Abstract: The present invention relates to a convolutional artificial neural network-based image and video recognition system, and a method therefor. The recognition system comprises: a mobile device for performing lower layer analysis, transmitting a user question to a server, receiving an answer thereto from the server, and managing the same; the server connected to the mobile device over a network so as to perform data processing in respective neural network layers corresponding to middle and upper layers of a convolutional artificial neural network, and register, analyze, search for, and classify a particular object (image) and video; and a retriever for comparing an FC layer value for artificial neural network processing of an image and video transmitted by the mobile device with an FC layer value for artificial neural network processing of an image and video registered by a content provider.

Abstract: The present invention relates to a convolutional artificial neural network-based image and video recognition system, and a method therefor. The recognition system comprises: a mobile device for performing lower layer analysis, transmitting a user question to a server, receiving an answer thereto from the server, and managing the same; the server connected to the mobile device over a network so as to perform data processing in respective neural network layers corresponding to middle and upper layers of a convolutional artificial neural network, and register, analyze, search for, and classify a particular object (image) and video; and a retriever for comparing an FC layer value for artificial neural network processing of an image and video transmitted by the mobile device with an FC layer value for artificial neural network processing of an image and video registered by a content provider.

Abstract: An apparatus for and a method of analyzing components using microscopic regions. The apparatus includes: an image obtainer obtaining an image of a part of a living body; a color information generator generating color information of an image having a matrix pattern from a size of the microscopic regions and the obtained image; a first data transformer transforming the color information of the image into a product of first and second matrixes; and a component analyzer analyzing components of the obtained image using at least one of the first and second matrixes. When the obtained image is divided into a plurality of microscopic regions, the size of the microscopic regions is determined so that concentrations of components of the part fluctuate in each of the microscopic regions.

Abstract: Disclosed are a droplet ejection device and method using an electrostatic field. The droplet ejection method includes: setting a separate electric field direction in each of a plurality of nozzles; supplying one of ink and ink containing particles to each nozzle; and forming and ejecting a plurality of separate ink droplets. The droplet ejection device includes a deposition part having electrode layers and insulating layers deposited toward a nozzle. Therefore, it is possible to readily perform droplet ejection without a heater or diaphragm vibration device. In addition, it is possible to reduce impact applied to the ink and obtain good print quality, since the ink is ejected using the electrostatic field.

Abstract: In a method of determining a regression model, and a method of predicting a component concentration of a test mixture using the regression model, the method of determining the regression model includes selecting a regression model, generating a plurality of observation points, each one of the plurality of observation points having a prediction target value and a measured value, determining a weight of the prediction target value and the measured value at each one of the plurality of observation points, and obtaining a calculation amount by reflecting the weights in differences between predicted values acquired by applying the measured values to the regression model and the prediction target values, and determining coefficients of the regression model to minimize the calculation amount.

Abstract: In a method of estimating a component concentration of a mixture, and an apparatus for performing the method, the method includes generating a global calibration model with respect to a calibration data set using a concentration value determined by a plurality of independent variables including a predetermined specific component, a concentration of which is intended to be estimated, as a dependent variable, dividing the calibration data set into at least two small groups according to a value of the dependent variable and generating local calibration models for each of the at least two small groups using the calibration data set included in the divided at least two small groups, and determining a small group to which a spectrum of the mixture belongs and estimating a concentration of the specific component using a local calibration model of the determined small group.

Abstract: In an apparatus and method of noninvasively measuring a concentration of a blood component, the method includes (a) varying a thickness of a body part of a subject, measuring absorption spectrums at different thicknesses of the body part, obtaining a first differential absorption spectrum between the absorption spectrums measured at different thicknesses, actually measuring concentrations of the blood component, and establishing a statistical model using the first differential absorption spectrum and the actually measured concentrations; and (b) estimating the concentration of the blood component using a second differential absorption spectrum obtained with respect to the body part based on the statistical model.

Abstract: Disclosed are a droplet ejection device and method using an electrostatic field. The droplet ejection method includes: setting a separate electric field direction in each of a plurality of nozzles; supplying one of ink and ink containing particles to each nozzle; and forming and ejecting a plurality of separate ink droplets. The droplet ejection device includes a deposition part having electrode layers and insulating layers deposited toward a nozzle. Therefore, it is possible to readily perform droplet ejection without a heater or diaphragm vibration device. In addition, it is possible to reduce impact applied to the ink and obtain good print quality, since the ink is ejected using the electrostatic field.

