Patents Examined by Jiangeng Sun
-
Patent number: 11417006Abstract: A stereo matching method includes extracting feature points of a first image and feature points of a second image, the first image and the second image together constituting a stereo image, determining reference points by matching the feature points of the second image to the feature points of the first image, classifying the reference points, and performing stereo matching on pixels of which disparities are not determined in the first image and the second image based on disparities of the reference points in the pixels determined based on a result of the classifying.Type: GrantFiled: September 22, 2020Date of Patent: August 16, 2022Assignee: Samsung Electronics Co., Ltd.Inventors: Seungin Park, Minsu Ahn, Jiyeon Kim
-
Patent number: 11410351Abstract: Embodiments of the present disclosure may disclose a method for improving CT image quality. The method for improving the CT image quality may include obtaining SFS data. The SFS data may include SFS scan data or an SFS image. The SFS scan data may be acquired by a CT device in an SFS state. The SFS image may be generated by reconstructing scan data acquired by the CT device via scanning in the SFS state. The method may include generating a corresponding optimized image by processing the SFS data based on an image quality optimization model. The image quality optimization model may be a machine learning model. The present disclosure may simulate the SFS image as an FFS image using a deep neural network model, thereby improving a resolution of the SFS image and reduce artifacts in the SFS image.Type: GrantFiled: June 3, 2020Date of Patent: August 9, 2022Assignee: SHANGHAI UNITED IMAGING HEALTHCARE CO., LTD.Inventors: Yuan Bao, Guotao Quan, Rengcai Yang
-
Patent number: 11403735Abstract: A method for structure simulation for super-resolution fluorescence microscopy, the method including receiving a first image having a first resolution, which is indicative of a distribution of fluorophores; applying a Markov model to the fluorophores to indicate an emission state of the fluorophores; generating a plurality of second images, having the first resolution, based on the first image and the Markov model; adding DC background to the plurality of second images to generate a plurality of third images, having the first resolution; downsampling the plurality of third images to obtain a plurality of fourth images, which have a second resolution, lower than the first resolution; and generating a time-series, low-resolution images by adding noise to the plurality of fourth images. The time-series, low-resolution images have the second resolution.Type: GrantFiled: December 4, 2018Date of Patent: August 2, 2022Assignee: KING ABDULLAH UNIVERSITY OF SCIENCE AND TECHNOLOGYInventors: Xin Gao, Yu Li, Renmin Han
-
Patent number: 11403731Abstract: Disclosed is an image upscaling apparatus that includes: multiple convolution layers, each configured to receive an input image or a feature map outputted by a previous convolution layer and extract features to output a feature map; and a multilayer configured to receive a final feature map outputted from the last convolution layer and output an upscaled output image. The multilayer includes: a first partition layer including first filters having a minimum size along the x-axis and y-axis directions and the same size as the final feature map along the z-axis direction; and at least one second partition layer, each including second filters, having a size greater than that of the first filter in the x-axis and y-axis directions and having a number and size of the first filter in the z-axis direction, and configured to shuffle features in the x-axis and y-axis directions of the first shuffle map.Type: GrantFiled: September 23, 2020Date of Patent: August 2, 2022Assignee: INDUSTRY-ACADEMIC COOPERATION FOUNDATION, YONSEI UNIVERSITYInventors: Seong Ook Jung, Sung Hwan Joo, Su Min Lee
-
Patent number: 11394956Abstract: An information processing device encodes a plurality of documents into a first plurality of encoded documents, respectively based on first encoding information in which a plurality of words and a plurality of first codes of a first code group are associated, words included in the plurality of documents and included in the first encoding information being encoded in the plurality of encoded documents. The information processing device performs frequency counts for each of a plurality of codes in the first encoded documents encoded in the first encoding. The information processing device encodes the plurality of first encoded documents into a plurality of second encoded documents respectively, utilizing a result of the frequency counts.Type: GrantFiled: June 12, 2020Date of Patent: July 19, 2022Assignee: Fujitsu LimitedInventors: Masao Ideuchi, Masahiro Kataoka, Kosuke Tao
-
Patent number: 11361442Abstract: A machine learning algorithm is trained on a number of microscopic images and a measure of outcome of each image. Each image is divided into tiles. The measure of outcome is assigned to each tile of the image. The tiles are then used to train the machine learning algorithm. The trained algorithm may then be used to evaluate images.Type: GrantFiled: November 9, 2018Date of Patent: June 14, 2022Assignee: OSLO UNIVERSITETSSYKEHUSInventors: Ole Johan Skrede, Tarjei Sveinsgjerd Hveem, John Robert Maddison, Havard Emil Greger Danielsen, Knut Liestol
-
