Abstract: Both pixel-oriented analysis and the more accurate yet slower object-oriented analysis are used to recognize patterns in images of stained cancer tissue. Images of tissue from other patients that are similar to tissue of a target patient are identified using the standard deviation of color in the images. Object-oriented segmentation is then used to segment small portions of the images of the other patients into object exhibiting object characteristics. Pixelwise descriptors associate each pixel in the remainder of the images with object characteristics based on the color of pixels at predetermined offsets from the characterized pixel. Pixels in the image of the target patient are assigned object characteristics without performing the slow segmentation of the image into objects. A pixel heat map is generated from the target image by assigning pixels the color corresponding to the object characteristic that the pixelwise descriptors indicate is most likely associated with each pixel.

Abstract: In implementations of the subject matter described herein, a solution for object detection is proposed. First, a feature(s) is extracted from an image and used to identify a candidate object region in the image. Then another feature(s) is extracted from the identified candidate object region. Based on the features extracted in these two stages, a target object region in the image and a confidence for the target object region are determined. In this way, the features that characterize the image from the whole scale and a local scale are both taken into consideration in object recognition, thereby improving accuracy of the object detection.

Abstract: Systems, and methods are provided for computer based method for determining a source of a registration error in a surgical navigation system, by registering at least two patient reference markers in a common coordinate space, positioning at least one reference marker in a known position with respect to at least one of two patient reference markers. A relative position of the at least two patient reference markers with respect to each other in the common coordinate space is monitored, as well as a position of said at least one reference marker with respect to a first of the at least two patient reference markers.

Abstract: Systems and methods are provided for image re-orientation. A computing system for image re-orientation may comprise one or more processors and a memory for storing instruction that, when executed by the processors, cause the system to perform: obtaining positive samples including text of a predetermined type in a predetermined orientation range and negative samples not including text of the predetermined type in the predetermined orientation range; training a model with the obtained positive and negative samples; receiving an image in an original orientation; applying the trained model to the received image; in response to determining that the received image does not comprise any text of the predetermined type in the predetermined orientation range, rotating the image by a predetermined angle and re-applying the trained model to the rotated image to determine if the rotated image comprises at least some text of the predetermined type in the predetermined orientation range.

Abstract: Both object-oriented analysis and the faster pixel-oriented analysis are used to recognize patterns in an image of stained tissue. Object-oriented image analysis is used to segment a small portion of the image into object classes. Then the object class to which each pixel in the remainder of the image most probably belongs is determined using decision trees with pixelwise descriptors. The pixels in the remaining image are assigned object classes without segmenting the remainder of the image into objects. After the small portion is segmented into object classes, characteristics of object classes are determined. The pixelwise descriptors describe which pixels are associated with particular object classes by matching the characteristics of object classes to the comparison between pixels at predetermined offsets. A pixel heat map is generated by giving each pixel the color assigned to the object class that the pixelwise descriptors indicate is most probably associated with that pixel.

Abstract: A line-of-sight detection method includes detecting an eye region of a subject from an image, determining whether a position of a boundary between a pupil and an iris is detected from the image based on a brightness change in the eye region, executing, in accordance with a result of the determining, at least one of first processing in which a position of a center of the pupil is detected based on an outline of the pupil and second processing in which the position of the center of the pupil is detected based on another outline of the iris, and detecting a line of sight of the subject based on the position of the center of the pupil detected by the at least one of the first processing and the second processing.

Abstract: According to an embodiment, an object detection apparatus includes an imaging unit, an extraction unit, an estimation unit, and a detection unit. The imaging unit acquires an image of a plurality of polyhedrons which are loaded, each of the polyhedrons having a marker on a surface, each of marker including a first straight line portion and a second straight line portion. The extraction unit that extracts a plurality of markers included in the image. The estimation unit estimates a region included in the surface by using the first straight line portion and the second straight line portion of the extracted markers. The detection unit detects a region having all corners provided with the markers as a surface of a desired polyhedron.

Abstract: A method for tomosynthesis volume reconstruction acquires at least a prior projection image of the subject at a first angle and a subsequent projection image of the subject at a second angle. A synthetic image corresponding to an intermediate angle between the first and second angle is generated by a repeated process of relating an area of the synthetic image to a prior patch on the prior projection image and to a subsequent patch on the subsequent projection image according to a bidirectional spatial similarity metric, wherein the prior patch and subsequent patch have n×m pixels; and combining image data from the prior patch and the subsequent patch to form a portion of the synthetic image. The generated synthetic image is displayed, stored, processed, or transmitted.

Abstract: Techniques for segmentation of organs and tumors and cells in image data include revising a position of a boundary by evaluating an evolution equation that includes differences of amplitude values for voxels on the boundary from a statistical metric of amplitude of voxels inside, and from a statistical metric of amplitude of voxels outside, for a limited region that lies within a distance r of the boundary. The distance r is small compared to a perimeter of the first boundary. Some techniques include determining a revised position of multiple boundaries by evaluating an evolution equation that includes differences in a first topographical distance from a first marker and a second topographical distance from a second marker for each voxel on the boundary, and also includes at least one other term related to boundary detection.