Abstract: A method of measuring a concentration of a component in a subject includes setting an intensity relationship equation between a positive-order beam and a negative-order beam with respect to a reference matter at a particular wavelength, applying a light having a first wavelength band absorbed by the component and detecting an intensity of a positive-order beam output from the subject and an intensity of a negative-order beam output from the reference matter, the positive-order beam and the negative-order beam having a second wavelength band, calculating an intensity of a positive-order beam input to the subject by applying the intensity of the negative-order beam output from the reference matter to the intensity relationship equation, and calculating absorbance using the intensity of the positive-order beam output from the subject and the intensity of the positive-order beam input to the subject and measuring a concentration of the component using the absorbance.

Abstract: An apparatus for and a method of analyzing components using microscopic regions. The apparatus includes: an image obtainer obtaining an image of a part of a living body; a color information generator generating color information of an image having a matrix pattern from a size of the microscopic regions and the obtained image; a first data transformer transforming the color information of the image into a product of first and second matrixes; and a component analyzer analyzing components of the obtained image using at least one of the first and second matrixes. When the obtained image is divided into a plurality of microscopic regions, the size of the microscopic regions is determined so that concentrations of components of the part fluctuate in each of the microscopic regions.

Abstract: In a method of determining a regression model, and a method of predicting a component concentration of a test mixture using the regression model, the method of determining the regression model includes selecting a regression model, generating a plurality of observation points, each one of the plurality of observation points having a prediction target value and a measured value, determining a weight of the prediction target value and the measured value at each one of the plurality of observation points, and obtaining a calculation amount by reflecting the weights in differences between predicted values acquired by applying the measured values to the regression model and the prediction target values, and determining coefficients of the regression model to minimize the calculation amount.

Abstract: In a mobile terminal having a skin permeation device, and an apparatus and a method for analyzing a body fluid using the same, the skin permeation device includes a control unit for outputting an oscillation start signal in response to an input signal, an oscillating unit for receiving the oscillation start signal and for outputting an oscillation signal having a predetermined frequency, and an output unit for outputting an ultrasonic signal to a diagnosis region of a user in response to the oscillation signal, the output unit being operable to perform skin permeation.

Abstract: In a method of estimating a component concentration of a mixture, and an apparatus for performing the method, the method includes generating a global calibration model with respect to a calibration data set using a concentration value determined by a plurality of independent variables including a predetermined specific component, a concentration of which is intended to be estimated, as a dependent variable, dividing the calibration data set into at least two small groups according to a value of the dependent variable and generating local calibration models for each of the at least two small groups using the calibration data set included in the divided at least two small groups, and determining a small group to which a spectrum of the mixture belongs and estimating a concentration of the specific component using a local calibration model of the determined small group.

Abstract: In an apparatus and method of noninvasively measuring a concentration of a blood component, the method includes (a) varying a thickness of a body part of a subject, measuring absorption spectrums at different thicknesses of the body part, obtaining a first differential absorption spectrum between the absorption spectrums measured at different thicknesses, actually measuring concentrations of the blood component, and establishing a statistical model using the first differential absorption spectrum and the actually measured concentrations; and (b) estimating the concentration of the blood component using a second differential absorption spectrum obtained with respect to the body part based on the statistical model.

Abstract: A method of measuring a concentration of a component in a subject includes setting an intensity relationship equation between a positive-order beam and a negative-order beam with respect to a reference matter at a particular wavelength, applying a light having a first wavelength band absorbed by the component and detecting an intensity of a positive-order beam output from the subject and an intensity of a negative-order beam output from the reference matter, the positive-order beam and the negative-order beam having a second wavelength band, calculating an intensity of a positive-order beam input to the subject by applying the intensity of the negative-order beam output from the reference matter to the intensity relationship equation, and calculating absorbance using the intensity of the positive-order beam output from the subject and the intensity of the positive-order beam input to the subject and measuring a concentration of the component using the absorbance.