Sony Corporation

Omron Corporation

Sakon Yamamoto

Tetsujiro Kondo

Naoki Tsuchiya

Kennosuke Hayashi

Yoshihisa Ijiri

Yohei Okawa

Yu Maruyama

Tsutomu Watanabe

Masaaki Hattori

Hiroto Kimura

Takeharu Nishikata

Shizuo Chikaoka

Akihiko Arimitsu

Masashi Kurita

Yutaka Kato

Yasuyuki Ikeda

Nobuyuki Asakura

Takuo Morimura

Kei Hiraizumi

Yuki Hanzawa

Hirosuke Nagano

Shigenori Nagae

Yuki Hasegawa

An inspection apparatus according to one or more embodiments extracts an attention area from a target image using a first estimation model, performs a computational process with a second estimation model using the extracted attention area, and determines whether a target product has a defect based on a computational result from the second estimation model. The first estimation model is generated based on multiple first training images of defect-free products in a target environment. The second estimation model is generated based on multiple second training images of defects. The computational process with the second estimation model includes generating multiple feature maps with different dimensions by projecting the target image into different spaces with lower dimensions. The extracted attention area is integrated into at least one of the multiple feature maps in the computational process with the second estimation model.

Inspection apparatus, unit selection apparatus, inspection method, and computer-readable storage medium storing an inspection program

An image determination device includes feature extractors which output, on the basis of an image to be examined, each piece of feature data indicating a specific feature of the image; a first training part which trains a first determiner so as to output first output data indicating first label data associated with a first training image on the basis of first feature data output when the first training image is input to the feature extractors; a second training part which trains a second determiner so as to output second output data indicating second label data associated with a second training image on the basis of second feature data output when the second training image is input to the feature extractors; and an output part which outputs, on the basis of the first output data and the second output data, output data indicating an overall determination result about an image.

Image determination device, image determination method, and non-transitory computer readable medium storing program

Provided is an image determination device. The image determination device is provided with: feature extractors which output, on the basis of an image to be examined, each piece of feature data indicating a specific feature of the image; a determiner which outputs, on the basis of the feature data output from the extractors, output data indicating the determination result pertaining to the image; and a training part which trains the determiner so as to output, output data indicating the label data associated with the training image on the basis of the feature data output when the training image is input to the extractors, wherein the training part further trains, by using new training data, the determiner so that the output data indicating the label data associated with the image is output by the determiner.

Image determination device, training method and non-transitory computer readable medium storing program

An image determination device includes: a training model which outputs, on the basis of an image to be examined, output data indicating a determination result about the image; a training part which trains the training model to output, by using training data including a training image and label data, output data indicating the label data associated with the training image, when the training image is input to the training model; a dividing part which divides the training data into a plurality of pieces of sub-training data; a measurement part which measures accuracy of determination when the training part trains the training model by using each of the plurality of pieces of sub-training data; and selection part which selects at least any one among the plurality of pieces of sub-training data on the basis of the accuracy of determination.

This image processing device, by superimposing a defect imageon a background image, with the original image of an inspection subject as the background image, and performing image processing in which the display format of the defect imageor the display format of the background imageis altered, a plurality of composite imageshaving different ways of seeing the defect imagewith respect to the background imageare generated (step), the advisability of detecting the defect imagefrom each of the plurality of composite imagesis verified (step), the detectable rangeof the defect imageis estimated on the basis of the detection advisability verification results (step), and the detectable rangeis displayed (step).

Image processing device, image processing method, and image processing non-transitory computer readable medium for verifying detectable range of defect image

An apparatus of processing an image contains a scene-change-detecting unit that detects a scene change of input image based on an image signal relative to the input image, a display-position-controlling unit that determines a display position of an image after the scene change to a display position associated with an image before the scene change based on a detection result in the scene-change-detecting unit, and a display-position-setting unit that sets the display position determined in the display-position-controlling unit as a display position in which the image after the scene change is displayed. The display-position-controlling unit detects a feature value of the images before and after the scene change and determines the display position of the image after the scene change based on the feature value of the images.

Apparatus and method of processing image as well as apparatus and method of generating reproduction information with display position control using eye direction

The present invention relates to an image processing device that can reproduce a state in which, using a first display device, an image is displayed on a second display device having a characteristic different from that of the first display device, an image processing method, and a program. A motion detecting unitdetects motion of an image from an input image signal. A sub-field developing unitdevelops the input image signal over sub-fields. A light-intensity integrating unitcalculates, in a simulated manner, the light intensity integrated on the retinas of a person at the time the person sees the input image signal, from the direction of the motion of the image and a light emitting pattern of the sub-fields, and displays the image with the calculated light intensity on the first display device. The present invention is applicable to the case where, for example, using an LCD, a state in which an image is displayed on a PDP having a characteristic different from that of the LCD is reproduced.

Image processing device, image processing method, and program

An image signal processing apparatus includes: a block division unit inputting an image signal encoded by orthogonal transformation and dividing the image signal into second blocks different from first blocks encoded using the orthogonal transformation; an activity calculation unit calculating an activity of each second block; a re-encoding unit encoding the second blocks using the orthogonal transformation; a re-encoded difference calculation unit calculating difference values between the second blocks and the first blocks; a class determination unit determining a class of each second block on the basis of a distribution of the difference value and the activity; a prediction coefficient generating unit generating a prediction coefficient corresponding to each class; a prediction tap construction unit constructing a prediction tap for calculating a target pixel in each second block; and a predictive operation unit operating the prediction tap and the prediction coefficient to calculate a pixel value of each target pixel.

