Abstract: The present invention is directed to a method of detecting stray fibers emanating from a substrate comprising: a) Taking an image or a series of consecutive images of stray fibers in front of a contrasted background, that is created using one or more light sources, using a digital device; b) Segmenting of stray fibers and background from objects not directly contrasted against the background resulting in a binary image mask; c) Applying an edge detection technique to the image or modification of original image to enhance contrast of stray fibers against the background; and Applying the binary image mask from step b) to an image created in step c) to create an image in which only the stray fibers and/or outline of any other object contrasted against the background are present.

Abstract: A computer extracts a plurality of candidate lines that are observed from a position of an imaging device that captures an image of an object from among a plurality of candidate lines included in shape information of the object. The computer generates plural pieces of association information indicating a prescribed number of combinations obtained by respectively associating the prescribed number of candidate lines of the observed plurality of candidate lines with the prescribed number of feature lines of a plurality of feature lines detected from the image. The computer determines an association result according to respective errors of the plural pieces of association information.

Abstract: Techniques to improve edge detection in images. Some techniques include logic to process image data into patched image data in accordance with a colorspace model where the patched image data includes color data in a plurality of patches and identify an image group corresponding to the patched image data. The logic may be further configured to select, based upon the image group, a colorspace transform mechanism being operative to transform the image data into transformed image data in accordance with another colorspace model, the other colorspace model having a higher likelihood than the colorspace model at edge detection for the image group. The logic may be further configured to apply the colorspace transform mechanism to the image data to generate the transformed image data in accordance with the other colorspace mode and then, apply an edge detection technique to the transformed image data. Other embodiments are described and claimed.

Abstract: A learning method for segmenting an image having one or more lanes is provided to be used for supporting collaboration with HD maps required to satisfy level 4 of autonomous vehicles. The learning method includes steps of: a learning device instructing a CNN module (a) to apply convolution operations to the image, thereby generating a feature map, and apply deconvolution operations thereto, thereby generating segmentation scores of each of pixels on the image; (b) to apply Softmax operations to the segmentation scores, thereby generating Softmax scores; and (c) to (I) apply multinomial logistic loss operations and pixel embedding operations to the Softmax scores, thereby generating Softmax losses and embedding losses, where the embedding losses is used to increase inter-lane differences among averages of the segmentation scores and decrease intra-lane variances among the segmentation scores, in learning parameters of the CNN module, and (II) backpropagate the Softmax and the embedding losses.

Abstract: An apparatus and method for detecting a foreground in an image is provided, and the foreground detecting apparatus includes a context information estimator configured to estimate context information on a scene from an image frame of the image, a background model constructor configured to construct a background model of the image frame using the estimated context information, and a foreground detector configured to detect a foreground from the image frame based on the constructed background model.

Abstract: A bed area extraction method includes specifying, by a computer, a first part of a bed candidate area in an image based on a first line segment extracted from the image; setting a search range for searching a second line segment with reference to the bed candidate area which is determined based on the specified first part of the bed candidate area; and correcting the bed candidate area based on a disposition of the second line segment included in the set search range.

Abstract: Apparatus and methods for display control and image processing that enable easy and noninvasive observation of an object are described. Display control of a cell image may include generating a motion amount for subregions of the cell image and displaying the motion amount superimposed on the cell image. Image processing of a cell image may include generating a motion amount for subregions of a cell image and evaluating a state of propagation of the motion of the cells.

Abstract: The present disclosure relates to a data processing apparatus and a data processing method capable of generating evaluation index data for performing accurate and detailed evaluation of cultured cardiomyocytes. A motion detecting unit divides frame image data obtained by photographing the cultured cardiomyocytes for a predetermined time into blocks and obtains motion detection data in units of blocks per each frame period. A feature amount calculating unit calculates a feature amount for each block at the same position in a frame image using the motion detection data. A classification processing unit classifies each of the blocks into any one of a plurality of classification categories using the calculated feature amount. On the basis of the classification result, evaluation index data made of individual classification result data that represent correspondences between the blocks and the classification categories is generated.

