Patents Examined by Qian Yang
  • Patent number: 10529075
    Abstract: This disclosure relates generally to computer vision, and more particularly to method and system for tracking objects within a video. In one embodiment, a method for tracking objects within a video is disclosed. The method includes receiving one or more regions of interest (ROIs) corresponding to one or more objects in an initial frame of the video, extracting a set of scale and rotation invariant interest data points in each of the ROIs, clustering the set of scale and rotation invariant interest data points in a ROI into a set of clusters based on corresponding locations in the ROI, determining an optimal set of interest data points from each of the set of clusters based on corresponding feature response values and spread values, and initiating tracking of the optimal set of interest data points in subsequent frames of the video to track the one or more objects in the video.
    Type: Grant
    Filed: March 15, 2018
    Date of Patent: January 7, 2020
    Assignee: Wipro Limited
    Inventor: Ghulam Mohiuddin Khan
  • Patent number: 10528845
    Abstract: Method, device, and storage medium for generating a GIF file. The method for generating a GIF file includes: obtaining an image set including a plurality of images; selecting at least two images from the image set, content of each of the at least two images satisfying a preset similarity condition, and a difference between photographing times of any two of the at least two images not exceeding a preset time threshold; and generating a GIF file according to the at least two images.
    Type: Grant
    Filed: December 15, 2017
    Date of Patent: January 7, 2020
    Assignee: TENCENT TECHNOLOGY (SHENZHEN) COMPANY LIMITED
    Inventor: Yuyu Zheng
  • Patent number: 10521663
    Abstract: Herein is disclosed an image location iteration system comprising one or more processors configured to receive from one or more unmanned aerial vehicles a plurality of images and corresponding detected positions; determine an alignment of the plurality of images according to one or more image features; calculate a first set of measured locations for the plurality of images according to the alignment and the detected positions; and calculate a second set of measured locations the plurality of images according to the alignment and a first subset of the detected positions.
    Type: Grant
    Filed: December 28, 2017
    Date of Patent: December 31, 2019
    Assignee: Intel IP Corporation
    Inventors: Cornelius Claussen, Marco Moeller
  • Patent number: 10521890
    Abstract: A method and device for detecting image artifacts. Pixel transition values of camera images are computed in a first direction by comparing intensity differences between neighboring pixels with an upper threshold and a lower threshold and setting a corresponding pixel transition value. A low pass filter with respect to time and a spatial filter are applied to the pixel transition values. Pixels of potential artifact regions are identified by comparing magnitude values of the smoothed pixel transition values with a predetermined detection threshold.
    Type: Grant
    Filed: December 18, 2017
    Date of Patent: December 31, 2019
    Assignee: Continental Automotive GmbH
    Inventor: Rui Guerreiro
  • Patent number: 10509942
    Abstract: The present disclosure discloses a fingerprint identification device, a fingerprint identification method and a touch screen. The fingerprint identification device comprises: an optical sensor with a receiving surface configured to receive fingerprint reflected light; and a lens assembly configured to converge the fingerprint reflected light toward the receiving surface of the optical sensor, wherein the lens assembly includes a liquid crystal lens configured to converge the fingerprint reflected light toward the receiving surface of the optical sensor by rotating liquid crystal molecules at different angles upon energization.
    Type: Grant
    Filed: September 5, 2017
    Date of Patent: December 17, 2019
    Assignee: BOE TECHNOLOGY GROUP CO., LTD.
    Inventors: Haisheng Wang, Lijun Zhao, Yingming Liu, Rui Xu, Changfeng Li, Yanan Jia, Yuzhen Guo, Pinchao Gu, Yunke Qin
  • Patent number: 10509962
    Abstract: Camera platform techniques are described. In an implementation, a plurality of digital images and data describing times, at which, the plurality of digital images are captured is received by a computing device. Objects of clothing are recognized from the digital images by the computing device using object recognition as part of machine learning. A user schedule is also received by the computing device that describes user appointments and times, at which, the appointments are scheduled. A user profile is generated by the computing device by training a model using machine learning based on the recognized objects of clothing, times at which corresponding digital images are captured, and the user schedule. From the user profile, a recommendation is generated by processing a subsequent user schedule using the model as part of machine learning by the computing device.
    Type: Grant
    Filed: December 29, 2017
    Date of Patent: December 17, 2019
    Assignee: eBay Inc.
    Inventors: Shuai Zheng, Fan Yang, Mohammadhadi Kiapour, Qiaosong Wang, Japjit S. Tulsi, Robinson Piramuthu
  • Patent number: 10503832
    Abstract: Systems and methods provided herein for a determining a meaning of an ambiguous term in a text segment based on a context term, a static knowledge graph and a temporal knowledge graph. These systems and methods access a first knowledge graph associated with the context term to determine a potential term that is the meaning of the unknown term. Upon determining that there are multiple potential terms in the first knowledge graph that could be the meaning of the unknown term, the systems and methods take into account the temporal dimension of relationships between entities for disambiguating the meaning of the unknown term. The systems and methods achieve this by determining a time stamp of the text segment and accessing a second knowledge graph associated with the first context term and related to the time stamp to determine the potential term that is the meaning of the first unknown term.
