Abstract:
An aspect of the disclosure relates to a training device including a memory storing a program, and at least one processor configured to execute the program stored in the memory, in which the processor is configured to acquire pulse wave data to which biological reaction information is imparted, extract a local maximum point of a baseline or a local minimum point of a baseline derived from the pulse wave data as an identification reference point and set a correct answer label for the identification reference point based on the biological reaction information, set an analysis window for the extracted identification reference point and determine a feature vector of the identification reference point in the analysis window, and train a discriminator that identifies a cyclic alternating pattern (CAP) indicating a periodic brain wave activity by training data including the feature vector and the correct answer label.
Abstract:
A bioinformation acquiring apparatus includes at least one processor; and a memory configured to store a program to be executed in the processor. The processor acquires bioinformation in a chronological order; derives outlier level parameters, the outlier level parameter indicating a level of inclusion of outliers of the bioinformation in pieces of bioinformation acquired within a first duration; derives correction terms based on the bioinformation after removal of the outliers of the bioinformation from pieces of bioinformation acquired within a second duration that is longer than the first duration; selects one or both of a first correction procedure and a second correction procedure based on the outlier level parameters, as a correction procedure, the first correction procedure using the correction terms, the second correction procedure involving interpolation irrelevant to the correction terms; and corrects the outliers of the bioinformation within the first duration by the selected correction procedure.
Abstract:
An acceleration acquirer of a controller of an in-bed and out-of-bed determination device acquires, in a time series, acceleration of a subject from a sensor that is an acceleration sensor attached by the subject, a body motion determiner determines, based on the acquired acceleration, whether there is a body motion of the subject at each time, an evaluator evaluates a distribution of the acceleration at a time at which it is determined that there is not a body motion, an estimator estimates a body axis of the subject based on the evaluated distribution of the acceleration, and an in-bed and out-of-bed determiner determines at least one of in bed and out of bed of the subject based on the estimated body axis of the subject that is estimated by the estimator.
Abstract:
A voice input unit has predetermined directivity for acquiring a voice. A sound source arrival direction estimation unit operating as a first direction detection unit detects a first direction, which is an arrival direction of a signal voice of a predetermined target, from the acquired voice. Moreover, a sound source arrival direction estimation unit operating as a second direction detection unit detects a second direction, which is an arrival direction of a noise voice, from the acquired voice. A sound source separation unit, a sound volume calculation unit, and a detection unit having an S/N ratio calculation unit detect a sound source separation direction or a sound source separation position, based on the first direction and the second direction.
Abstract:
A feature extractor extracts feature quantities from a digitized speech signal and outputs the feature quantities to a likelihood calculator. A distance determiner determines the distance between a user providing speech and a speech input unit. The likelihood calculator selects registered expressions for speech recognition from a recognition target table based on the determined distance, to be used in calculation of likelihoods at the likelihood calculator. The likelihood calculator calculates likelihoods for the selected registered expressions based on the feature quantities extracted by the feature extractor, and outputs one of the registered expressions having the maximum likelihood as a result of speech recognition.
Abstract:
A speech determiner determines whether or not a target individual is speaking when facial images of the target individual are captured. An emotion estimator estimates the emotion of the target individual using the facial images of the target individual, on the basis of the determination results of the speech determiner.
Abstract:
An imaging apparatus generates a 3D model using a photographed image of a subject and generates a 3D image based on the 3D model. When a corresponding point corresponding to a point forming the 3D model does not form a 3D model generated using a photographed image photographed at a different photographing position, the imaging apparatus determines that the point is noise, and removes the point determined as noise from the 3D model. The imaging apparatus generates a 3D image based on the 3D model from which the point determined as noise is removed.