Abstract: Provided is a method to easily detect eye motions in the eye moving direction, which has resistance to a fluctuation in background light and to a displacement of the position of a camera. The method includes the steps of: detecting a white area of the eye from the eye area; determining a blood vessel image to be used for a template blood vessel image in the white area of the eye; detecting a nth blood vessel image, which matches the template blood vessel image; calculating a nth moving amount of the eye based on a difference between the reference position and the detection position of the nth blood vessel image, which matches the template blood vessel image; and calculating a nth rotation angle of the eye to detect eye motions in the information processing apparatus.

Abstract: A head mounted display comprises an eye tracking system configured to enable eye tracking using polarization. The eye tracking system includes one or more illumination sources and an optical detector comprising polarization sensitive pixels. The one or more illumination sources are configured to illuminate a user's eye and generate reflections directed towards the optical detector. The eye tracking system determines, for each polarization sensitive pixel in a subset of the polarization sensitive pixels, one or more estimation parameters. The eye tracking system determines, for the subset of the polarization sensitive pixels, depth information for one or more glints associated with one or more surfaces of the eye, based in part on the polarization of the reflections and the one or more estimation parameters. The determined depth information is used to update a model of the eye. The eye tracking system determines eye tracking information based on the updated model.

Abstract: Embodiments of the present invention are directed to methods and a mechanism for manipulating images generated by radiotherapy machines used in radiation diagnostic and treatment applications. In one embodiment, a method is provided for intelligent automatic propagation of manual or automatic contouring across linked (e.g., registered) images and image data sets by acquiring one or more images of one or more image data sets; determining the correlation between the images with respect to identified structures; and generating a deformation map that establishes a correspondence for each point in the source image with a point in the target image. Subsequently, the intelligent propagation mechanism applies this deformation map individually to each structure of the source image and propagates the deformed structure to the target image.

Abstract: Systems and methods for co-registering medical images obtained with different imaging modalities are provided. For instance, images obtained with x-ray imaging, such as x-ray computed tomography (“CT”), can be co-registered with images obtained with magnetic resonance imaging (“MRI”). The different imaging modalities generate images that have different visualization characteristics for tissues; thus, in general, co-registration is accomplished by identifying different anatomical features in the different images and then utilizing a known spatial relationship between those anatomical features to co-register the different images.

Abstract: A registration technique is provided that can combine one or more related registrations to enhance accuracy of a registration of image volumes. A registration relationship between a first source volume and a target volume and a registration relationship between the first source volume and a second source volume are concatenated to provide an estimate of a registration relationship between the second source volume and the target volume. The estimate is utilized to inform the direct registration of the second source volume to the target volume or utilized in place of the direct registration.

Abstract: There is provided in accordance with an aspect of the presently disclosed subject matter, a method operating a depth sensing device. The method includes obtaining a current pulse timing on each of a first and a second depth sensing devices; and allocating an exclusive pulse timeslot to each of said first and second devices, so that during a portion of a frame which corresponds to any one of the pulse timeslots only a respective one of said first device and said second device emits a pulse.

Abstract: One embodiment of the present invention sets forth a technique for estimating a head pose of a user. The technique includes acquiring depth data associated with a head of the user and initializing each particle included in a set of particles with a different candidate head pose. The technique further includes performing one or more optimization passes that include performing at least one iterative closest point (ICP) iteration for each particle and performing at least one particle swarm optimization (PSO) iteration. Each ICP iteration includes rendering the three-dimensional reference model based on the candidate head pose associated with the particle and comparing the three-dimensional reference model to the depth data. Each PSO iteration comprises updating a global best head pose associated with the set of particles and modifying at least one candidate head pose. The technique further includes modifying a shape of the three-dimensional reference model based on depth data.

Abstract: A method for gathering information about a work environment comprising using two laser rangefinders to obtain distances to two obstacles. A camera captures an image of the area and an image processor extracts the color depth at the two points of the laser rangefinders. The image processor then extracts the color depth of the line connecting the two points of the laser rangefinders. If the color depth of the two points of the laser rangefinders and the line connecting them is within a predetermined range, the system identifies the area as a substantially flat surface. When a substantially flat surface is found, the system can calculate the length and angle of the line connecting the two points of the laser rangefinders.