Abstract: An image processing apparatus includes a distance information acquisition unit that acquires distance information, associated with an object image included in each of a first image and a second image, and corresponding to a distance in a depth direction of the object, a calculation unit that calculates a movement vector of the object image, at least in a horizontal direction and a vertical direction, based on position information of the object image, and a generation unit that generates stereoscopic print data of the acquired second image, based on the second image, the distance information, and the movement vector, by increasing a depth component of the position information of the object image included in the second image based on at least part of the movement vector. In addition, a storing unit stores the generated stereoscopic print data in one of a second memory and a recording medium.

Abstract: Provided are a system and method of detecting moving objects. The system stores pixel information regarding each of pixels included in frames of the video in a storage, sets a background model comprising at least one background model element, the at least one background model element indicating at least one of a reference visual property and a reference usage time, determines whether the pixels are background pixels or foreground pixels by comparing the pixel information with the at least one background model element, and updating the background model based on a result of the comparing.

Abstract: Provided herein is technology relating to analysis of images and particularly, but not exclusively, to methods and systems for determining the area and/or volume of a region of interest using optical coherence tomography data. Some embodiments provide for determining the area and/or volume of a lesion in retinal tissue using three-dimensional optical coherence tomography data and a two-dimensional optical coherence tomography fundus image.

Abstract: Disclosed are an image processing apparatus, an image processing method and a recording medium thereof, the image processing apparatus including: a storage configured to store standard information about at least one anatomical entity; and at least one processor configured to detect regions corresponding to a plurality of anatomical entities based on a medical image obtained by scanning an object including the plurality of anatomical entities, to estimate a volume of a first anatomical entity at a predetermined point in time based on object information measured from the detected regions of the anatomical entity and the standard information stored in the storage, and to provide information about condition of the first anatomical entity based on the estimated volume. Thus, it is possible to make a diagnosis more simply and accurately by determining condition information of an anatomical entity at a point in time for the diagnosis based on a randomly taken medical image.

Abstract: Techniques for segmentation include determining an edge of voxels in a range associated with a target object. A center voxel is determined. Target size is determined based on the center voxel. In some embodiments, edges near the center are suppressed, markers are determined based on the center, and an initial boundary is determined using a watershed transform. Some embodiments include determining multiple rays originating at the center in 3D, and determining adjacent rays for each. In some embodiments, a 2D field of amplitudes is determined on a first dimension for distance along a ray and a second dimension for successive rays in order. An initial boundary is determined based on a path of minimum cost to connect each ray. In some embodiments, active contouring is performed using a novel term to refine the initial boundary. In some embodiments, boundaries of part-solid target objects are refined using Markov models.

Abstract: The image capturing part 201 captures a ball in flight with a camera continuously. The image corresponding part 203 generates a first registered ball image obtained by making a size of a first ball image in a first capturing image captured first correspond to a size of a second ball image in a second capturing image captured second. The 3D model constructing part 204constructs a 3D model of the first registered ball image obtained by converting a camera coordinate system of the generated first registered ball image into a world coordinate system. The virtual rotating part 205 rotates virtually the constructed 3D model of the first registered ball image by using a rotation parameter estimated in advance and rotation matrix information.

Abstract: An image obtaining unit obtains a plurality of projection images taken by imaging a subject with different radiation source positions. A frequency decomposition unit performs frequency decomposition on each of the projection images to obtain a plurality of band projection images representing frequency components for individual frequency bands for each of the projection images. A reconstruction unit reconstructs the band projection images to generate band tomographic images of the subject for the individual frequency bands. A frequency synthesis unit performs frequency synthesis on the band tomographic images to generate a tomographic image of the subject.

Abstract: Aspects of the present invention are related to methods and systems for vision-based computation of ego-motion.

Abstract: Embodiments of a method for detection of plurality of three-dimensional cephalometric landmarks in volumetric data are disclosed. In some embodiments, a three-dimensional matrix is developed by stacking of volumetric data and the bony structure is segmented through thresholding. Initially a seed point is searched for initializing the process of landmark detection. Two three-dimensional distance vectors are used to define and obtain the Volume of Interest (VOI). First 3-D distance vector helps to identify Empirical Point and consecutively second gives dimensions of the VOI. Three-dimensional contours of anatomical structure are traced in the estimated VOI. Cephalometric landmarks are identified on the boundaries of traced anatomical geometry, based on corresponding Mathematical Entities. Detected landmark can be used as a Reference Point for further detection of landmarks. Estimating the VOI and detection of points continues till all desired landmarks are detected.

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 system and method for automatically analyzing an image of cells to detect cancer. Some embodiments include eliciting and receiving a digital photomicrograph image of cells; determining a boundary of a cell in the image; identifying a plurality of characteristics of the cell from image-pixel data from within the identified boundary of the cell; reading a plurality of cell characteristics of a plurality of types of cells from a database; comparing the identified characteristics of the cells in the image to the plurality of cell characteristics read from the database; and determining a pathology based on the comparing. Some embodiments further include automatically identifying an appropriate treatment and applying the identified treatment.