    Type: Grant
    Filed: July 29, 2016
    Date of Patent: December 10, 2019
    Assignee: Rovi Guides, Inc.
    Inventors: Manik Malhotra, Sashikumar Venkataraman, Ahmed Nizam Mohaideen P.
  • Patent number: 10499865
    Abstract: A system for measuring muscle mass of a patient has a dual-energy radiation emission source. A radiation detector is configured to detect radiation emitted from the dual-energy radiation emission source passed through the patient. A processor has a memory with storing instructions, that when executed by the processor, perform a set of operations. The operations include receiving radiation detection data; generating a scan representation; identifying a primary fat target; determining an amount of fat in the primary fat target; comparing the amount of fat in the primary fat target to a reference; and based on the comparison, correcting an estimated amount of lean tissue to generate a corrected muscle mass value.
    Type: Grant
    Filed: February 25, 2016
    Date of Patent: December 10, 2019
    Assignee: Hologic, Inc.
    Inventors: Kevin Wilson, Thomas Kelly
  • Patent number: 10489921
    Abstract: Provided is a technology for preventing a frame-out of a person from an image acquired through a camera in a behavior analysis technology using the acquired image. A behavior analysis apparatus includes: an image acquisition unit; a person image extraction unit; a person skeleton detection unit; a person behavior analysis unit; a frame-out determination unit configured to determine a frame-out of the person toward an outside of the predetermined field angle as to one of whether or not the frame-out has occurred and whether or not the frame-out is predictable through use of any one of the person image area and the person skeleton; and a frame-out notification unit configured to notify the person of the determination in one of a case where the frame-out has occurred and a case where the frame-out is predictable.
    Type: Grant
    Filed: March 30, 2017
    Date of Patent: November 26, 2019
    Assignee: Hitachi, Ltd.
    Inventors: Toshio Kamimura, Yuuichi Nonaka, Akinobu Watanabe, Yasuyuki Mimatsu
  • Patent number: 10483006
    Abstract: A computer-implemented method for providing a personalized evaluation of assessment of atherosclerotic plaques for a patient acquiring patient data comprising non-invasive patient data, medical images of the patient, and blood biomarkers. Features of interest are extracted from the patient data and one or more machine learning models are applied to the features of interest to predict one or more measures of interest related to atherosclerotic plaque.
    Type: Grant
    Filed: May 19, 2017
    Date of Patent: November 19, 2019
    Assignee: Siemens Healthcare GmbH
    Inventors: Lucian Mihai Itu, Tiziano Passerini, Saikiran Rapaka, Puneet Sharma, Dorin Comaniciu
  • Patent number: 10475184
    Abstract: From a plurality of medical images in time phases, a target site is extracted t least one medical image, a reference point is set on each of a target-site side, and a periphery side of the target site which are on across from each other over an outline of the extracted target site, and movement information for the reference points is calculated.
    Type: Grant
    Filed: September 9, 2015
    Date of Patent: November 12, 2019
    Assignee: CANON KABUSHIKI KAISHA
    Inventors: Ryo Sakamoto, Koji Sakai, Gakuto Aoyama, Kiyohide Satoh
  • Patent number: 10460152
    Abstract: Proposed are a method and apparatus for pushing information, the method comprises: detecting face information, and acquiring control information; acquiring pushing information according to the face information and the control information; and presenting the pushing information.
    Type: Grant
    Filed: June 17, 2015
    Date of Patent: October 29, 2019
    Assignee: BAIDU ONLINE NETWORK TECHNOLOGY (BEIJING) CO., LTD.
    Inventors: Yang Li, Jiawei Gu, Kai Yu
  • Patent number: 10460193
    Abstract: An image processing apparatus approximates a three-dimensional shape of a target object, acquired from an image of of the target object, by a quadric surface, applies a plane expansion process that expands the quadric surface into an expansion plane, to data of the image, computes corresponding points before and after the plane expansion process, and generates normalized data of the image from the corresponding points that are computed. The corresponding points are computed based on a point on a straight line of the expansion plane in contact with a point on the quadric surface and corresponding to a reference point on the target object, and an intersection point of an imaginary straight line, passing through the point on the straight line and extending in a direction of a first normal vector with respect to the straight line, and a quadratic curve of the quadric surface.
    Type: Grant
    Filed: November 6, 2017
    Date of Patent: October 29, 2019
    Assignee: FUJITSU LIMITED
    Inventor: Takahiro Aoki
  • Patent number: 10460436
    Abstract: Disclosed are an inspection method, an inspection system, and a method of fabricating a semiconductor package using the same. The inspection method comprises obtaining a reference value by measuring a surface profile of a reference pattern, scanning reference images of the reference pattern by using a plurality of optical inspection conditions, obtaining estimation values of the reference pattern that are measured from the reference images, selecting an desired optical inspection condition among the plurality of optical inspection conditions by comparing the reference value with the estimation values, scanning a target image of a target pattern by using the desired optical inspection condition, and obtaining an error value by quantitatively comparing the target image with a design image of the target pattern.