Abstract: A displacement detecting apparatus includes: a detector which detects displacement, which is spatial displacement over time, of each of a plurality of measurement points which have been set on an object, using a plurality of images of the object captured at a plurality of time points; an extractor which extracts characteristic displacement specific to the object, based on the displacement detected by the detector; and a calculator which calculates overall displacement indicating displacement of the entirety of the object, from the characteristic displacement extracted by the extractor.

Abstract: An image processing method includes acquiring a plurality of image data items by continuously capturing images of an object by an image capturing apparatus (S1), transmitting the plurality of image data items to a computing section (S2), computing image feature values each of which relates to positional information and information correlated with illumination conditions of each of the image data items by using the computing section (S3), comparing degrees of matching between reference image feature values, each of which corresponds to positional information and information correlated with illumination conditions, and the plurality of image data items (S4), and selecting, from the plurality of image data items, an image having positional information and illumination conditions that match the reference image feature values and outputting the image (S5).

Abstract: A system for identifying specific instances of objects in a three-dimensional (3D) scene, comprising: a camera for capturing an image of multiple objects at a site; at least one processor executable to: use a location and orientation of the camera to create a 3D model of the site including multiple instances of objects expected to be in proximity to the camera, and generate multiple candidate clusters each representing a different projection of the 3D model, detect at least two objects in the image, and determine a spatial configuration for each detected object; and match the detected image objects to one of the multiple candidate cluster using the spatial configurations, associate the detected objects with the expected object instances of the matched cluster, and retrieve information of one of the detected objects that is stored with the associated expected object instance; and a head-wearable display configured to display the information.

Abstract: A method for verifying positions of a plurality of monitoring devices includes selecting a second monitoring device by a first monitoring device. A first number of monitoring devices out of the plurality of monitoring devices is supervised, the plurality of monitoring devices being positioned such that each monitoring device of the plurality of monitoring devices can be monitored by at least one other monitoring device of the plurality of monitoring devices. Correct positions of the plurality of monitoring devices have been determined and stored prior to verification, and security information for secure communication has been exchanged between two of the plurality of monitoring devices for all of the plurality of monitoring devices.

Abstract: An image processing device includes an acquisition unit configured to acquire an image data and a depth data correspond to the image data; a calculating unit configured to calculate a position and attitude change for each depth from the image data and the depth data; and a determining unit configured to determine a position and attitude change of a whole image by a position and attitude change data calculated by the calculating unit based on a statistic of the position and attitude change calculated by the calculating unit.

Abstract: An image processing device that detects a marker, in which the marker includes a first figure in which three or more figure centers are overlapped with each other and a data area provided on an internal or external portion of the first figure, and the image processing device detects the first figure and detects the data area based on the figure centers of the first figure.

Abstract: The invention provides an apparatus for determining a bonding position of a die. The apparatus includes a bond head for picking up and bonding the die. The apparatus further includes a plurality of cameras positioned and configured for capturing at least a first image including a first side surface of the die, a second image including a second side surface of the die and a third image including the first and second side surfaces of the die. Further, the invention provides a method of determining a bonding position of a die. The method captures at least a first image including a first side surface of the die, a second image including a second side surface of the die and a third image including the first and second side surfaces of the die. The method further determines a bonding position of the die based on the captured images.

Abstract: A method for fully automated localization and channel identification of electroencephalography (EEG) electrodes. The electrode locations are automatically identified from three dimensional images stored in an electronic format, wherein the images may be derived from magnetic resonance imaging (MRI) that render the electrodes visible and object shapes and properties are used to locate the electrodes in the three dimensional images. The three dimensional images also show the brain in detail, such that the relationship of the electrodes to the brain is available, thereby making it possible to better identify electrical sources within the brain that create the EEG signals.

Abstract: A control system for a hybrid machine is provided. The control system includes a controller communicably coupled to an energy storage unit of the hybrid machine. The controller is configured to receive data from the energy storage unit. The controller is configured to evaluate a current storage charge state of the energy storage unit based on the received data. The controller is configured to receive historical data related to idle events associated with an engine of the hybrid machine. The controller is configured to receive data related to one or more machine operating parameters. The controller is configured to pre-emptively control at least one of an engine speed and an engine power based on the received data for at least one of shutting down the engine during an idle state and restarting the engine.

