Abstract: A server device includes: a reception unit configured to receive a RAW image from a terminal; and an image pipeline processing unit configured to develop the RAW image, wherein the image pipeline processing unit has: a first image processing unit configured to performing one or more image processes including at least pixel value adjustment on the RAW image and outputting an image-processed RAW image; and a second image processing unit configured to perform one or more image processes including at least demosaic on the RAW image image-processed by the first image processing unit and outputting a developed image.

Abstract: Proposed is a technique for providing assistance so that an image is captured with the intended composition. An image processing device according to the present invention includes a correction means capable of carrying out image blur correction, and can output an image generated on the basis of a cut-out region which has been cut out from an image based on imaging performed by an imaging device. The image processing device includes a control means for controlling notification related to a range from which the cut-out region can be cut out while the image blur correction performed by the correction means is activated.

Abstract: A calculation scheme decision system includes a pre-calculation unit performing, in an execution environment in which calculation is performed, calculation for each of respective layers of the network structure using at least one of calculation schemes prepared in advance for the respective layers, a cost acquisition unit acquiring a calculation cost of at least one calculation scheme for each layer based on a result of the calculation by the pre-calculation unit, a decision unit selecting one calculation scheme for each layer based on the calculation cost from among at least one of the calculation schemes prepared in advance for the respective layers to associate the layer with the selected one calculation scheme, and a calculation unit performing the calculation for each of the respective layers of the network structure on input data in the execution environment using the calculation scheme associated with each layer.

Abstract: Examples of an image processing device includes a first camera, a second camera having a different angle of view from the first camera, a processor configured to generate a preview image from an image captured by the first camera and an image captured by the second camera, and a display configured to display the preview image. In a case of a change in camera usage from the first camera to the second camera due to a change in a zoom magnification, a tone of an image that was captured by the second camera and is to be displayed first on the display is brought closer to a tone of an image that was captured by the first camera and was displayed last on the display, in comparison with a case where the captured images are used as-is.

Abstract: Examples of a depth estimation method include acquiring a plurality of depth maps, and outputting one output depth map obtained by compositing the plurality of depth maps with a lower average difference between distance values of adjacent pixels than in a case of directly using distance values included in the plurality of depth maps.

Abstract: An analysis device for visualizing an accuracy of a trained determination device includes an acquisition unit acquiring an image pair of a non-defective product image and a defective product image, an extraction unit extracting an image region of a defective part of the defective product, a generation unit generating a plurality of image regions of pseudo-defective parts, a compositing unit synthesizing each of the image regions of the plurality of pseudo-defective parts with the non-defective product image to generate a plurality of composite images having different feature quantities, an unit outputting the plurality of composite images to the determination device and acquiring a label corresponding to each of the plurality of composite images from the determination device, and a display control unit displaying an object indicating the label corresponding to each of the plurality of composite images in an array based on the feature quantities.

Abstract: A training support device includes a derivation unit that derives a feature quantity of the training data, on the basis of a model trained using the training data, and training data having first data given a first label and second data given a second label, and derives, a feature quantity of the training candidate data, on the basis of at least one piece of training candidate data given any one of the first label and the second label, and the model, a calculation unit that calculates, at least one of a distance between the training candidate data and the first data and a second distance between the training candidate data and the second data, and a selection unit that selects data to be added as the training data from among the pieces of training candidate data, on the basis of the distance.

Abstract: The image generation apparatus includes: an acquisition unit configured to acquire a base image, a reference image, the respective exposure values associated with the base image and the reference image, and a pseudo-exposure value different from the respective exposure values associated with the base image and the reference image; an estimation unit configured to estimate a conversion parameter; and a generation unit configured to generate a pseudo-image regarded as an image photographed with the pseudo-exposure value by converting the luminance value of the base image for each of the pixels based on the respective exposure values associated with the base image and the reference image, the pseudo-exposure value, and the conversion parameter.

Abstract: An image identification apparatus includes an acquisition unit configured to acquire an input image, a classification unit configured to calculate a classification score of the input image based on a neural network having a weight coefficient in each layer determined to calculate the classification score indicating a degree of similarity between a training image and an image to be processed, an anomaly determination unit configured to calculate an anomaly score of the input image based on a function approximator constructed by machine learning based on the training image of correct answer, and an identification unit configured to classify the input image into a good image having a high degree of similarity to the training image of the correct answer or a bad image having low degree of similarity to the training image of correct answer based on the classification score and the anomaly score.

Abstract: A training system comprises a training device, and a training data creation device for the training device. The training device trains a neural network by means of a backpropagation algorithm. The training data creation device acquires any one of a positive evaluation indicating that content of the input data coincides with the label, a negative evaluation indicating that content of the input data does not coincide with the label, and an ignorable evaluation indicating exclusion from a training target label, for each label regarding the input data to create training data.

Abstract: A device includes an acquiring unit acquiring a plurality of images including a first image, a second image, and a third image; a region detecting unit comparing pixel values of the first image and the third image to detect a moving object region; and a compositing unit generating a composite image by synthesizing a target image selected from a plurality of images and the second image wherein the compositing unit performs a ghost removing process for pixels of the target image corresponding to the moving object region and pixels of the second image corresponding to the moving object region to bring the pixel values of one of these images closer to the pixel values of the other image, and synthesizes the target image and the second image after the ghost removing process.

