Patents Examined by Michelle M Entezari
-
Patent number: 11354542Abstract: A mechanism is described for facilitating on-the-fly deep learning in machine learning for autonomous machines. A method of embodiments, as described herein, includes detecting an output associated with a first deep network serving as a user-independent model associated with learning of one or more neural networks at a computing device having a processor coupled to memory. The method may further include automatically generating training data for a second deep network serving as a user-dependent model, where the training data is generated based on the output. The method may further include merging the user-independent model with the user-dependent model into a single joint model.Type: GrantFiled: February 6, 2020Date of Patent: June 7, 2022Assignee: Intel CorporationInventor: Raanan Yonatan Yehezkel Rohekar
-
Patent number: 11334741Abstract: A method and apparatus for sorting is described, and which includes providing a product stream formed of individual objects of interest having feature aspects which can be detected; generating multiple images of each of the respective objects of interest; classifying the feature aspects of the objects of interest; identifying complementary images by analyzing some of the multiplicity of images; fusing the complementary images to form an aggregated region representation of the complementary images; and sorting the respective objects of interest based at least in part upon the aggregated region representation which is formed.Type: GrantFiled: March 24, 2020Date of Patent: May 17, 2022Assignee: Key Technology, Inc.Inventors: Peter Lodewyckx, Marc Van Daele, Timothy Justice
-
Patent number: 11328523Abstract: The present disclosure relates to an image composite system that employs a generative adversarial network to generate realistic composite images. For example, in one or more embodiments, the image composite system trains a geometric prediction neural network using an adversarial discrimination neural network to learn warp parameters that provide correct geometric alignment of foreground objects with respect to a background image. Once trained, the determined warp parameters provide realistic geometric corrections to foreground objects such that the warped foreground objects appear to blend into background images naturally when composited together.Type: GrantFiled: June 9, 2020Date of Patent: May 10, 2022Assignee: Adobe Inc.Inventors: Elya Shechtman, Oliver Wang, Mehmet Yumer, Chen-Hsuan Lin
-
Patent number: 11324570Abstract: The present invention provides digitally driven methods to conveniently and efficiently monitor and evaluate the positions of a patient's teeth during the course of orthodontic treatment. In an exemplary embodiment, an initial set of 3D digital data representing a patient's dental structure is provided by means of an X-ray radiograph and/or an intraoral scan. Then during the course of treatment, a reduced image representing part of the dental structure may be quickly provided using a bite plate or a low resolution intraoral scan. A three-dimensional (3D) image of the full dental structure is subsequently rendered or “reconstructed” by registering elements of the initial image with corresponding elements of the reduced image. By reconstructing the dental structure from the reduced image, the treating professional can provide a precise, manipulable 3D image of the patient's complete dental structure for diagnostic and treatment planning purposes.Type: GrantFiled: April 18, 2019Date of Patent: May 10, 2022Assignee: 3M INNOVATIVE PROPERTIES COMPANYInventor: David K. Cinader, Jr.
-
Patent number: 11321821Abstract: The present disclosure provides a computer implemented method and system for compensating for cloud cover to allow the build-up of composite, cloud free images of a particular geographic area. The disclosure takes satellite or aerial image data as an input and temporally stacks the images for a particular geographic area, with the most recent image at the top. The time for the composite image to be produced (which may be the present time, or may be a time from the past) is defined as a temporal projection plane. The most recent image prior to or temporally proximate to this plane is analysed to identify any areas obscured by clouds. Cloud free areas of the most recent image are projected onto the temporal projection plane. For the identified obscured areas, older, cloud free images in the stack are projected onto the temporal projection plane to fill the obscured areas. This forms a cloud free composite image at the defined temporal projection plane.Type: GrantFiled: February 10, 2020Date of Patent: May 3, 2022Assignee: Ordnance Survey LimitedInventors: Lynne Allan, Miles Austen
-
Patent number: 11317108Abstract: The present disclosure generally relates to a device and a method of decoding a color picture from a bitstream.Type: GrantFiled: October 17, 2019Date of Patent: April 26, 2022Assignee: INTERDIGITAL VC HOLDINGS, INC.Inventors: Sebastien Lasserre, Fabrice Leleannec, Yannick Olivier
-
Patent number: 11302117Abstract: A method of detecting an anomaly is provided, including dividing each frame of a video stream into a plurality of cells; in each cell formulate statistics based on metadata generated for the frame, the metadata related to presence of an object in the cell, velocity of objects in the cell, direction of motion of objects in the cell, and classification of objects in the cell; and using the formulated statistics to determine when the anomalous activity has occurred in one of the cells of the plurality of cells.Type: GrantFiled: April 9, 2020Date of Patent: April 12, 2022Assignee: AVIGILON CORPORATIONInventors: Dharanish Kedarisetti, Peter Venetianer
-
