Abstract: Methods and systems are provided for using temporal supersampling to increase a displayed resolution associated with peripheral region of a foveated rendering view. A method for enabling reconstitution of higher resolution pixels from a low resolution sampling region for fragment data is provided. The method includes an operation for receiving a fragment from a rasterizer of a GPU and for applying temporal supersampling to the fragment with the low resolution sampling region over a plurality of prior frames to obtain a plurality of color values. The method further includes an operation for reconstituting a plurality of high resolution pixels in a buffer that is based on the plurality of color values obtained via the temporal supersampling. Moreover, the method includes an operation for sending the plurality of high resolution pixels for display.

Abstract: A method of operating a data processor when rendering a sequence of frames to produce a sequence of frames for display is disclosed. Each frame represents a view of a scene of one or more objects. The motion of one or more objects is tracked between frames in the sequence of frames. When data for a new frame in the sequence of frames is to be rendered, for a region of the new frame, the method involves determining, based on the tracked motion of one or more objects between frames in the sequence of frames, an amount of motion within the region between frames. The amount of motion is used to assign the region to a category of no, low, intermediate, or high motion. A rendering technique is selected for rendering the region based upon the category to which the region is assigned.

Abstract: Remote distribution of multiple neural network models to various client devices over a network can be implemented by identifying a native neural network and remotely converting the native neural network to a target neural network based on a given client device operating environment. The native neural network can be configured for execution using efficient parameters, and the target neural network can use less efficient but more precise parameters.

Abstract: Methods, systems, and apparatus, including computer programs encoded on computer storage media, for generating output images. One of the methods includes obtaining data specifying (i) a partitioning of the H by W pixel grid of the output image into K disjoint, interleaved sub-images and (ii) an ordering of the sub-images; and generating intensity values sub-image by sub-image, comprising: for each particular color channel for each particular pixel in each particular sub-image, generating, using a generative neural network, the intensity value for the particular color channel conditioned on intensity values for (i) any pixels that are in sub-images that are before the particular sub-image in the ordering, (ii) any pixels within the particular sub-image that are before the particular pixel in a raster-scan order over the output image, and (iii) the particular pixel for any color channels that are before the particular color channel in a color channel order.

Abstract: One aspect of the current disclosure provides a method of upscaling an image. The method includes: rendering an image, wherein the rendering includes generating color samples of the image at a first resolution and depth samples of the image at a second resolution, which is higher than the first resolution; and upscaling the image to an upscaled image at a third resolution, which is higher than the first resolution, using the color samples and the depth samples.

Abstract: An image processing apparatus including an input unit configured to input image data of a target frame, a first processing unit configured to perform a first processing for output image data of a reference frame related to the target frame, which is stored in the memory, to change positions of pixels in the output image data, an image processing unit configured to perform image processing for the image data of the target frame input by the input unit based on the output image data of the reference frame processed by the first processing unit, and a second processing unit configured to perform a second processing, the second processing being according to the first processing, for the image data of the target frame for which image processing has been performed by the image processing unit and to store the processed image data in the memory as output image data of the target frame.

Abstract: The present technology is generally directed to filtering systems and methods for suppression of reverberation artifacts in ultrasound images. In some embodiments, a method of obtaining a filtered ultrasound image includes taking a first ultrasound image of a target tissue using an applicator. At least a portion of the applicator is moved such that the reverberation artifact ultrasound path length changes relative to the first position of the applicator. A second ultrasound image of the target tissue is then taken. The first and second ultrasound images are synthesized using at least one filtering method. The filtering method attenuates or removes reverberation artifacts in the synthesized ultrasound image.

Abstract: In one aspect, there is provided a method of denoising one or more inspection images comprising a plurality of pixels, comprising: receiving an inspection image generated by an inspection system configured to inspect one or more containers, the inspection image being corrupted by a Poisson-Gaussian noise and a variance of the noise being non-constant in the plurality of pixels, and denoising the received inspection image by applying, to the inspection image, a variance-stabilizing transformation for transforming the variance of the noise into a constant variance in the plurality of pixels, wherein the variance-stabilizing transformation is based on a descriptor associated with the angular divergence of the inspection radiation and the Poisson-Gaussian noise.

