Abstract: A non-oriented electrical steel sheet having: low high-frequency iron loss and high magnetic flux density; an inner layer and surface layers provided on both sides of the inner layer, the surface layers and the inner layer having specific chemical compositions; the thickness t of 0.01 mm to 0.35 mm; a multilayer ratio of t1 to t of 0.10 to 0.70, t1 denoting a total thickness of the surface layers; a difference between [Si]1 and [Si]0 of 1.0 mass % to 4.5 mass % or less, [Si]1 denoting a Si content in each of the surface layers and [Si]0 denoting a Si content in the inner layer; and a difference between [Mn]0 and [Mn]1 of 0.01 mass % to 0.40 mass %, [Mn]0 denoting a Mn content at a mid-thickness position t/2 and [Mn]1 denoting an average Mn content in a region from a surface to a position at a depth of ( 1/10)t.

Abstract: When producing a non-oriented electrical steel sheet by subjecting a slab containing predetermined amounts of C, Si, Mn, Al, N, and Cr to hot rolling, hot-band annealing, cold rolling, and finishing annealing in a continuous annealing furnace, the following conditions are used: a maximum reached temperature in the finishing annealing is set to be lower than 900° C.; an average cooling rate from a temperature (maximum reached temperature—50° C.) to 500° C. in a cooling process of the finishing annealing is set to 40° C./s or higher; and a parameter ?/t defined from a plastic elongation ratio ? (%) in a rolling direction between before and after the finishing annealing and a soaking time t (s) in the finishing annealing is set to 0.10 or higher. Thus, a non-oriented electrical steel sheet is obtained.

Abstract: Provided is a non-oriented electrical steel sheet having good fatigue resistance suitable for rotor cores and excellent magnetic properties suitable for stator cores. The non-oriented electrical steel sheet has a chemical composition of C: 0.01% or less, Si: 2.0% to 5.0%, Mn: 0.05% to 5.00%, P: 0.1% or less, S: 0.01% or less, Al: 3.0% or less and N: 0.005% or less, with the balance being Fe and inevitable impurities, where Si+Al is 4.5% or more. For the crystal grains in the steel sheet, the average grain size X1 is 60 ?m to 200 ?m, the standard deviation S1 of the crystal grain size distribution satisfies the specified formula (1), and the skewness ?1 of the crystal grain size distribution is 1.50 or less.

Abstract: Provided are a high-strength non-oriented electrical steel sheet having good fatigue resistance suitable for rotor cores and a non-oriented electrical steel sheet having excellent magnetic properties suitable for stator cores. The non-oriented electrical steel sheet has a chemical composition of C: 0.01% or less, Si: 2.0% to 5.0%, Mn: 0.05% to 5.00%, P: 0.1% or less, S: 0.01% or less, Al: 3.0% or less and N: 0.005% or less, with the balance being Fe and inevitable impurities, where Si+Al is 4.5% or more. For the crystal grains in the steel sheet, the average grain size X1 is 50 ?m or less, the standard deviation S1 of the grain size distribution satisfies the specified formula (1), and the skewness ?1 of the grain size distribution is 2.00 or less.

Abstract: When producing a non-oriented electrical steel sheet by subjecting a slab containing a predetermined composition to hot rolling, hot-band annealing, cold rolling, and finishing annealing in a continuous annealing furnace, the following conditions are used: a maximum reached temperature in the finishing annealing is set to be lower than 900° C.; an average cooling rate from a temperature to 500° C. in a cooling process of the finishing annealing is set to 40° C./s or higher; a parameter ?/t defined from a plastic elongation ratio ? (%) in a rolling direction between before and after the finishing annealing and a soaking time t (s) in the finishing annealing is set to 0.10 or higher; an average ferrite grain size is 50 ?m or larger; and compressive residual stresses ?s and ?c in a sheet width direction at a steel sheet surface and a sheet-thickness center part, respectively, are both 2.0 MPa or higher.

Abstract: A receiver convolves an impulse response for compensating for frequency characteristics and a complex impulse response for wavelength dispersion compensation with each of a real component and an imaginary component of each polarized wave of a reception signal. For each polarized wave, the receiver performs complex signal processing of multiplying the imaginary component by the imaginary unit, then branching the resulting component, and adding one imaginary component to the real component. For each polarized wave, the receiver uses, as input signals, the real component and the imaginary component of each polarized wave after complex signal processing and the phase conjugates of them.

Abstract: An image analysis method according one or more embodiments may analyze a form of a cell using a deep learning algorithm with a structure of a neural network. The image analysis method may include: generating data for analysis from an image for analysis in which an analysis target cell is captured; inputting the data for analysis into the deep learning algorithm; and generating data indicating a form of the analysis target cell using the deep learning algorithm.