Abstract: Setting interface element colors is disclosed including rendering an interface element, determining first color values, assigning a value to a color attribute of the interface element based on the first color values, and updating the interface element according to the assigned value.

Abstract: Embodiments can provide a computer-implemented method for 3D motion correction for diffusion weighted imaging images, the method comprising acquiring a series of image slices from a medical imaging device; binning the series of image slices into bins, each bin comprising a plurality of slice locations; identifying, for each of the B-values, a dominating breathing state wherein at least one of the plurality of slice locations of the dominating breathing state contains an image slice from the series of images; identifying, for each of the B-values, one or more non-dominating breathing states; and registering, for each of the B-values, all of the one or more non-dominating breathing states to the dominating breathing state.

Abstract: Provided are a computed tomography device and a computed tomography method. The computed tomography device includes a gantry and an image processing processor. The gantry includes a light source for irradiating light, a detector disposed facing the light source and for receiving the light, and an arm for supporting the light source and the detector. The image processing processor receives a two-dimensional detection image for a subject from the detector. The image processing processor converts the received two-dimensional detection image to two-dimensional detection image data. The image processing processor generates three-dimensional reconstruction image data from the two-dimensional detection image data. A computed tomography device and a computed tomography method according to the inventive concept correct an error of a gantry movement path to provide a stable three-dimensional reconstruction image.

Abstract: Method of correcting PET image attenuation and PET/MRI devices are provided. In one aspect, an MRI image of a subject is segmented to obtain respective MRI sub-images corresponding to attenuation units of the subject, respective pseudo CT sub-images corresponding to the MRI sub-images are obtained based on an MRI-CT atlas, the respective pseudo CT sub-images corresponding to the MRI sub-images are spliced to obtain a pseudo CT image corresponding to the MRI image, and attenuation correction is performed for a PET image of the subject according to the pseudo CT image, wherein the PET image and the MRI image are obtained through a same scan process for the subject.

Abstract: A method for generating attenuation map is disclosed. The method includes acquiring an anatomic image and PET data indicative of a subject, wherein the anatomic image comprises a plurality of voxels. The method also includes fetching a reference image to register the anatomic image, the reference image includes voxel segmentation information. The method further includes segmenting the anatomic image into a plurality of regions based on the voxel segmentation information. The method further includes generating a first attenuation map corresponding to the anatomic image by assigning attenuation coefficients to the plurality of regions. The method further includes calculating a registration accuracy between the anatomic image and the reference image. The method further includes determining a probability distribution of attenuation coefficient.

Abstract: Provided are an iterative image reconstruction method, an iterative image reconstruction program, and a tomography apparatus capable of effectively performing image reconstruction by shorting computation time when performing image reconstruction using an iterative method. In this iterative image reconstruction method, in cases where the image quality after image reconstruction is insufficient when the image was obtained with the number of iterations of 100 times in the first image reconstruction processing, when resuming the image reconstruction, the number of iterations and the state variables at the end of the reconstruction in which the number of iterations is 100 times are read, and then the image reconstruction is resumed.

Abstract: Provided are improved referenceless multi-material beam hardening correction methods, with an emphasis on maintaining data quality for real-world imaging of geologic materials with a view towards automation. A referenceless post reconstruction (RPC) correction technique is provided that applies the corrections in integrated attenuation space. A container-only pre-correction technique also is provided to allow automation of the segmentation process required for beam hardening correction methods.

Abstract: Embodiments of the present invention provide a method, system and computer program product for chart decomposition and sequencing in a limited display device. In an embodiment of the invention, a method of chart decomposition and sequencing in a limited display device is provided. The method includes generating a data visualization for raw data and partitioning the data visualization into a set of a discrete number of segments. The method also includes establishing a sequence for the segments in the set. Finally, the method includes displaying the segments sequentially in frames of an animation according to the sequence in a display of the limited display device.

