Mitsubishi Electric Corporation

Daiki Ata

Satoshi Namematsu

Kazuhiko Nakamura

Daisuke Ammi

Keiichi Yamada

Tatsumi Yabusaki

Tomo Ikuyama

Yuji Hattori

A learning device and other techniques allow accurate diagnosis of a production facility. A learning device () includes a data acquirer that acquires data for learning, and a model generator that generates a learning model for inferring a condition of a workpiece () handled in a production facility () on the basis of the data for learning. The data for learning includes setting data indicating a setting of the production facility (), image data indicating an image of the production facility () captured by a camera (), temperature data indicating a surface temperature of the production facility () measured by a temperature sensor (), distance data indicating a distance from a range sensor () to the production facility () measured by the range sensor (), and condition data indicating the condition of the workpiece () handled in the production facility ().

Learning device, diagnostic system, and model generation method to diagnose abnormality based on temperature measurement in a production facility

An article image data acquirer acquires article image data as a target of optical character recognition (OCR). An inference result data generator generates first inference result data and second inference result data by inputting the article image data as the target of the OCR into a trained model. An inference result data outputter outputs the first inference result data and the second inference result data. An image filter generator generates a first image filter based on the first inference result data and a second image filter based on the second inference result data. An image filter outputter outputs the first image filter and the second image filter.

Inference device, inference method, and recording medium

A training data acquirer acquires training data including article image data, image-filter-related data indicating a combination of a plurality of image filters used for image processing of the article image data and a value of a parameter for each of the plurality of image filters, and optical character recognition (OCR) score data indicating a score of character recognition output through OCR when image processing is performed on the article image data using the image filters based on the image-filter-related data. A trained model generator generates a trained model indicating a relationship between the article image data, the image-filter-related data, and the OCR score data through machine learning using the training data.

Image-filter generation system, image-filter generation device, learning device, and learning method

A parameter adjustment device () adjusts a parameter relating to control of laser light emitted from a measurement sensor () onto an object. A parameter calculator () calculates the parameter by applying, to a trained model generated through machine learning using training data sets each including waveform data of an amount of light received by the measurement sensor () and data indicating the parameter used to acquire the waveform data, waveform data newly acquired in a new state. The parameter calculated by the parameter calculator () enables measurement of the object using the measurement sensor () in the new state. A parameter outputter () outputs data indicating the parameter calculated by the parameter calculator ().

Parameter adjustment device, training device, and measurement system

A setpoint adjustment apparatus for a displacement meter () includes a determiner () to determine whether a measurement value acquired by an acquirer () in measurement of a reference workpiece using an applying setpoint, to be used in measurement of the reference workpiece, is within the range of a desired measurement value (), and a changer () to change the applying setpoint. When the measurement value is within the range of the desired measurement value (), the applying setpoint used in acquisition of the measurement value is employed as an applying setpoint for inspection of a measurement target (). When the measurement value is out of this range, the applying setpoint used in acquisition of the measurement value is changed to a different applying setpoint, and whether the measurement value from the reference workpiece using this applying setpoint is within the range of the desired measurement value () is determined.

Setting value adjustment device for displacement meter

A filtering device () includes an acquirer () and a filtering unit (). The acquirer () acquires an input value that is repeatedly input. The filtering unit () obtains an output value by filtering for reduction of noise included in the input value. When a difference between the current input value and the past output value exceeds a first threshold, the filtering unit () obtains a new output value by weighting the current input value with a weight greater than a weight for obtaining the past output value.

Filtering device, sensor device, filtering method and program

A peak location calculation device includes a peak value detection block that detects a plurality of candidates for a peak value of intensity of laser light received by an imaging unit consisting of an imaging element formed of a plurality of pixels, and a peak location calculation block that calculates an approximation function that approximates an intensity distribution of the laser light using the plurality of candidates for peak values detected by the peak value detection block. The peak location calculation block calculates the approximation function that minimizes an error between intensity values of the plurality of candidates for the peak value and values of the approximation function.

Display device

A signal processing device for performing processing for reducing noise in a displacement amount measured on a basis of light reflected from a detection object includes a moving-average calculating unit that performs moving average calculation for an input signal to reduce a noise component included in the input signal and an infinite impulse response filter that reduces a noise component included in an input signal by digital signal processing. A filter coefficient of the infinite impulse response filter is determined on a basis of a difference between a first calculation result output by the moving-average calculating unit and a second calculation result output by the infinite impulse response filter when a same signal is input to the moving-average calculating unit and the infinite impulse response filter.

Growing community of inventors

Tokyo, Japan

Daiki Ata