Patent number: 11341732Abstract: First intraoral images of a first portion of a three-dimensional intraoral object are received. The first intraoral images correspond to an intraoral scan of the three-dimensional intraoral object during a current patient visit. A pre-existing model that corresponds to the three-dimensional intraoral object is identified. The pre-existing model is based on intraoral data of the three-dimensional intraoral object captured during a previous patient visit. A first intraoral image of the first intraoral images is registered to a first portion of the pre-existing model. A second intraoral image of the first intraoral images is registered to a second portion of the pre-existing model.Type: GrantFiled: September 12, 2019Date of Patent: May 24, 2022Assignee: Align Technology, Inc.Inventor: Gilad Furst
-
Patent number: 11341657Abstract: Systems and methods for monitoring and controlling dynamic multi-phase flow phenomena, capable of sensing, detecting, quantifying, and inferring characteristics, properties, and compositions (including static and dynamic characteristics, properties and compositions). The systems combine machine vision and mathematical models, which enables direct observation and detection of static and dynamic multi-phase fluid flow properties and phenomena (e.g. voids, waves, shadows, dimples, wrinkles, foam, bubbles, particulates, discrete materials, collections of materials, and position) and inferring other properties and phenomena (e.g. flow regimes, bubble velocities and accelerations, material deposition rates, erosion rates, phasic critical behavioral points as related to heat transfer, and the volumetric and mass flow rates of the phases) that are used to monitor and control systems applied to a multi-phase fluid flow system.Type: GrantFiled: March 31, 2020Date of Patent: May 24, 2022Assignee: STRATOS PERCEPTION LLCInventor: Rube Ben Williams
-
Patent number: 11322245Abstract: A medical image processing apparatus includes: a superimposed image generation unit configured to generate a superimposed image by superimposing a subject image and a fluorescent image in areas corresponding to each other; a determination unit configured to determine whether or not at least one of a subject and an observation device moves from timing before timing at which one of the subject image and the fluorescent image is captured to the timing; and a superimposition controller configured to cause the superimposition image generation unit to prohibit a superimposition in an area of at least a part of the subject image and the fluorescent image when the determination unit determines that at least one of the subject and the observation device moves.Type: GrantFiled: July 5, 2019Date of Patent: May 3, 2022Assignee: SONY OLYMPUS MEDICAL SOLUTIONS INC.Inventor: Taihei Michihata
-
Patent number: 11320508Abstract: The invention relates to a magnetic resonance imaging data processing system (126) for processing motion artifacts in magnetic resonance imaging data sets using a deep learning network (146, 502, 702) trained for the processing of motion artifacts in magnetic resonance imaging data sets. The magnetic resonance imaging data processing system (126) comprises a memory (134, 136) storing machine executable instructions (161, 164) and the trained deep learning network (146, 502, 702). Furthermore, the magnetic resonance imaging data processing system (126) comprises a processor (130) for controlling the magnetic resonance imaging data processing system.Type: GrantFiled: October 22, 2018Date of Patent: May 3, 2022Assignee: Koninklijke Philips N.V.Inventors: Karsten Sommer, Tom Brosch, Tim Philipp Harder, Jochen Keupp, Ingmar Graesslin, Rafael Wiemker, Axel Saalbach
-
Patent number: 11315255Abstract: The present disclosure includes methods and systems for identifying and manipulating a segment of a three-dimensional digital model based on soft classification of the three-dimensional digital model. In particular, one or more embodiments of the disclosed systems and methods identify a soft classification of a digital model and utilize the soft classification to tune segmentation algorithms. For example, the disclosed systems and methods can utilize a soft classification to select a segmentation algorithm from a plurality of segmentation algorithms, to combine segmentation parameters from a plurality of segmentation algorithms, and/or to identify input parameters for a segmentation algorithm. The disclosed systems and methods can utilize the tuned segmentation algorithms to accurately and efficiently identify a segment of a three-dimensional digital model.Type: GrantFiled: June 22, 2020Date of Patent: April 26, 2022Assignee: Adobe Inc.Inventors: Vladimir Kim, Aaron Hertzmann, Mehmet Yumer
-
Patent number: 11315273Abstract: The present invention discloses a disparity estimation method for weakly supervised trusted cost propagation, which utilizes a deep learning method to optimize the initial cost obtained by the traditional method. By combining and making full use of respective advantages, the problems of false matching and difficult matching of untextured regions in the traditional method are solved, and the method for weakly supervised trusted cost propagation avoids the problem of data label dependency of the deep learning method.Type: GrantFiled: March 5, 2020Date of Patent: April 26, 2022Assignees: DALIAN UNIVERSITY OF TECHNOLOGY, PENG CHENG LABORATORYInventors: Wei Zhong, Hong Zhang, Haojie Li, Zhihui Wang, Risheng Liu, Xin Fan, Zhongxuan Luo, Shengquan Li
-
Patent number: 11308638Abstract: A depth estimation method includes: taking a single image as a first image in binocular images, and obtaining a second image in the binocular images based on the first image via a first neural network; and obtaining depth information corresponding to the first image via a second neural network by performing binocular stereo matching on the first image and the second image.Type: GrantFiled: March 31, 2020Date of Patent: April 19, 2022Assignee: SHENZHEN SENSETIME TECHNOLOGY CO., LTD.Inventors: Yue Luo, Sijie Ren