Image signal processing apparatus, image signal processing method, prediction coefficient generating apparatus, prediction coefficient generating method, and programs for causing computer to execute the methods

An image processing apparatus includes an input unit configured to input a characteristic parameter indicating the characteristics of a filter process; a detection unit configured to detect, on the basis of the characteristic parameter input by the input unit, a phase shift amount between the image-captured signal and an obtained signal; a first forming unit configured to form a first prediction tap composed of a plurality of obtained pixels used to predict a target image-captured pixel value; a coefficient obtaining unit configured to obtain a first prediction coefficient generated in accordance with the characteristic parameter and the phase shift amount in order to predict the target image-captured pixel value by product-sum computation with the value of the first prediction tap; and a first computation unit configured to generate a first output signal corresponding to the image-captured signal by performing product-sum computation between the first prediction coefficient.

Image processing apparatus, image-capturing apparatus, image processing method, and program

A signal processing system, a signal processing apparatus and method, a recording medium, and a program in which high frequency components of a narrow-range signal having suppressed high frequency components can be restored with high accuracy. A narrow-range audio signal having low frequency components is divided into frames by a frame-unit dividing unit, and is spectral-transformed into a signal on a frequency domain by a DCT transform unit. A learning-produced-data table extracts, at an address corresponding to a class code supplied from a classifying unit, high-frequency-component DCT spectrum data. In a spectrum combining unit, the extracted high-frequency-component DCT spectrum data is combined with low-frequency-component DCT spectrum data output from the DCT transform unit. Subsequently, a broad-range audio signal having combined high frequency components and low frequency components is output.

Signal processing system, signal processing apparatus and method, recording medium, and program

In encoding input image data including first and second images, the first image is divided into blocks. A second block corresponding to a first block that is one of the blocks is detected from the second image. Pixels forming the first and second blocks are sorted in order according to pixel values. Block difference data indicating the amount of change in the pixel values of the sorted pixels between the first and second blocks is determined and encoded. In decoding, block information identifying the second block, sort-order information indicating the sorted pixels forming the first block, and the encoded block difference data are obtained. The encoded block difference data is decoded. The decoded block difference data and the second block identified by the block information are used to generate pixel values of the first block. The generated pixel values are reverse-sorted according to the sort-order information.

Encoding apparatus and method, decoding apparatus and method, image processing system and method, and recording medium

An image processing system includes a coding device configured to code input image data, and a decoding device configured to decode the coded input image data, wherein, if coding and decoding are repeated on the input image data, the image quality of an image corresponding to the input image data is deteriorated. The decoding device includes an input section inputting quantized data in which the input image data is quantized; and a decoding section dequantizing the quantized data that is input by the input section and converting a dequantized value obtained as a result of the quantization into a value in the vicinity of the boundary of a quantization threshold value within a range corresponding to the dequantized value when the quantization is performed.

Systems and methods for image processing coding/decoding

The present invention relates to a signal processing system, a signal processing apparatus and method, a recording medium, and a program in which high frequency components of a narrow-range signal having suppressed high frequency components can be restored with high accuracy. A narrow-range audio signal having low frequency components is divided into frames by a frame-unit dividing unit, and is spectral-transformed into a signal on a frequency domain by a DCT transform unit. A learning-produced-data tableextracts, at an address corresponding to a class code supplied from a classifying unit, high-frequency-component DCT spectrum data. In a spectrum combining unit, the extracted high-frequency-component DCT spectrum data is combined with low-frequency-component DCT spectrum data output from the DCT transform unit. Subsequently, a broad-range audio signal having combined high frequency components and low frequency components is output. The present invention can be applied to cellular phones.

In an information processing apparatus, for example, a video tape recorder, a recording section records an input signal on a recording medium. Substantially at the same time, a playback section plays back the input signal from the recording medium to output a playback signal. A learning section generates prediction coefficients based on characteristics of the playback signal and the original input signal. Then, when the playback section plays back the input signal recorded on the recording medium to output the playback signal, a classification-adaptation section corrects the playback signal based on the prediction coefficients generated by the learning section, outputting a corrected playback signal.

Information processing apparatus and method, recording medium, and program

The present invention relates to a signal processing apparatus and method, a recording medium, and a program, in which a more preferable class can be easily selected. A region extracting unitextracts a class tap from an input composite video signal, and pattern detecting unitstodetect patterns for classes by using different predetermined methods. A class-code determining unitdetermines a class based on information from an evaluation table, and a coefficient memorysupplies a predictive coefficient corresponding to the class to a predictive computing unit. The predictive computing unitperforms predictive computing based on a predictive tap supplied from a region extracting unitand the predictive coefficient supplied from the coefficient memory, so that a component Y signal is generated and is output therefrom. The present invention can be applied to a television receiver.

Growing community of inventors

Tokyo, Japan

Sakon Yamamoto