Abstract: Systems and methods for ensuring that images are compressed in a format that may be decompressed at a speed that is acceptable to a particular user and/or viewing environment are disclosed herein.

Abstract: A medical image-processing apparatus that corrects divergence, even in a case where divergence in CT numbers occurs throughout a plurality of sets of image data that have been acquired for anatomically identical areas because the observed object is undergoing shape-alteration. The apparatus includes an image data storage, an extractor, and a corrector. The image data storage stores respective sets of image data acquired by scanning a targeted region undergoing chronological shape-alteration in a subject, the scanning having been executed at predetermined respective timings. The extractor extracts image data that correspond to the targeted region out of each of the respective sets of image data. The corrector calculates degrees of shape-alteration of the targeted region in the respective extracted sets of images data and to correct CT numbers of the respective sets of image data extracted for the targeted region, based on calculated degrees.

Abstract: The subject of the present invention is a method for characterizing the distribution of the volumes of the surface of the skin of an individual, more particularly applicable in the evaluation of the effect of firming cosmetic products. This method uses the isolines and the gradients of the slopes computed from a three-dimensional representation of the surface of the skin.

Abstract: A method of obtaining fiber structure information of a nerve fiber in an object by using an MRI system includes: obtaining image information of voxels of the object; setting a tracking start area including at least one voxel of the object; setting at least one voxel adjacent to the tracking start area as a tracking processing area; determining a similarity between image information of a voxel included in the set tracking processing area and a voxel included in the tracking start area; and obtaining the fiber structure information according to the determined similarity.

Abstract: A method for producing a graph representation of an input image, the method including the procedures of applying convolutional layers of a trained convolutional neural network on the input image, defining a receptive field of a last convolutional layer of the trained convolutional neural network as a vertex of the graph representation, defining a vector of a three dimensional output matrix of the last convolutional layer that is mapped to the receptive field as a descriptor for the vertex and determining an edge between a pair of vertices of the graph representation by applying an operator on a pair of descriptors respective of the pair of vertices.

Abstract: A scene correlation-based target system and related methods are provided. A reference image depicts a remotely-positioned object having identifiable characteristics, wherein a reference directional vector is established relative to the reference image. A target image of a general vicinity of the remotely-positioned object has an unknown directional vector, the target image having at least a portion of the identifiable characteristics. An inertial measuring unit has a scene correlation system, wherein the scene correlation system matches the portion of the identifiable characteristics of the target image with the identifiable characteristics of the reference image, wherein a slew angle between the reference image and the target image is calculated. A target image directional vector is derived from the calculated slew angle and the reference directional vector.

Abstract: A discriminant function specifying device includes a setting part for calculating a discriminant function by machine learning using a learning input parameter, the discriminant function being for separating a distribution of a genuine similarity from a distribution of impostor similarity, the learning input parameter being set before executing the machine learning, and for setting a plurality of discriminant functions, each of the plurality of discriminant functions being obtained by changing values of the learning input parameter and a learning output parameter, the learning output parameter being a parameter of the calculated discriminant function; a calculation part for calculating an error rate for each of the plurality of discriminant functions set by the setting part; and an extraction part for extracting one discriminant function among the plurality of discriminant functions, the one discriminant function having the error rate calculated by the calculation part which meets a predetermined error conditi

Abstract: Systems and methods for performing face recognition and image searching are provided. A system for face recognition and image searching includes an ingestion system, a search system, a user device, and a database of galley files that include feature vectors. The ingestion system crawls the internet starting with a seed URL to scrape image files and generate feature vectors. Feature vectors of images input by a user may be compared by the search system to feature vectors in the gallery files. A method for generating feature vectors includes landmark detection, component aligning, texture mapping, vector computation, comparing cluster centers defined by vectors stored in a database with vectors generated based on an input image, linear discriminant analysis, and principal component analysis.

Abstract: A method is provided for estimating a parameter of physiological significance. One or more images are provided of a tissue in a subject to whom a dose of a contrast agent (CA) has been administered, using a computer equipped with image processing software, the concentration or relative concentration of the agent in a region or regions of interest in the tissue is determined, thus generating concentration data. The time-based behavior of concentrations of CA within the tissue is determined using a pharmacokinetic model that is based on a set of pharmacokinetic model parameters. Using computer code, the pharmacokinetic model is fit to the concentration data, varying one or more parameters, such that a best fit estimate of a parameter of physiological significance is provided.

Abstract: A fully automatic method for segmentation of the mitral leaflets in 3D transesophageal echocardiographic (3D TEE) images is provided. The method combines complementary probabilistic segmentation and geometric modeling techniques to generate 3D patient-specific reconstructions of the mitral leaflets and annulus from 3D TEE image data with no user interaction. In the model-based segmentation framework, mitral leaflet geometry is described with 3D continuous medial representation (cm-rep). To capture leaflet geometry in a target 3D TEE image, a pre-defined cm-rep template of the mitral leaflets is deformed such that the negative log of a Bayesian posterior probability is minimized. The likelihood of the objective function is given by a probabilistic segmentation of the mitral leaflets generated by multi-atlas joint label fusion, while the validity constraints and regularization terms imposed by cm-rep act as shape priors that preserve leaflet topology and constrain model fitting.