-
Patent number: 11307361Abstract: Methods for detecting contaminants on optical fiber connectors incorporate statistically based routines that include processing of one or more digital images to dynamically determine fiber locations and boundaries, detect contamination in fiber locations, dynamically detect ferrule boundaries, detect contamination in ferrule locations, and collect resulting data. Sequential masking and detection permits detection of contaminants in a first region non-coincident with a first mask image, and in a second region non-coincident with a second mask image. Dynamic determination of locations and boundaries for structures such as fibers, ferrules, and pins or pinhole features address limitations associated with static determination of locations and boundaries for these structures. Specific routines or routines are provided for single- and multi-fiber connectors, including single fiber connectors having physical contact and angled physical contact polishes.Type: GrantFiled: May 2, 2019Date of Patent: April 19, 2022Assignee: Corning Research & Development CorporationInventor: Adam James Ruggles
-
Patent number: 11308637Abstract: A distance detection method, a distance detection system and a computer program product are provided. The method includes: comparing a street view image with a real-time image according to a first distance; determining a distance between a vehicle and a target position in response to the comparison result of the street view image and the real-time image; and outputting the distance between the vehicle and the target position to prompt a user of the vehicle.Type: GrantFiled: April 2, 2019Date of Patent: April 19, 2022Assignee: Wistron CorporationInventors: Chong-Wei Li, Chih-Pin Liao
-
Patent number: 11301999Abstract: A computer-implemented method of training a neural network for organ segmentation may be provided. The method may include: collecting a set of digital sample images from a database; inputting the collected set of digital images into a neural network recognition model; and training the neural network recognition model to recognize a first object in a first digital image as a specific object based on the first object being similar to a second object in a second digital image. The method may comprise predicting a signed distance map (SDM) in conjunction with a segmentation map.Type: GrantFiled: May 7, 2020Date of Patent: April 12, 2022Assignee: TENCENT AMERICA LLCInventors: Hui Tang, Chao Huang, Shih-Yao Lin, Zhen Qian, Wei Fan
-
Patent number: 11295412Abstract: An image processing apparatus applies an image to a first learning network model to optimize the edges of the image, applies the image to a second learning network model to optimize the texture of the image, and applies a first weight to the first image and a second weight to the second image based on information on the edge areas and the texture areas of the image to acquire an output image.Type: GrantFiled: April 2, 2020Date of Patent: April 5, 2022Assignee: SAMSUNG ELECTRONICS CO., LTD.Inventors: Cheon Lee, Donghyun Kim, Yongsup Park, Jaeyeon Park, Iljun Ahn, Hyunseung Lee, Taegyoung Ahn, Youngsu Moon, Tammy Lee
-
Patent number: 11278236Abstract: Methods and apparatuses are disclosed for assessing pigmentation of skin based on images thereof. In disclosed implementations, a cross-polarized blue image of skin is obtained and processed to extract pigment distribution information. Useful applications include assessing depigmentation, as in the skin condition vitiligo, as well as assessing re-pigmentation such as due to treatment. In addition to the spatial extent of depigmentation, implementations of the present disclosure can also provide the degree of depigmentation. The degree and area of depigmentation can be measured with better accuracy and sensitivity than known techniques.Type: GrantFiled: April 3, 2019Date of Patent: March 22, 2022Assignee: Canfield Scientific, IncorporatedInventor: Sachin V. Patwardhan
-
Patent number: 11270446Abstract: The present disclosure is related to systems and methods for image processing. The method includes obtaining a first image of a first modality. The method includes generating a second image of a second modality by processing, based on a trained machine learning model, the first image. The second modality may be different from the first modality.Type: GrantFiled: December 28, 2019Date of Patent: March 8, 2022Assignee: SHANGHAI UNITED IMAGING INTELLIGENCE CO., LTD.Inventors: Shu Liao, Yunhao Ge, Dongming Wei
-
Patent number: 11257219Abstract: Imaging systems and methods are provided, which involve acquiring static volume data using a first imaging technique; segmenting the static volume data to generate a static segmentation; annotating the static segmentation with at least one annotation; acquiring initial dynamic volume data using a second imaging technique different to the first imaging technique; segmenting the initial dynamic volume data to generate a plurality of dynamic segmentations; comparing the static segmentation to each one of the plurality of dynamic segmentations and determining, using the comparisons, a single dynamic segmentation that most closely corresponds to the static segmentation; storing the corresponding single dynamic segmentation in the memory as a reference segmentation; acquiring subsequent dynamic volume data; segmenting the subsequent dynamic volume data to generate at least one subsequent dynamic segmentation; determining a difference between the reference segmentation and the subsequent dynamic segmentation; updatiType: GrantFiled: November 29, 2018Date of Patent: February 22, 2022Assignee: KONINKLIJKE PHILIPS N.V.Inventors: Marco Verstege, Pieter Gerben Eshuis, Cherif Sahyoun