duration range starting from the first moment; and displaying a preset two-dimensional code page when the user performs a wrist turning action within the preset duration range. In the method, a process of presenting a two-dimensional code by the user can be simple and efficient.