Abstract: A method of attenuation correction for a positron emission tomography (PET) system includes obtaining PET scan data representative of a volume scanned by the PET system, obtaining a plurality of magnetic resonance (MR) scan data sets representative of the volume, each MR scan data set being acquired in a respective time period during acquisition of the PET scan data by the PET system, detecting motion of the volume that occurred during the acquisition of the PET scan data based on an assessment of the plurality of MR scan data sets, the PET scan data, or the plurality of MR scan data sets and the PET scan data, determining attenuation correction data from the plurality of MR scan data sets based on the detected motion for alignment of the attenuation correction data and the PET scan data, and correcting the PET scan data with the attenuation correction data.

Abstract: The invention relates to a method for updating and/or for initializing, in a system comprising an image sensor oriented towards the horizon such as a pair of binoculars coupled to a geographical north seeker incorporating a gyrometer, a value representative of the orientation of the image sensor with respect to north, in which the shift of two images arising from the digital image sensor is determined at a first instant and at a second instant that are distinct so as to construct an item of information representative of an angular movement, and in which this angular movement information is utilized to: update the value of orientation of the image sensor;—and/or to discriminate the movements of the image sensor with respect to the ground from the movements of the image sensor that are due to the rotation of the earth, in a process for initializing the value of orientation of the image sensor using a north seeker of gyroscopic or gyrometric type rigidly tied to the image sensor.

Abstract: A method and system may assess the damage to infrastructure using aerial images captured from an unmanned aerial vehicle (UAV), a manned aerial vehicle (MAV) or from a satellite device. Specifically, an item of infrastructure may be identified for assessing damage. The UAV, MAV, or satellite device may then capture aerial images within an area which surrounds the identified infrastructure item. Subsequently, the aerial images may be analyzed to determine a condition and the extent and/or severity of the damage to the infrastructure item. Furthermore, the aerial images along with indications of the extent of the damage may be displayed on a computing device.

Abstract: Computer-readable storage media, computing devices and methods are discussed herein. In embodiments, a computing device may be configured to perform facial recognition based on gradient based feature extractions of images of faces. In embodiments, the computing device may be configured to determine directional matching patterns of the images from the gradient based feature extraction and may utilize these directional matching patterns in performing a facial recognition analysis of the images of faces. Other embodiments may be described and/or claimed.

Abstract: A method for determining a projection angle for use in tomographic imaging comprises obtaining projection data including at least one projection view from at least one projection angle. The at least one projection view is generated by scanning a test object with a tomographic imaging device. Each projection view comprises at least one observation value obtained at least one detection location, providing a plurality of candidate projection angles, for each candidate projection angle, calculating a function value indicative of an amount of information that may be gained by adding to the projection data a further projection view generated by scanning of the test object with the tomographic imaging device from the candidate projection angle, and selecting a candidate projection angle from the plurality of candidate projection angles taking into account the function values. A corresponding system and computer program product also is provided.

Abstract: An apparatus for estimating movement of a vehicle using stereo image matching is provided, which divides the stereo image information into ground, background and object labels and measures information about rotation and movement of the vehicle using the label and background labels.

Abstract: The invention relates to a deviation indicator with infrared imagery and a system for automatically aiming at and tracking a target. According to the invention, the deviation indicator with infrared imagery comprises an infrared pulsed laser (11) and the device (12) for controlling and processing infrared images also processes the laser pulse echoes.

Abstract: The present invention relates to an imaging technique and an imaging system and more particularly to an automatic assessment of quantitative measure(s)/properties(s) of an object wherein, for example, an imaging system is used to capture an image, following which image properties are quantified using image processing techniques. An imaging technique obtains an image of an object along a first axis, or by way of a first technique, and subsequently obtains an image of the object along a second axis, or by way of a second technique. One or more pixels from the first image are selected and designated as reference pixels. An automated comparison between corresponding regions of the first and second images are then performed which is based upon the reference pixels, so as to indicate regions of interest.

Abstract: An example method includes, responsive to identifying an object and a reflection surface: identifying a reflection incidence point on a current position of the reflection surface; reprojecting the reflection incidence point using a surface motion vector of the reflection incidence point to determine a previous reflection incidence point. The previous reflection incidence point is a reflection incidence point on a previous position of the reflection surface. The method further includes, reprojecting a reflected point (a reflection of the object in accordance with the reflection surface's current position) using a motion vector to determine a previous location of the reflected point on the previous position of the reflection surface; projecting a current view point onto the previous position of the reflection surface to produce a previous view point; and presenting to a user a presentation of the previous location of the reflected point on the previous position of the reflection surface.

Abstract: A method for processing one or more reconstructed images includes receiving a first reconstructed image having a first image resolution and receiving a second reconstructed image having a second image resolution. The first resolution is greater than the second resolution. The method further includes generating a contrast-dependent resolution image by combining the first and second reconstructed images single image so that the first reconstructed image contributes to a greater degree to the contrast-dependent resolution image for higher contrasts regions of the images and the second reconstructed image contributes to a greater degree to the contrast-dependent resolution image for lower contrasts regions of the images.