Patent number: 11288507Abstract: An electronic device includes circuitry that determines probability map information for a first image, based on application of a neural network model on the first image. The neural network model is trained to detect one or more objects based on a plurality of images associated with the one or more objects. The probability map information indicates a probability value for each pixel in the first image. A region corresponding to the one or more objects is detected in the first image based on the probability map information. A first set of sub-images is determined from the detected region, based on application of a stochastic optimization function on the probability map information. The one or more objects are detected from a second set of sub-images of the first set of sub-images, based on application of the neural network model on the second set of sub-images.Type: GrantFiled: September 27, 2019Date of Patent: March 29, 2022Assignee: SONY CORPORATIONInventors: Michiro Hirai, Nobutaka Mitsuma, Nikolaos Georgis
-
Patent number: 11284849Abstract: Methods and apparatus for grading coronary calcium, including an example apparatus that includes a set of logics and a memory that stores a set of radiographic images. The set of logics includes an image acquisition logic that acquires a set of radiographic images of a region of tissue, the set of radiographic images including a low-energy reference x-ray image and a higher-energy floating x-ray image, a registration logic that produces a set of registered images from a plurality of members of the set of radiographic images, an enhancement logic that produces a set of calcium-enhanced images from the set of registered images, a post-enhancement logic that produces a bone image from the set of calcium-enhanced images, and a grading logic that computes a coronary heart disease prediction score based, at least in part, on the bone image and a set of features extracted from the bone image.Type: GrantFiled: September 30, 2019Date of Patent: March 29, 2022Assignee: Case Western Reserve UniversityInventors: David Wilson, Robert Gilkeson, Di Wen
-
Patent number: 11288410Abstract: Described herein are systems and methods for marking a dental aligner. A method for marking a dental aligner includes receiving a digital model corresponding to a dental aligner. The digital model includes a dental arch comprising a plurality of teeth. The method further includes determining whether a section of teeth on a right side or a left side of the dental arch includes flatter occlusal surfaces and selecting the section with the flatter occlusal surfaces, identifying surfaces on the teeth of the selected section that are flat relative to other surfaces on the teeth of the selected section, determining a line of best fit between the flat surfaces, and marking the dental aligner with a marking based on the line of best fit.Type: GrantFiled: November 15, 2019Date of Patent: March 29, 2022Assignee: SDC U.S. SmilePay SPVInventors: Josh Long, Christopher Yancey, Tony Solarek, Clete Culp
-
Patent number: 11281903Abstract: An image processing system or electronic device may implement processing circuitry. The processing circuitry may receive an image, such as financial document image. The processing circuitry may determine a character count for the financial document image or particular portions of the financial document image without recognizing any particular character in the financial document image. In that regard, the processing circuitry may determine a top left score for pixels in the financial document, the top left score indicating or representing a likelihood that a particular pixel corresponds to a top left corner of a text character. The processing circuitry may also determine top right score for image pixels. Then, the processing circuitry may identify one or more text chunks using the top left and top rights scores for pixels in the financial document image. The processing circuitry may determine a character count for the identified text chunks.Type: GrantFiled: December 9, 2019Date of Patent: March 22, 2022Assignee: UNITED SERVICES AUTOMOBILE ASSOCIATION (USAA)Inventor: Adam S. Newman
-
Patent number: 11282259Abstract: Aspects of the present invention provide an approach for non-visually mapping an environment. In an embodiment, a set of images that is within the field of view of the user is captured from a video capture device worn by the user. Environmental text that is within the set of images is identified. An aural contextual indicator that corresponds to the environmental text is then generated. This aural contextual indicator indicates the informational nature of the environmental text. An aural map of the environment is created using a sequence of the generated aural contextual indicators. This aural map is delivered to the user via an aural output device worn by the user in response to a user request.Type: GrantFiled: November 26, 2018Date of Patent: March 22, 2022Assignee: International Business Machines CorporationInventors: Aaron K. Baughman, Gary F. Diamanti, Mauro Marzorati, Moncef Benboubakeur
-
Patent number: 11278656Abstract: A pressure measuring device 30 installs on a tube 11 for transferring a medium (e.g., blood in a extracorporeal blood circulator) so as to measure a pressure of the medium inside the tube 11. The pressure measuring device 30 includes a main body portion 31 mountable to the tube 11, an image acquisition unit 32 disposed in the main body portion 31 so as to acquire image information on a pressure receiver that is deformed in response to the received pressure of the medium inside the tube 11, and a control unit 100 that converts the image information acquired by the image acquisition unit into pressure information about the pressure.Type: GrantFiled: December 4, 2019Date of Patent: March 22, 2022Assignee: TERUMO KABUSHIKI KAISHAInventors: Kozo Hisamatsu, Yosuke Itamochi
-