    Type: Grant
    Filed: August 31, 2017
    Date of Patent: October 29, 2019
    Assignee: Samsung Electronics Co., Ltd.
    Inventors: Younghoon Sohn, Ilsoo Kim, Yusin Yang
  • Patent number: 10441438
    Abstract: A method according to one embodiment includes locating distal lateral and medial condyles and posterior lateral and medial condyles a femoral bone model, and determining a posterior resection plane based on locations of the distal lateral and medial condyles, one or more dimensions of a prospective femoral implant, and one or more surgical preferences. The method may further include determining a first distance between the posterior medial condyle and the posterior resection plane and a second distance between the posterior lateral condyle and the posterior resection plane, determining a difference between a target value and an average of the first distance and the second distance, wherein the target value is associated with a thickness of the prospective femoral implant, and selecting a desired femoral implant size based on the difference between the target value and the average determined for each of a plurality of possible femoral implant sizes.
    Type: Grant
    Filed: August 25, 2017
    Date of Patent: October 15, 2019
    Assignee: Smith & Nephew, Inc.
    Inventors: Mohammed Rahman, Yangqiu Hu, Kyle Lonidier
  • Patent number: 10445877
    Abstract: A computer-implemented method for determining farm boundary delineations within a target geographic area, comprising extracting data from pixels of a satellite image of the target geographic area, evaluating the data using a classification algorithm to generate one or more line segments between adjacent pixels, the one or more line segments being representative of a portion of a boundary delineation, connecting the one or more line segments to an adjacent line segment to form a boundary delineation defining at least one parcel of land within the target geographic area, and generating a boundary delineation map including the boundary delineation.
    Type: Grant
    Filed: December 30, 2016
    Date of Patent: October 15, 2019
    Assignee: International Business Machines Corporation
    Inventors: Conrad M. Albrecht, Hendrik F. Hamann, Levente Klein, Siyuan Lu, Fernando J. Marianno
  • Patent number: 10444362
    Abstract: Systems and processes for increasing the effective sampling density of a LADAR data set are disclosed. LADAR data points are merged with data regarding edges of objects within the physical space represented by the LADAR data points to form a merged LADAR-edge point cloud. Each data point within the merged LADAR-edge point cloud is examined to identify co-planar neighboring data points within a defined search area. Additional data points are added to the LADAR-edge point cloud by interpolating between the identified, co-planar neighboring data points.
    Type: Grant
    Filed: January 13, 2015
    Date of Patent: October 15, 2019
    Assignee: Raytheon Company
    Inventor: Matthew A. Schaefer
  • Patent number: 10434555
    Abstract: Systems and methods for using metrology to assist a user at a workstation to form a part into a desired contour include scanning the part to obtain scanned data indicative of an actual contour of the part. Distance errors are determined based on a comparison of the scanned data to a computer model of the desired contour. A contour map of deviation is determined based on the distance errors, with the contour map of deviation indicating magnitudes of the distance errors for at least a selected set of points on the actual contour of the part. Visible indicium is projected onto the part that represents the contour map of deviation, thereby assisting the user of the workstation to identify where further modification of the actual contour is needed.
    Type: Grant
    Filed: December 11, 2015
    Date of Patent: October 8, 2019
    Assignee: The Boeing Company
    Inventors: Brenda C. Gross, John W. Dorsey-Palmateer
  • Patent number: 10438378
    Abstract: The present disclosure relates to systems and methods for determining a target activity map and a target attenuation map for producing a PET image. The systems may execute the methods to acquire, based on a PET system, a first dataset relating to coincidence events with TOF information, and a second dataset relating to single events. The systems may also execute the methods to determine a target activity map and a target attenuation map based on the first dataset and the second dataset through a plurality of iterations. The systems may further execute the methods to generate the PET image based on the target activity map and the target attenuation map.
    Type: Grant
    Filed: August 25, 2017
    Date of Patent: October 8, 2019
    Assignee: UIH AMERICA, INC.
    Inventors: Tao Feng, Jizhe Wang, Wentao Zhu, Hongdi Li
  • Patent number: 10430951
    Abstract: The present disclosure discloses a method and device for straight line detection and image processing. The straight line detection method includes: dividing a horizontal axis and a vertical axis of a straight line parameter space equally, so as to divide the straight line parameter space into a plurality of parameter areas; voting for the plurality of parameter areas utilizing a coordinate of each sample pixel to obtain a vote amount of each of the parameter areas; extracting a straight line parameter and the vote amount of each of the parameter areas having the vote amount larger than a voting threshold, and grouping the straight line parameters into a group; and weighting and averaging the straight line parameter of each group and the vote amount respectively to obtain the straight line parameter of a detected straight line. The present disclosure also discloses a robot and a numerical control machine.
    Type: Grant
    Filed: November 13, 2015
    Date of Patent: October 1, 2019
    Assignee: BEIJING A&E TECHNOLOGIES CO., LTD.
    Inventor: Li Wang