Abstract: A virtual reality apparatus includes a virtual reality display screen. Further, the virtual reality apparatus includes an eye tracking system that tracks a gaze direction of one or more eyes of a user to determine an object of focus in a virtual reality display. In addition, the virtual reality apparatus includes a processor. The virtual reality apparatus also includes a memory having a set of instructions that when executed by the processor causes the virtual reality apparatus to estimate one or more pupil dimensions of the one or more eyes based upon a luminance of virtual reality imagery displayed by the virtual reality display. The virtual reality apparatus is further caused to determine a focal plane based upon the estimated one or more pupil dimensions.

Abstract: A method is provided to perform local de-noising of an image. The method includes obtaining a region of interest and a region of noise within a scan. The method also includes determining, for a first image based on the region of interest and a second image based on the region of noise, sample blocks and atoms for each image, where each atom contributes to a weighted sum that approximates a sample block in the image. The method also includes determining a measure of similarity of each atom from the first image with atoms from the second image and removing an atom from the first image if the measure of similarity exceeds a predetermined threshold value. The method also includes reconstructing a de-noised image based on atoms remaining in the first image after removing the atom from the first image, and presenting the de-noised image on a display device.

Abstract: An image forming apparatus includes an image forming device to form a correction image, an image density detector to detect image densities of a plurality of areas in the correction image, and circuitry to correct an image formation condition of the image forming device based on detected image densities of the plurality of areas. The circuitry replaces a detected image density of an area of interest selected from the plurality of areas with an average value of detected image densities of two or more areas including adjacent areas adjacent to the area of interest and corrects the image formation condition of the image forming device based on the detected image densities of the plurality of areas after replacement when a difference between the detected image density of the area of interest and at least one of the detected image densities of the adjacent areas exceeds a predetermined threshold.

Abstract: In an embodiment, an image processing apparatus for performing geometrical transformation of data stored in accordance with two-dimensional coordinate positions includes: a control unit configured to decompose a transformation processing from first coordinates to second coordinates into linear transformation and non-linear transformation, the control unit also being configured to determine transformation parameters for the linear transformation and the non-linear transformation; a linear transformation unit configured to perform the linear transformation of the first coordinates to calculate third coordinates; a non-linear transformation unit configured to perform the non-linear transformation of the first coordinates to calculate fourth coordinates; and a combining unit configured to combine the third coordinates and the fourth coordinates to calculate the second coordinates.

Abstract: A method for controlling brightness of an image according to the present disclosure comprises the steps of: determining a light source area on the basis of a light signal intensity of pixels included in the image; determining a surrounding area of the light source area in a predetermined manner; and compensating the brightness property of at least one of the light source area and the surrounding area on the basis of at least one of the property of the image, the property of a display device, or property of the surrounding environment.

Abstract: Color trim data is used in an approximation function to approximate one or more non-linear transformations of image data in an image processing pipeline. The color trim data is derived in one embodiment through a back projection on a colorist system, and the color trim data is used at the time of rendering an image on a display management system.

Abstract: An apparatus and method for insuring the proper alignment of a detected vein pattern and a projected vein pattern are disclosed. The apparatus enhances the visual appearance of veins so that an error that can lead to improper patient care or injury can be avoided.

Abstract: Provided are a regulation method, terminal equipment, and non-transitory computer-readable storage medium for automatic exposure control (AEC) of a region of interest. In the method, a luminance histogram of each color channel in a region of interest is obtained based on statistics on luminance of a plurality of sub-region blocks in the region of interest; a first luminance of the each color channel is determined according to the luminance histogram of the each color channel in the region of interest and the corresponding relationship between the luminance and the number of the sub-region blocks; a reference luminance is determined based on the first luminance of the each color channel, the reference luminance corresponding to a reference color channel; a target luminance corresponding to the present AEC luminance is obtained; and a luminance regulation is performed on the reference color channel according to the target luminance.