Abstract: Disclosed is a method of authorizing a user for accessing a server and/or for receiving of an on-line service and the steps of: capturing biometric data of the user using the sensor on a ME; forming from the biometric data a biometric template on the IDS and storing the biometric template on the MED; and via the IDS allowing access to a server by the user providing to the IDS, via the MED, matching biometric data and a biometric template. On the MED, a local check can be made for a match between biometric data of the user that are captured using the sensor on the MED and biometric data read out of the memory.

Abstract: A device includes an acquiring unit acquiring a plurality of images including a first image, a second image, and a third image; a region detecting unit comparing pixel values of the first image and the third image to detect a moving object region; and a compositing unit generating a composite image by synthesizing a target image selected from a plurality of images and the second image wherein the compositing unit performs a ghost removing process for pixels of the target image corresponding to the moving object region and pixels of the second image corresponding to the moving object region to bring the pixel values of one of these images closer to the pixel values of the other image, and synthesizes the target image and the second image after the ghost removing process.

Abstract: A training system comprises a training device, and a training data creation device. The training device trains a neural network. The training data creation device acquires any one of a positive evaluation indicating that content of the input data coincides with the label, a negative evaluation indicating that content of the input data does not coincide with the label, and an ignorable evaluation indicating exclusion from a training target label, for each label regarding the input data to create training data. In training the neural network for training, the training system adjusts the weight coefficient for the intermediate layer and makes the recognition score of the label with the ignorable evaluation not affect the adjustment of the weight coefficient for the intermediate layer.

Abstract: An image processing apparatus includes: an alignment unit configured to perform alignment for a group of images including a plurality of short-time-exposure images that are captured with a first exposure time and a long-time-exposure image that is captured with a second exposure time that is longer than the first exposure time; and an output unit configured to output the long-time-exposure image in a case where the long-time-exposure image satisfies a predetermined condition and configured to output a composite image which is generated by compositing a plurality of images selected from among the plurality of short-time-exposure images, in a case where the long-time-exposure image does not satisfy the predetermined condition.

Abstract: An image processing apparatus for obtaining an output image by performing a filter processing of an input image, includes a setting unit configured to set a plurality of local regions in the input image, a deriving unit configured to derive a plurality of coefficients corresponding to each of the plurality of local regions based on statistics indicating variations in pixel values in the plurality of local regions, and a filter processing unit configured to obtain an output pixel value by performing a liner transformation of the input image based on the plurality of the coefficients, wherein the plurality of local regions include a region that is partitioned diagonally to a raster scan direction; wherein the filter processing unit performs the liner transformation of the input image based on a coefficient adaptively selected from the plurality of coefficients.

Abstract: Identification structure (1) with a portrait image (3) of a holder, the identification structure having a substrate (8,50), a laser sensitive layer (11,55) and a lens array (12,56) overlying the laser sensitive layer (11,55), the substrate including a printed pattern (9,54), the laser sensitive layer including a laser engraved image pattern (15,15?; 16,16?). The laser engraved image pattern forms a three dimensional portrait image (A,B) of the holder and the printed pattern forms a two-dimensional color pattern (C) of the portrait image of the holder, wherein the two-dimensional color pattern (C) and the three-dimensional portrait image (A,B) are adapted to be observed in a superimposed manner.

Abstract: A calculation scheme decision system includes a pre-calculation unit performing, in an execution environment in which calculation is performed, calculation for each of respective layers of the network structure using at least one of calculation schemes prepared in advance for the respective layers, a cost acquisition unit acquiring a calculation cost of at least one calculation scheme for each layer based on a result of the calculation by the pre-calculation unit, a decision unit selecting one calculation scheme for each layer based on the calculation cost from among at least one of the calculation schemes prepared in advance for the respective layers to associate the layer with the selected one calculation scheme, and a calculation unit performing the calculation for each of the respective layers of the network structure on input data in the execution environment using the calculation scheme associated with each layer.

Abstract: One aspect of the present disclosure is an image processing device performing image stabilization of frame images shot by an imaging device. This device is provided with an acquiring unit, an image generating unit and a cropping unit. The acquiring unit acquires the plurality of frame images. The image generating unit stitches, among the plurality of frame images acquired by the acquiring unit, a first frame image and a second frame image, the second frame image being one frame image among several frames before and after the first frame image, to generate a composite image larger than a size of the first frame image and the second frame image. The cropping unit sets a cropping region to the composite image and outputs an image in the cropping region as an output frame image.

Abstract: Disclosed is a security document including an image layer with in a first image area an identification image. The document has in a second image area an authentication image that is composed of a first authentication image that is a positive version of the first image and a second authentication image that is a negative version of the first image. The authentication images can be observed at different angles and form a tilt image. Mutual comparison of the positive and negative tilt images and comparison with the identification image provides improved detection of unauthorized alterations.