Abstract: The present invention provides a coating agent composition capable of making interference fringes inconspicuous and suppressing the generation of cracks during the production of a hard coat layer without blending fine particles of metal oxides such as zirconium oxide and titanium oxide in an optical article in which a hard coat layer or a functional layer including a primer layer and a hard coat layer is formed on a plastic layer having a high refractive index. The coating agent composition comprises a trifunctional silane coupling agent and a bifunctional silane coupling agent in a solid content ratio of 95:5 to 10:90.

Abstract: There is provided the turbulence prediction system or the turbulence prediction method that predicts a zone in a prediction area where the possibility of the occurrence of a turbulence is high at a determination time point.

Abstract: A photoelectric conversion apparatus comprising: a pixel array; a signal line; a processing circuit; a switch for controlling conduction between the signal line and an input node of the processing circuit; and a control unit. The control unit performs a first transition in which, after resetting of the processing circuit, the switch is transitioned to an OFF state and then, during a period after the floating diffusion is reset during which the pixel signals are read out of the pixels into the signal line, the switch is transitioned at least from the OFF state to an ON state; keeps the switch in the OFF state during a period during which the transfer transistor is performing the transfer; and performs a second transition in which, after the transfer, the switch is transitioned from the OFF state to the ON state.

Abstract: A display device according to the present invention, includes: a first substrate including a first single-crystal semiconductor substrate provided with a plurality of light emitting portions and with a first drive circuit that drives the plurality of light emitting portions, wherein the first single-crystal semiconductor substrate includes a plurality of light guiding portions that transmit light so as to implement a see-through function.

Abstract: The present invention provides a composition for forming a hard coat and a laminate using the same. The composition for forming a hard coat includes at least metal oxide fine particles; a hydrolysate of an organosilicon compound; an adhesion promoting component having an alkoxysilyl group; a curing catalyst; and a solvent, wherein by introducing, as the organosilicon compound, an organosilicon compound having a hydrocarbon group having (6 to 18) carbon atoms substituted with a glycidoxy group, and an organosilicon compound having a hydrocarbon group having (1 to 5) carbon atoms substituted with a glycidoxy group, even when applied to a flexible resin substrate, the composition for forming a hard coat has excellent adhesion and scratch resistance, and bending resistance (crack resistance) and surface hardness as well.

Abstract: To shorten the time required to sort target particles with sufficient purity, a particle sorter is provided.

Abstract: An image analysis method according one or more embodiments may analyze a form of a cell using a deep learning algorithm with a structure of a neural network. The image analysis method may include: generating data for analysis from an image for analysis in which an analysis target cell is captured; inputting the data for analysis into the deep learning algorithm; and generating data indicating a form of the analysis target cell using the deep learning algorithm.

Abstract: An image analysis method according one or more embodiments may analyze a form of a cell using a deep learning algorithm with a structure of a neural network. The image analysis method may include: generating data for analysis from an image for analysis in which an analysis target cell is captured; inputting the data for analysis into the deep learning algorithm; and generating data indicating a form of the analysis target cell using the deep learning algorithm.

Abstract: An image sensor with a simple configuration that generates an image to which a filter has been applied is disclosed. An image sensor comprises a plurality of pixels, each pixel including a photoelectric converter that detects incidence of single photons; and a counter that counts pulses present in an output signal from the photoelectric converter. The image sensor outputs a count value as a pixel value, wherein the count value is obtained by counting the pulses present in the output signals from a plurality of the photoelectric converters in a first period, or by counting the pulses present in the output signal from a single photoelectric converter in the first period and a second period.

Abstract: An image sensor with a simple configuration that generates an image to which a filter has been applied is disclosed. An image sensor comprises a plurality of pixels, each pixel including a photoelectric converter that detects incidence of single photons; and a counter that counts pulses present in an output signal from the photoelectric converter. The image sensor outputs a count value as a pixel value, wherein the count value is obtained by counting the pulses present in the output signals from a plurality of the photoelectric converters in a first period, or by counting the pulses present in the output signal from a single photoelectric converter in the first period and a second period.

Abstract: An image analysis method according one or more embodiments may analyze a form of a cell using a deep learning algorithm with a structure of a neural network. The image analysis method may include: generating data for analysis from an image for analysis in which an analysis target cell is captured; inputting the data for analysis into the deep learning algorithm; and generating data indicating a form of the analysis target cell using the deep learning algorithm.

Abstract: An image sensor provided with an imaging unit including a plurality of photoelectric converters, the image sensor comprises: an input unit to which compressed image data is input from the outside of the image sensor; a decompressor for decompressing the compressed image data input from the input unit; and an image processor that applies image processing to image pickup data obtained from the imaging unit and to the image data decompressed by the decompressor.