Abstract: A method of evaluating one or more kernels of an ear of maize using digital imagery that includes acquiring a digital image of the one or more kernels of the ear of maize without the use of spatial reference points, processing the digital image to estimate at least one physical property of the one or more kernels of the ear of maize from the digital image, and evaluating the at least one kernel of maize using the estimate of the at least one physical property of the at least one kernel of maize. The method includes using one or more such digital images to estimate yield on a plant, management zone, field, county and country level.

Abstract: Methods and systems for performing active learning for defect classifiers are provided. One system includes one or more computer subsystems configured for performing active learning for training a defect classifier. The active learning includes applying an acquisition function to data points for the specimen. The acquisition function selects one or more of the data points based on uncertainty estimations associated with the data points. The active learning also includes acquiring labels for the selected one or more data points and generating a set of labeled data that includes the selected one or more data points and the acquired labels. The computer subsystem(s) are also configured for training the defect classifier using the set of labeled data. The defect classifier is configured for classifying defects detected on the specimen using the images generated by the imaging subsystem.

Abstract: According to one embodiment, an analysis apparatus includes an acquisition unit and a processor. The acquisition unit acquires first information with a first time length between a first time and a second time. The first information is based on motion of an object person. The processor extracts multiple similarity points from the first information. The multiple similarity points are similar to each other in the first information. The processor calculates a time interval between the similarity points.

Abstract: Methods and systems for inspection of wafers and reticles using designer intent data are provided. One computer-implemented method includes identifying nuisance defects on a wafer based on inspection data produced by inspection of a reticle, which is used to form a pattern on the wafer prior to inspection of the wafer. Another computer-implemented method includes detecting defects on a wafer by analyzing data generated by inspection of the wafer in combination with data representative of a reticle, which includes designations identifying different types of portions of the reticle. An additional computer-implemented method includes determining a property of a manufacturing process used to process a wafer based on defects that alter a characteristic of a device formed on the wafer. Further computer-implemented methods include altering or simulating one or more characteristics of a design of an integrated circuit based on data generated by inspection of a wafer.

Abstract: Disclosed is a measurement processing device including: a processing unit configured to control an imaging device to image a substrate, on which a processing film is removed from the peripheral edge portion, and an enclosure member that surrounds the substrate. A captured image obtained by the imaging device is processed to measure a cut width in which the processing film is absent in the peripheral edge portion of the substrate, and a gap width between a peripheral edge end of the substrate and the enclosure member.

Abstract: An imaging and analysis system for a component of a rotary machine includes an image capture device operable to capture image data from at least one selected type of electromagnetic radiation that is at least one of reflected from and transmitted through the component. The system also includes an image processor configured to generate processed data from the captured image data. The system further includes a control system configured to automatically identify a condition of the component by comparing the processed data to stored reference data. The reference data is stored in a format that enables direct comparison to the processed data.

Abstract: Disclosed is an assessment method for making an assessment of laser welding between first and second cylindrical metal members, wherein the first and second cylindrical metal members are arranged coaxially to define an overlap range where the first and second cylindrical metal members overlap each other; and wherein the laser welding is performed on the overlap range along a circumferential direction of the overlap range. The assessment method includes: during the laser welding, carrying out a measurement of a position of a contour of at least one of the first and second cylindrical metal members; and judging the occurrence or non-occurrence of a position deviation of the at least one of the first and second cylindrical metal members based on a result of the measurement.

Abstract: An item inspection apparatus includes an image obtaining section and a control section. The image obtaining section obtains a captured image of at least a part of an item. The control section determines whether the item is defective by using the captured image of the item. The control section performs a first inspection for at least one of position, shape and size with respect to at least one of a first element and a second element. The first element has a predetermined shape and is formed on the item and the second element includes at least one of an opening, a depression and a through-hole formed in the item. The control section performs a second inspection for at least one of foreign substance adhesion, scratch and surface stain on the item. Thus, the defect inspection may be performed more precisely and effectively.