Abstract: Systems and methods are provided for generating interactive chart visualizations that incorporate multiple datasets. The charts include different scales that correspond to the different datasets. At least one of the scales can also be selectively hidden. The hidden scale(s) will be rendered in response to user input selecting data/objects corresponding to the hidden scale(s).

Abstract: Embodiments for a method and system for the making, storage and display of virtual image edits are disclosed. Source/linked image(s) (201a) in a web or other computer device (100a) environment may be virtually edited and/or combined with one another (102), by individual users, in collaboration with others or an automated program. Only the virtual edit alteration code, including the source image location information, are displayed (108) and saved (410), eliminating image redundancy, as the source images are not truly edited, copied or resaved, but instead remain stored at the copyright holder's disclosed location. Just as the House Of Mirrors in an amusement park changes ones reflection without changing the person, separating the image(s) from the virtual edits improves image load speeds and saves storage space. The said display is dynamically affected should the image copyright holder change or delete one of the source images.

Abstract: A cartoon data generating system includes a receiving unit configured to receive a layer selection signal for selecting at least one layer from cartoon data including a cut generated from a combination of at least one layer including different stories and an effect application signal for applying a predetermined effect to the at least one selected layer; a determination unit configured to determine a layer satisfying an expansion condition among the at least one selected layer; and a controller configured to expand the layer satisfying the expansion condition to a predetermined size and generate an effect cut from a combination of the expanded layer to which the predetermined effect is applied and another layer.

Abstract: In one embodiment, a method includes rendering an image comprising a plurality of tags, wherein each tag is associated with a particular location within the image, and wherein at least two tags of the plurality of tags are aggregated into an aggregated tag, calculating for each tag the distance of the location of the tag from the location of every other tag of the image responsive to the first user changing a zoom level for the image, and updating the aggregated tag based on the calculated distance of the locations of the tags to each other such that if the change in zoom level is zooming in, then the updated aggregated tag is aggregated with at least one fewer tag, and if the change in zoom level is zooming out, then the updated aggregated tag is aggregated with at least one more tag.

Abstract: Provided is a display device including a first display unit, a second display unit arranged on a back side of the first display unit, a generation unit configured to perform predetermined image processing on first image data to generate second image data, and a control unit configured to select one of a plurality of display modes including a first display mode on which the first display unit displays an image based on the first image data and the second display unit displays an image based on the second image data, and a second display mode on which the first display unit and the second display unit display the image based on the first image data.

Abstract: An electronic device is provided. The electronic device includes a processor and a memory electrically connected with the processor, wherein the memory may store instructions executed to enable the processor to obtain an image and output a message generated based on additional information and an image analysis result obtained by analyzing the obtained image.

Abstract: Applications and services providing customized realty renovation visualization are described. A realty renovation visualization service can, in response to receiving a request to customize a renovation visualization, query a data resource for a set of inspirational room images. The set of inspirational room images can be images of inspirational rooms with similar spatial characteristics as the room that is the subject of the customized renovation visualization. The request can include information about at least a room depicted in the renovation visualization and the data resource can be queried using parameters of a definite region of the room obtained from the information about at least the room. At least one customized room image can then be generated from the set of inspirational room images using one or more customizing criteria, and the at least one customized room image provided to the source of the request to include in the renovation visualization.

Abstract: A device and method for providing an augmented reality (AR) image are provided. The device includes a display configured to alternately display a first image including an object at an object position and a second image including a mask for blocking transmission of light incident on the display at the object position; a processor configured to determine a time period and control the display to alternately display the first image and the second image during the time period; and a polarization switch configured alternately polarize the light incident on the display at a switching period determined based on the time period.

Abstract: Methods and apparatus to generate augmented content regions for augmented reality (AR) systems are disclosed. An example method includes receiving from a plurality of AR devices data representing a plurality of sight lines captured using the plurality of AR devices, identifying a plurality of commonalities of the plurality of sight lines based on the data representing the plurality of sight lines, and defining an augmented content region based on the plurality of commonalities.

