Abstract: According to one embodiment, a gain adjustment unit executes gain adjustment for 3D B-mode data using gain adjustment values respectively decided for the depth on each scanning line, azimuth direction and elevation direction in a scanned region. A threshold decision unit decides a threshold used for discriminating lumen and non-lumen regions in the scanned region using the gain-adjustment-executed 3D B-mode data. A threshold processing unit executes threshold processing to discriminate data concerning the non-lumen region from the gain-adjustment-executed 3D B-mode data, using the threshold. An image generation unit generates an ultrasonic image concerning the lumen region based on the threshold-processing-executed 3D B-mode data.
Type:
Grant
Filed:
September 5, 2012
Date of Patent:
May 13, 2014
Assignees:
Kabushiki Kaisha Toshiba, Toshiba Medical Systems Corporation
Abstract: Systems, methods, and computer readable storage media are described that can provide a multi-level hierarchical framework to progressively associate detection responses, in which different methods and models are adopted to improve tracking robustness. A modified transition matrix for the Hungarian algorithm can be used to solve the association problem that considers not only initialization, termination and transition of tracklets but also false alarm hypotheses. A Bayesian inference approach can be used to automatically estimate a scene structure model as the high-level knowledge for the long-range trajectory association.
Abstract: In a magnetic resonance method and apparatus to determine an examination image data set describing the temperature distribution within an examination subject (in particular a patient) using magnetic resonance an absolute temperature image data set of an examination region of the subject is acquired by magnetic resonance temperature imaging, a normal temperature data set of the examination region is determined, and the examination image data set is determined in a computerized processor by a comparison of the absolute temperature image data set with the normal temperature data set.
Type:
Grant
Filed:
August 25, 2011
Date of Patent:
April 15, 2014
Assignee:
Siemens Aktiengesellschaft
Inventors:
Joerg Roland, Sebastian Schmidt, Anke Weissenborn
Abstract: A platform is proposed for automated analysis of retinal images, for obtaining from them information characterizing retinal blood vessels which may be useful in forming a diagnosis of a medical condition. A first aspect of the invention proposes that a plurality of characteristics of the retina are extracted, in order to provide data which is useful for enabling an evaluation of cardiovascular risk prediction, or even diagnosis of a cardiovascular condition. A second aspect uses fractal analysis of retinal images to provide vascular disease risk prediction, such as, but not limited to, diabetes and hypertension.
Abstract: The present invention enables the detecting and tracking of a specific subject using a simple structure and simplified operations without registering characteristic information of the subject. The invention comprises: a subject detection means that detects a subject image from an image; a subject-recognition-information storage means that stores a recognition ID of the subject image as well as a first position that is the position at which the subject image was last detected within the image; and a recognition ID assigning means that assigns the recognition ID to the subject image, wherein if the first position is within a predefined distance from the position at which the subject detection means detects the subject image, the recognition ID assigning means assigns the recognition ID corresponding to the first position to the subject image.
Abstract: In a lane boundary detection device, a plurality of edge components are extracted from a captured image capturing the periphery of the own vehicle. Candidates of a curve (including straight lines) that is to be the boundary of a driving area are extracted as boundary candidates based on the placement of the plurality of edge components. Then, an angle formed by a tangent in a predetermined section of each extracted boundary candidate and a vertical line in the captured image is calculated. Boundary candidates of which the formed angle is less than an angle reference value are set to have low probability. The boundary candidate having the highest probability among the boundary candidates is set as the boundary of the driving area.
Abstract: A road estimation device receives data including a core point being assigned along a road and assigned with attributes for identifying the road. An input unit inputs map data including links having attributes corresponding to the attributes of the core point. An extraction unit first narrows down links into candidate links each having an attribute corresponding to a non-shape-relevant attribute of the core point. The extraction unit subsequently narrows down the candidate links into a candidate link having an attribute corresponding to a shape-relevant attribute of the core point, thereby to extract the candidate link pertinent to the road represented by the core point from the map data for estimating the road on a map.
Abstract: Systems and methods for facilitating video fingerprinting are provided. In one embodiment, a system can include: a memory, a microprocessor, a communication component that receives a video, and a video fingerprinting component that fingerprints the video with a subfingerprint (SFP). The video fingerprinting component can employ an SFP component stored in the memory and that comprises: a feature extraction component that determines local descriptors for at least one frame of a video; and a quantization component that quantizes the local descriptors to generate first frame information including a set of values for the at least one frame. The SFP component can also include: an accumulation component that accumulates first frame information over a snippet of the video; and an SFP generation component that computes the SFP associated with the snippet. The SFP can be computed based on a hash based on the accumulated first frame information over the snippet.
Abstract: The present disclosure relates generally signal processing. One claim recites an apparatus comprising: memory for storing a color video signal comprising first data and second data; and a processor. The processor is programmed for: modifying first color information and second color information of the first data by encoding a signal in the first color information such that the signal includes a first signal polarity, and encoding the signal in the second color information such that signal includes a second signal polarity that is inversely related to the first signal polarity, and modifying first color information and second color information of the second data by encoding the signal in the first color information such that signal includes the second signal polarity, and encoding the signal in the second color information such that the signal includes the first signal polarity. Of course, different combinations and claims are provided too.
Abstract: A method for deriving a representation of an image is described. The method involves processing signals corresponding to the image. At least one region of the image that is representative of the image is identified. In one embodiment, each identified region corresponds to a feature of the image. For each identified image region, a two dimensional function of at least part of the image region is derived and processed to obtain an intermediate representation of the at least part of the image region. A representation of the image is derived from the intermediate representation of at least one of the image regions.