Patent number: 11270478Abstract: A method for reconstructing an image may include obtaining scan data relating to a subject. The method may also include determining a first field of view (FOV) and determining a second FOV. The method may further include reconstructing a first image based on a first portion of the scan data corresponding to the first field of view, and reconstructing a second image based on a second portion of the scan data corresponding to the second field of view. The method may also include generating a third image based on the first image and the second image.Type: GrantFiled: July 22, 2019Date of Patent: March 8, 2022Assignee: SHANGHAI UNITED IMAGING HEALTHCARE CO., LTD.Inventors: Xin Wang, Guotao Quan
-
Patent number: 11270526Abstract: A teaching assistance method and a teaching assistance system using said method, the teaching assistance method comprising implementing behaviour detection of students in classroom images by means of using a trained depth tensor column network model, thus providing higher image recognition precision and reducing the hardware requirements for algorithms, and being able to be used on an embedded device, reducing the usage costs of the teaching assistance method; in addition, a teaching assistance system using said teaching assistance method has the same advantages.Type: GrantFiled: August 7, 2017Date of Patent: March 8, 2022Assignees: SHENZHEN INSTITUTES OF ADVANCED TECHNOLOGY CHINESE ACADEMY OF SCIENCES, SHENZHEN SIBIKU TECHNOLOGY CO., LTDInventors: Shuqiang Wang, Yongcan Wang, Yue Yang, Yanyan Shen, Minghui Hu
-
Patent number: 11263784Abstract: Examples disclosed herein relate to determining image capture position information based on a quasi-periodic pattern. For example, a processor may determine whether a target area is within a captured image based on the detection of a quasi-periodic pattern in a first detection area and in a second detection area of the captured image.Type: GrantFiled: December 10, 2019Date of Patent: March 1, 2022Assignee: Hewlett-Packard Development Company, L.P.Inventors: Robert Ulichney, Matthew Gaubatz, Stephen Pollard
-
Patent number: 11256923Abstract: Particular embodiments extract features for frames from a video. Then, a sequence of frames is identified based on a pattern analysis of frames based on the features. This pattern may be used to select sentinel features from the sequence of frames. For example, a pattern may include a first sentinel sequence followed by a transitional frame and then a second sentinel sequence. The transitional frame may be a black frame that is used to identify back-to-back sentinel sequences. Then, a sentinel frame that demarks a transition from a first content type to a second content type is identified. For example, the frames on either side of the black frame may be very similar and be identified as sentinel frames that mark a transition from program content to advertisement content. The above process may allow automated detection of sentinel frames that can run without user supervision.Type: GrantFiled: May 12, 2016Date of Patent: February 22, 2022Assignee: ARRIS Enterprises LLCInventors: Stephen P. Emeott, Renxiang Li
-
Patent number: 11256946Abstract: The present disclosure relates to a method and system of processing original handwriting input, the system and method being capable of recognize a plurality of strokes provided on the input recognition interface, the method including: determining a stroke box around each stroke; determining overlap between the stroke boxes; correlating overlapping stroke boxes to one or more characters; providing a character box around each of the one or more characters; determining overlap between character boxes; correlating overlapping character boxes to one or more words; providing a word box around each of the one or more words; provide a word margin around each of the one or more word boxes; determining overlap between each word box to determine a line; wherein each of the characters, words, or lines can be individually selected and rearranged, the system automatically adjusting spacing or placement of surrounding elements to allow for the rearrangement.Type: GrantFiled: March 25, 2019Date of Patent: February 22, 2022Assignee: Boston Inventions, LLCInventor: Samuel David Provencher
-
Patent number: 11256963Abstract: A surgical instrument detection system is provided that can determine the kinds of surgical instruments without special processing, such as application of an optically readable symbol, to the surgical instruments. A surgical instrument detection system 100 includes: an image input section 31 to input an image taken by a camera 1; an object extraction section 32 to clip an object image of a small steel article from the input image; a determination section 33 to input the object image to a learned classification model 331 and determine a kind of the small steel article based on features included in the object image; and an output image generation section 34 to generate an image representing the result of determination by the determination section and output such image to a monitor 2.Type: GrantFiled: May 30, 2018Date of Patent: February 22, 2022Assignee: EIZO CorporationInventors: Takuya Katayama, Nobuaki Sugii
-
Patent number: 11244442Abstract: The correlation of optical images with SEM images includes acquiring a full optical image of a sample by scanning the sample with an optical inspection sub-system, storing the full optical image, identifying a location of a feature-of-interest present in the full optical image with an additional sources, acquiring an SEM image of a portion of the sample that includes the feature at the identified location with a SEM tool, acquiring an optical image portion at the location identified by the additional source, the image portions including a reference structure, correlating the image portion and the SEM image based on the presence of the feature-of-interest and the reference structure in both the image portions and the SEM image, and transferring a location of the feature-of-interest in the SEM image into the coordinate system of the image portion of the full optical image to form a corrected optical image.Type: GrantFiled: July 11, 2019Date of Patent: February 8, 2022Assignee: KLA CorporationInventors: Hucheng Lee, Lisheng Gao, Jan Lauber, Yong Zhang