Abstract: One variation of a method for predicting manufacturing defects includes: accessing a first set of inspection images of a first set of assembly units recorded by an optical inspection station over a first period of time; generating a first set of vectors representing features extracted from the first set of inspection images; grouping neighboring vectors in a multi-dimensional feature space into a set of vector groups; accessing a second inspection image of a second assembly recorded by the optical inspection station at a second time succeeding the first period of time; detecting a second set of features in the second inspection image; generating a second vector representing the second set of features in the multi-dimensional feature space; and, in response to the second vector deviating from the set of vector groups by more than a threshold difference, flagging the second assembly unit.

Abstract: A plant growth measurement and prediction system uses drone-captured images to measure the current growth of particular plant species and/or to predict future growth of the plant species. For example, the system instructs a drone to fly along a flight path and capture images of the land below. The captured images may include both thermographic images and high-resolution images. The system processes the images to create an orthomosaic image of the land, where each pixel in the orthomosaic image is associated with a brightness temperature. The system then uses plant species to brightness temperature mappings and the orthomosaic image to identify current plant growth. The system generates a diagnostic model using the orthomosaic image to then predict future plant growth.

Abstract: A semiconductor defect classification device includes feature extractors that are configured to receive images of semiconductor patterns on a wafer and to extract features of the images from the images, and a classifier that is configured to receive the features of the images and first meta information about the wafer and to use machine learning to classify a defect of the semiconductor patterns associated with the images based on the features of the images and the first meta information.

Abstract: A method of inspecting a display panel, which is capable of substantially minimizing a color difference of a display panel according to a viewing angle, includes: setting a reference area having a center point on a chromaticity diagram; emitting a white light on the display panel; measuring first color coordinates of a target point of the display panel at a first viewing angle; measuring second color coordinates of the target point at a second viewing angle; calculating a difference value between the first color coordinates and the second color coordinates; converting the difference value into target polar coordinates with the center point as an origin; and determining whether the target polar coordinates are within the reference area.

Abstract: The present disclosure provides a method and a computer-readable storage medium for color quality assessment of a batch-sample image based on multispectral imaging. The method comprises: selecting an operation area on a reference image and locating a corresponding operation area on a batch-sample image, wherein the reference image is used for color comparison with the batch-sample image to determine whether the batch-sample image satisfies a pass requirement; generating and comparing the color-information items obtained from the reference image and from the batch-sample image respectively and determine whether the batch-sample image satisfies the pass requirement based on one or more thresholds. This method can be applied to automatically and precisely assess the quality of mixed-color fabrics in textile industry, which is more objective, reliable, and cost-effective, as compared to conventional methods for color quality assessment.

Abstract: Aspects of the invention generally relate to illumination gas imaging and detection. Camera systems can illuminate a target scene with light sources configured to emit absorbing and non-absorbing wavelengths with respect to a target gas. An image of the target scene illuminated with a non-absorbing wavelength can be compared to a non-illuminated image of the target scene in order to determine information about the background of the target scene. If sufficient light of the non-absorbing wavelength is scattered by the scene toward a detector, the target scene comprises an adequate background for performing a gas imaging process. A camera system can alert a user of portions of the target scene suitable or unsuitable for performing a gas imaging process. If necessary, the user can reposition the system until sufficient portions of the target scene are recognized as suitable for performing the gas imaging process.

Abstract: Disclosed are systems, devices, and methods for detecting characteristics of an unhatched egg. A set of images of an unhatched egg are obtained, where each of the spectral images is obtained in a particular wavelength range. The set of images is processed to extract image features, where the image features includes an image texture feature. The extracted image features are processed to classify the unhatched egg according to at least one characteristic. The at least one characteristic may include fertility and/or gender.

Abstract: Neural network classification may be performed by inputting a training data set into each of a plurality of first neural networks, the training data set including a plurality of samples, obtaining a plurality of output value sets from the plurality of first neural networks, each output value set including a plurality of output values corresponding to one of the plurality of samples, each output value being output from a corresponding first neural network in response to the inputting of one of the samples of the training data set, inputting the plurality of output value sets into a second neural network, and training the second neural network to output an expected result corresponding to each sample in response to the inputting of a corresponding output value set.