Abstract: Method and apparatus for enhancing representations of micro-calcifications in a digital mammogram image. The method includes smoothing the digital mammogram image using a fuzzy smoothing technique to yield a fuzzy smoothed image. The method includes subtracting the fuzzy smoothed image from the digital mammogram image to yield a micro-calcifications enhanced image. The method includes scaling gray level values of pixels in the micro-calcifications enhanced image by a predetermined amount to provide a digital mammogram image with enhanced representations of the micro-calcifications. Apparatus for enhancing representations of micro-calcifications includes an image processing unit that uses an image acquisition unit to receive an image and a digital signal processor to process the image to provide a digital mammogram image with enhanced representations of micro-calcifications.
Type:
Grant
Filed:
October 7, 2010
Date of Patent:
January 21, 2014
Assignee:
Texas Instruments Incorporated
Inventors:
Hrushikesh Garud, Debdoot Sheet, Amit Suveer, Manjunatha Mahadevappa, Ajoy Kumar Ray
Abstract: A digital image signal processing apparatus includes a plurality of line memories, a line memory control unit that controls input image data corresponding to an input image to be input to each of the plurality of line memories and to be output from the plurality of line memories, a data patch generating unit that sequentially accesses the input image data stored in each of the plurality of line memories and generates a data patch for noise reduction filtering, and a filtering unit that performs the noise reduction filtering on the data patch and generates output image data. A digital image signal processing apparatus and method thereof thus may perform a high-speed noise reduction filtering operation and generate an output image having a same size as an input image while maintaining a same quality as the input image.
Abstract: Trajectories of objects are estimated by determining the optimal solution(s) of a tracking model on the basis of an occupancy probability distribution. The occupancy probability distribution is the probability of presence of objects over a set of discrete points in the spatio-temporal space at a number of time steps. The tracking model is defined by the set of discrete points, a virtual source location and a virtual sink location, wherein objects in the tracking model are creatable in the virtual source location and are removable in the virtual sink location.
Type:
Grant
Filed:
January 11, 2012
Date of Patent:
December 24, 2013
Assignee:
Ecole Polytechnique Federale de Lausanne (EPFL)
Inventors:
François Fleuret, Jérôme Berclaz, Engin Türetken, Pascal Fua
Abstract: What is disclosed is a system and method for determining whether a front seat in a motor vehicle is occupied based on seat pattern recognition. The present invention takes advantage of the observation that an unoccupied seat of a motor vehicle exhibits features which are distinguishable from an occupied seat. An unoccupied motor vehicle seat typically features long contiguous horizontal line segments and curve segments, and substantially uniform areas encompassed by these segments which are not present in an occupied seat. The present method provides a long horizontal edge test which uses location information within a defined window of the image, edge linking, softness of the edge, number of lines, line/curve fitting, and other techniques to locate horizontal edges in the image which define a seat, and a uniformity step which determines whether the area bounded by the horizontal edges is relatively uniform indicating an unoccupied seat.
Type:
Grant
Filed:
August 23, 2011
Date of Patent:
December 17, 2013
Assignee:
Xerox Corporation
Inventors:
Zhigang Fan, Abu Saeed Islam, Peter Paul, Beilei Xu, Lalit Keshav Mestha
Abstract: According to one embodiment, a search skip region setting function generation method includes estimating. The estimating includes estimating a relative position between a object and a template based on a distribution of surrounding search point similarities, and generates a function required to set a search skip region which allows to skip the object search on each model reduced-scale image based on the estimated relative position.
Abstract: A method of locating a check image region within a document image comprising the steps of locating a magnetic ink character recognition region of the check and calculating the top of the check relative to the magnetic ink character recognition region by detection of string literals having a historical and/or contextual relationship to the upper check boundary.
Type:
Grant
Filed:
March 16, 2012
Date of Patent:
December 10, 2013
Assignee:
Hyland Software, Inc.
Inventors:
Charles W. Jackson, Steven Mandel, Robert Scott Moriarty
Abstract: An automatic measurement system for plant features includes: at least one photographic device, for photographing a plant in a container, so as to generate a top view image and a side view image; and an operation processing device, electrically connected to the photographic devices, and at least including an image processing module and a storage unit. The image processing module analyzes at least one of the top view image and the side view image by using at least one image processing procedure and at least one analysis rule, so as to generate at least one feature data corresponding to appearance features of the plant, and records the at least one feature data in the storage unit.
Abstract: Disclosed is a method to generate data describing a product. The method includes the steps of comparing a digitized query image of the product to digitized pre-existing product images in a pre-existing product database. The pre-existing product database is organized using a taxonomy and an ontology. The pre-existing product images are linked to a corresponding node in the taxonomy and are also linked to attribute data and attribute value data in the ontology. At least one pre-existing product image is then retrieved that most closely matches the query image based on at least one matching criterion selected in whole or in part by a user. From the pre-existing product database is extracted the node in the taxonomy, the attribute data, or the attribute value data linked to the pre-existing product image retrieved earlier. In this fashion, product data relevant to the item depicted in the query image can be generated automatically from existing product data.
Type:
Grant
Filed:
October 8, 2010
Date of Patent:
November 12, 2013
Assignee:
Edgenet, Inc.
Inventors:
Ric Clippard, Rajesh Kommineni, Brian Rudolph, Scott Rudolph
Abstract: The processing comprises an analysis of a seismic image to estimate seismic horizons in an area of the subsurface and the calculation of an accumulation value associated with each pixel of the seismic image by accumulation along a set of estimated seismic horizons to form a synthesized image composed of accumulation values. This synthesized image is transformed to obtain a seismic representation in which signed amplitudes are allocated to the pixels of the synthesized image. The transformation may comprise a combination with the original seismic image or else a convolution with a wavelet or another convolution kernel.