Abstract: Included are: an acceptor configured to accept an operation result of success or failure of an operation by an operator, the operation result being obtained in a case where an optical see-through displayer worn by the operator displays a support image candidate for the operation by the operator superimposed on a reality space; a storing unit configured to store each support image candidate out of a plurality of the support image candidates for the operation and the operation result in a memory, associating the support image candidate with the operation result being in a case where the support image is displayed; and a selector configured to select the support image candidate to be set as the support image of the operation from among the plurality of the support image candidates, based on the operation result.

Abstract: A display assistance device detects the steering angle of a vehicle by using a steering angle sensor, and changes the position of a virtual viewpoint behind the host vehicle based on the detected steering angle. Then, the display assistance device converts images captured using a plurality of cameras such as a front camera and a rear camera into an overhead image of a downward view from the virtual viewpoint, and displays the converted overhead image on a display with a host vehicle icon superimposed on the overhead image, the host vehicle icon indicating the position of the vehicle.

Abstract: An online system customizes video conversations between users of the online system by providing graphics that are likely to interest the users. The online system may present composite views overlaying selected graphics on a video stream, which may be part of an augmented reality (AR) environment. The graphics include, for example, background graphics, masks, props, visual or particular effects, frames or borders, etc. The online system may use a machine learning model to predict whether a user is likely to select or interact with a particular graphic. The online system can also filter or rank graphics according to user affinities or user connections on the online system. Users may be encouraged to interact with the graphics that are customized to their specific interests, which can promote an engaging video conversation or AR experience.

Abstract: A system and method for creating multi-sided digital images involving at least one processor configured to run an image forming application, wherein the processor running said image forming application is configured to: (i) receive a digital sponsor image and a digital personal photo image from a user; (ii) generate an animated or video-based multi-sided digital image with at least one side depicting said digital sponsor image and at least another side depicting said personal photo, said animated or video-based multi-sided digital image configured to rotate and/or flip on a display of said one or more computer devices to show said digital sponsor image and said digital personal image; and (iii) transmit said animated or video-based multi-sided digital image to a user's computer device.

Abstract: Techniques of animating objects in VR involve applying a motion filter to the object that varies with vertices on an object. Along these lines, a VR computer generates an object for an interactive, three-dimensional game by generating a triangular mesh approximating the object surface and bones including vertices defining motion of the vertices based on motion of an anchor vertex. When a user selects a vertex of the object as an anchor vertex about which to move the object, the VR computer generates variable filters for each bone that restrict the motion of that bone based on the distance of that bone from the anchor vertex. Accordingly, when the user produces a gesture with a controller that defines a path of motion for the anchor vertex, the bone including the anchor vertex goes through an unfiltered motion while bones remote from the anchor vertex go through a more restricted motion.

Abstract: A method for virtual reality (VR) includes displaying a VR event in a view of a first VR position in a VR environment in a first period, wherein a user's VR avatar is at the first VR position in the first period; recording the VR event; and redisplaying the recorded VR event in a view of a second VR position in the VR environment in a second period.

Abstract: Disclosed is a real-time motion simulation method for hair and object collisions, which is based on a small amount of pre-computation training data and generates a self-adaptive simplified model for virtual hair style for real-time selection and interpolation and collision correction, thereby realizing real-time high-quality motion simulation for hair-object collisions. The method comprises the following steps: 1) reduced model pre-computation: based on pre-computation simulation data, selecting representative hairs and generating a reduced model; 2) real-time animation and interpolation: clustering the representative hairs simulated in real time; selecting the reduced model and interpolating; and 3) collision correction: detecting collision and applying a correction force on the representative hairs to correct the collisions. The present invention proposed a real-time simulation method for hair-object collision, which achieves similar effect as off-line simulation and reduces the computation time cost.

Abstract: Presented herein are systems and methods configured to generate virtual entities representing real-world users. In some implementations, the systems and/or methods are configured to capture user appearance information with imaging devices and sensors, determines correspondence values conveying correspondences between the appearance of the user's body or user's head and individual ones of default body models and/or default head models, modifies a set of values defining a base body model and/or base head model based on determined correspondence values and sets of base values defining the default body models and/or default head models. The base body model and/or base head model may be modified to model the appearance of the body and/or head of the user.