Abstract: To provide a technique of obtaining information for discriminating two substances included in a radiation image of a subject without performing imaging by changing radiation energy, a radiation imaging system includes a detector including a plurality of pixels which obtain pixel values corresponding to incident radiation transmitted through a subject, and an information processing unit configured to perform a process of estimating information on thicknesses and densities of substances included in the subject by performing a calculation using pixel values of an arbitrary one of the plurality of pixels and an average value of energy of radiation quanta of the arbitrary pixel calculated in accordance with the pixel values of the arbitrary pixel.

Abstract: A system and method includes generation of one or more motion-corrupted images based on each of a plurality of reference images, and training of a regression network to determine a motion score, where training of the regression network includes input of a generated motion-corrupted image to the regression network, reception of a first motion score output by the regression network in response to the input image, and determination of a loss by comparison of the first motion score to a target motion score, the target motion score calculated based on the input motion-corrupted image and a reference image based on which the motion-corrupted image was generated.

Abstract: The disclosure relates to stent detection and shadow detection in the context of intravascular data sets obtained using a probe such as, for example, and optical coherence tomography probe or an intravascular ultrasound probe.

Abstract: Systems and methods for applying an ensemble of segmentations to microscopy images of a tissue sample to determine if the tissue sample is representative of cancerous tissue. The ensemble of segmentations is applied to a plurality of greyscale or color microscopy images to generate a final image level segmentation and a final blob level segmentation. The final image level segmentation and final blob level segmentation are used to calculate a mean nuclear volume to determine if the tissue sample is representative of cancerous tissue.

Abstract: Systems, devices, and methods are described for providing patient anatomy models with indications of model accuracy included with the model. Accuracy is determined, for example, by analyzing gradients at tissue boundaries or by analyzing tissue surface curvature in a three-dimensional anatomy model. The determined accuracy is graphically provided to an operator along with the patient model. The overlaid accuracy indications facilitate the operator's understanding of the model, for example by showing areas of the model that may deviate from the modeled patient's actual anatomy.

Abstract: Automated detection of septal defects in cardiac computed tomography images is provided. In various embodiments, a blood pool is extracted from a CTA image of a heart. The blood pool is determined to be a left pool. A septal curve of the left pool is extracted. A plurality of features of the septal curve is determined. The heart is classified as having an atrial septal defect, ventricular septal defect, or no defect based on the plurality of features.

Abstract: Various embodiments provide a cardiac mapping method including: recording premature ventricular contraction (PVC) electrocardiogram (ECG) data during PVC of a heart; generating a PVC activation map of the heart, based on the PVC ECG data and a three-dimensional (3D) heart model generated based on two-dimensional (2D) images of the heart, the PVC activation map including an area of earliest activation; generating a reference image by modifying on one of the 2D images to identify the area of earliest activation; displaying both the PVC activation map and the reference image; and pacing the heart at a first pacing location disposed in the area of earliest activation.

Abstract: A method may include identifying a simulated three-dimensional representation corresponding to an internal anatomy of a subject based on a match between a computed two-dimensional image corresponding to the simulated three-dimensional representation and a two-dimensional image depicting the internal anatomy of the subject. Simulations of the electrical activities measured by a recording device with standard lead placement and nonstandard lead placement may be computed based on the simulated three-dimensional representation. A clinical electrogram and/or a clinical vectorgram for the subject may be corrected based on a difference between the simulations of electrical activities to account for deviations arising from patient-specific lead placement as well as variations in subject anatomy and pathophysiology.

Abstract: A system for processing a seller provided input image depicting an object. The system may process the input images to be in compliance with an online shopping website's image requirements. The image processing service may execute multiple modules on the input image. An edge detection module may detect one or more edges of the image of the object and determine border data indicative of a border around the object. An object of interest module may extract the image of the object within the border to generate an object of interest image. A background extraction module may create one or more closed shapes to separate the image of the object from background pixels. A hole detection module may traverse the one or more closed shapes to determine a presence of background pixels, remove the background pixels, and generate a resulting image.

