Abstract: A feature enhancement and data augmentation method for detecting at least one action of at least one tested subject is provided. The feature enhancement and data augmentation method includes obtaining at least one spectrogram; and performing a feature enhancement processing on the at least one spectrogram, to enhance at least one feature corresponding to at least one action in the at least one spectrogram and generate at least one feature enhanced spectrogram. The feature enhancement processing comprises a directional filtering.

Abstract: A wearable device with heart rate detection includes a heart rate detection assembly and a processor. The heart rate detection assembly is configured to obtain a plurality of dynamic heart rate signals according to a default frequency. The processor includes a finite state machine and a storage. The storage is configured to store the plurality of dynamic heart rate signals, and the finite state machine determines a heart rate state according to a sampling parameter and the plurality of dynamic heart rate signals. The processor obtains a weight according to the heart rate state, and obtains a dynamic heart rate value according to the weight.

Abstract: Abstract of Disclosure A feature enhancement and data augmentation method for detecting at least one action of at least one tested subject is provided. The feature enhancement and data augmentation method includes obtaining at least one spectrogram; and performing a feature enhancement processing on the at least one spectrogram, to enhance at least one feature corresponding to at least one action in the at least one spectrogram and generate at least one feature enhanced spectrogram. The feature enhancement processing comprises a directional filtering.

Abstract: A method for calculating blood oxygen saturation includes defining the extreme value point of a non-red light signal as the extreme value point of a red light signal, and calculating the blood oxygen saturation according to the extreme value point of the red light signal. The method for calculating blood oxygen saturation of the present disclosure proposes to use other light source signals to assist in finding the period of the red light signal, so that the calculation result of the red light signal is more accurate.

Abstract: The present invention provides a signal quality detection method for an ear-chip physiological measurement device. The signal quality detection method includes receiving a sensing signal from the ear-clip physiological measurement device; filtering the sensing signal to generate a pre-processed signal; calculating a physiological index according to the pre-processed signal; and calculating a similarity of a red light alternating current (AC) component and an infrared light AC component of the pre-processed signal and a plurality of correlation coefficients of the red light AC component, and generating a reliability index of the physiological index accordingly. The reliability index indicates one of a plurality of signal qualities.

Abstract: An object detection method, an electronic apparatus and an object detection system are provided. The method is adapted to the electronic apparatus and includes the following steps. A first image is obtained. A geometric transformation operation is performed on the first image to obtain at least one second image. The first image and the at least one second image are combined to generate a combination image. The combination image including the first image and the at least one second image is inputted into a trained deep learning model to detect a target object.

Abstract: An object detection method, an electronic apparatus and an object detection system are provided. The method is adapted to the electronic apparatus and includes the following steps. A first image is obtained. A geometric transformation operation is performed on the first image to obtain at least one second image. The first image and the at least one second image are combined to generate a combination image. The combination image including the first image and the at least one second image is inputted into a trained deep learning model to detect a target object.