Abstract: An image capture apparatus 1 is provided with a skin score calculation processing unit 53. The skin score calculation processing unit 53 segments a face region of a person included in an image into a plurality of segmented regions having a uniform size in a predetermined proportion to a size of the face region. Further, the skin score calculation processing unit 53 measures luminance values of a predetermined number of pixels, respectively. Further, the skin score calculation processing unit 53 determines a condition of at least any one of pores and freckles in the face region, based on a distribution condition of the luminance values measured in the segmented region, for each of the segmented regions segmented.

Abstract: In one embodiment, a method includes a computing system accessing a training image. The system may generate a feature map for the training image using a first neural network. The system may identify a region of interest in the feature map and generate a regional feature map for the region of interest based on sampling locations defined by a sampling region. The sampling region and the region of interest may correspond to the same region in the feature map. The system may generate an instance segmentation mask associated with the region of interest by processing the regional feature map using a second neural network. The second neural network may be trained using the instance segmentation mask. Once trained, the second neural network is configured to generate instance segmentation masks for object instances depicted in images.

Abstract: A method and electronic device for generating an index of segments of a polygon is disclosed. The method comprises segmenting a reference zone, which covers at least a portion of a map enclosing all segments of the polygon, into first level zones. Responsive to at least one segment being at least partially located within more than one first level zones, the method comprises indexing the at least one segment in association with the reference zone. The method also comprises, until a terminal condition is met, iteratively: (i) segmenting a given zone into subsequent level zones, where the given zone is a parent zone to the subsequent level zones, and (ii) responsive to at least one other segment being at least partially located within more than one subsequent level zones, indexing the at least one other segment in association with the given zone.

Abstract: Provided in accordance with the present disclosure are systems for identifying a position of target tissue relative to surgical tools using a structured light detector. An exemplary system includes antennas configured to interact with a marker placed proximate target tissue inside a patient's body, a structured light pattern source, a structured light detector, a display device, and a computing device configured to receive data from the antennas indicating interacting with the marker, determine a distance between the antennas and the marker, cause the structured light pattern source to project and detect a pattern onto the antennas. The instructions may further cause the computing device to determine, a pose of the antennas, determine, based on the determined distance between the antennas and the marker, and the determined pose of the antennas, a position of the marker relative to the antennas, and display the position of the marker relative to the antennas.

Abstract: It is determined, when a region of a moving object detected from an image overlaps a fixed region in the image, whether to superimpose a first mask image to be drawn in the region of the moving object on a second mask image to be drawn in the fixed region, based on an image feature amount of the region of the moving object. The drawing of the first mask image is controlled in accordance with this determination result.

Abstract: A system for detecting a motion event based on images. The system acquires a first image and a comparison image, and generates a first edge image and a comparison edge image based on the first image and the comparison image. The system also determines the difference between the first edge image and the comparison edge image. The system indicates a motion event if the difference between the first edge image and the comparison edge image meets or exceeds a threshold.

Abstract: An information processing apparatus including a memory, and a processor coupled to the memory and the processor configured to execute a process, the process including extracting a moving object area corresponding to a moving object included in a plurality of frames in a moving image based on a result of a comparison between the plurality of frames with each other, tracking a position of the moving object area among the plurality of frames, determining, for each of the plurality of frames, an image area excluding the moving object area to be a background area whose position moves more than a defined period of time, and updating a background image of the moving image based on a result of the determining.

Abstract: Systems and methods can be used for ultrasound-based geometry determination for cardiac mapping. A patient can be scanned with an ultrasound while wearing body surface electrodes. While the scanning takes place, the location of the ultrasound transducer can be tracked in three-dimensional space. The electrodes can be tracked and located in the same coordinate system as the image volume. Therefore, each electrode's location can be determined relative to the acquired image volume such that corresponding geometry data is generated for the heart and the electrodes.

Abstract: A method of the present disclosure includes performing, by a processing device, a first image registration between a reference image of a patient and a motion image of the patient to perform alignment between the reference image and the motion image, wherein the reference image and the motion image include a target position of the patient. The method further includes performing, by the processing device, a second image registration between the reference image and a motion x-ray image of the patient, via a first digitally reconstructed radiograph (DRR) for the reference image of the patient. The method further includes tracking at least a translational change in the target position based on the first registration and the second registration.