Patents Examined by Matthew C. Bella
  • Patent number: 10733482
    Abstract: Systems and methods for estimating a height of an object from a monocular image are described herein. Objects are detected in the image, each object being indicated by a region of interest. The image is then cropped for each region of interest and the cropped image scaled to a predetermined size. The cropped and scaled image is then input into a convolutional neural network (CNN), the output of which is an estimated height for the object. The height may be represented by a mean of a probability distribution of possible sizes, a standard deviation, as well as a level of confidence. A location of the object may be determined based on the estimated height and region of interest. A ground truth dataset may be generated for training the CNN by simultaneously capturing a LIDAR sequence with a monocular image sequence.
    Type: Grant
    Filed: March 8, 2017
    Date of Patent: August 4, 2020
    Assignee: Zoox, Inc.
    Inventors: Tencia Lee, James William Vaisey Philbin
  • Patent number: 10733431
    Abstract: In one embodiment, a system may access first, second, and third probability models that are respectively associated with predetermined first and second body parts and a predetermined segment connecting the first and second body parts. Each model includes probability values associated with regions in an image, with each value representing the probability of the associated region containing the associated body part or segment. The system may select a first and second region based on the first probability model and a third region based on the second probability model. Based on the third probability model, the system may compute a first probability score for regions connecting the first and third regions and a second probability score for regions connecting the second and third regions. Based on the first and second probability scores, the system may select the first region to indicate where the predetermined first body part appears in the image.
    Type: Grant
    Filed: December 31, 2018
    Date of Patent: August 4, 2020
    Assignee: Facebook, Inc.
    Inventors: Peizhao Zhang, Peter Vajda, Kevin Matzen, Ross Girshick
  • Patent number: 10726236
    Abstract: Embodiments of the present disclosure provide a display panel, a display apparatus and a method for identifying fingerprints. The display panel comprises a first number of pixel units, arranged in a first direction and a second direction perpendicular to the first direction, and each pixel unit comprises a display element and a sensor configured to sense and identify a fingerprint. The display panel may further comprise a receiving unit, configured to receive a fingerprint signal sensed by the sensor; a controlling unit, configured to control the receiving unit to receive the fingerprint signal sensed by a second number of sensors among a first number of sensors, in response to an instruction for fingerprint identification being received, wherein any two of the second number of sensors are not adjacent to each other in the first direction and the second direction.
    Type: Grant
    Filed: July 3, 2017
    Date of Patent: July 28, 2020
    Inventors: Changfeng Li, Xue Dong, Xiaochuan Chen, Haisheng Wang, Yingming Liu, Xiaoliang Ding, Shengji Yang, Weijie Zhao, Wei Liu, Pengpeng Wang
  • Patent number: 10721376
    Abstract: A system to identify vision sensor impairment, the system including: a memory with one or more executable instructions; a controller to execute the executable instructions; a vision sensor located on a vehicle to capture at least one image of at least a portion of an environment surrounding the vehicle; where the executable instructions enable the controller to: receive a request communicated from a mobile computing device; in response to the request, communicate an activation signal to the vision sensor, the activation signal to activate the vision sensor to capture an image of the environment surrounding the vehicle; receive the captured image from the vision sensor; determine whether the view of the captured image includes a substantially impairing weather debris object; and based on this determination, generate a notification and communicate the notification to the mobile computing device.
    Type: Grant
    Filed: April 24, 2018
    Date of Patent: July 21, 2020
    Inventor: Kunal Jaggi
  • Patent number: 10706501
    Abstract: The present disclosure provides a method and an apparatus for stretching an image. The method for stretching an image includes: selecting a corresponding stretching mode according to a mode selection parameter; generating a corresponding stretching filter group according to a stretching parameter and the selected stretching mode, and segmenting input image data into blocks; and processing the input image data segmented into blocks by the stretching filter group, to obtain stretched image data.
    Type: Grant
    Filed: March 1, 2018
    Date of Patent: July 7, 2020
    Inventors: Xiaoyu Li, Pablo Navarrete Michelini, Ran Duan, Lihua Geng, Xitong Ma
  • Patent number: 10697887
    Abstract: A measuring apparatus that measures an optical characteristic of a surface in accordance with a standard selected from a plurality of standards is provided. The apparatus includes an illumination device configured to illuminate the surface with light from a light source, an imaging device configured to image the light source with reflected light from the surface illuminated by the illumination device, and a processor configured to process image data having number of pixels obtained by reducing number of pixels of the imaging device by a reduction rate to obtain the optical characteristic of the surface. The processor is configured to determine the reduction rate based on the selected standard.
    Type: Grant
    Filed: July 13, 2016
    Date of Patent: June 30, 2020
    Inventors: Yusuke Kasai, Hiroshi Kawanago, Takashi Seki
  • Patent number: 10692231
    Abstract: Described are systems and methods for determining an agent that performed an event within a materials handling facility. When an event is detected, an overhead image and a side-view image corresponding to the event location are obtained. Depth information is determined for the overhead image and a depth pattern generated that represents an agent. The depth pattern from the overhead image is overlaid onto the side-view image and the side-view image is segmented so that a representation of the agent included in the side-view image is extracted. A distance between the extracted representation of the agent and the event location is then determined. An association between the agent and the event may be determined based on the distance between the extracted representation and the event location.
    Type: Grant
    Filed: August 14, 2019
    Date of Patent: June 23, 2020
    Assignee: Amazon Technologies, Inc.
    Inventors: Gang Hua, Gerard Guy Medioni
  • Patent number: 10686962
    Abstract: A graphical indicator comprising a plurality of first header blocks, a plurality of second header blocks and a plurality of data blocks for forming an indicator matrix is provided. Each of the first and second header blocks has a header graphical micro-unit, and each of the data blocks has a data graphical micro-unit. An array area is formed by the second header blocks and the data blocks. A first virtual line and a second virtual line are respectively formed by virtual centers of the first and second header blocks, and an included angle between the first and second virtual lines is less than 90 degrees.
    Type: Grant
    Filed: February 23, 2018
    Date of Patent: June 16, 2020
    Inventors: Cheng-Ta Lee, Jang-Jer Tsai, Tzu-Fong Huang
  • Patent number: 10679355
    Abstract: Described is a system for detecting moving objects. During operation, the system obtains ego-motion velocity data of a moving platform and generates a predicted image of a scene proximate the moving platform by projecting three-dimensional (3D) data into an image plane based on pixel values of the scene. A contrast image is generated based on a difference between the predicted image and an actual image taken at a next step in time. Next, an actionable prediction map is then generated based on the contrast mage. Finally, one or more moving objects may be detected based on the actionable prediction map.
    Type: Grant
    Filed: April 23, 2018
    Date of Patent: June 9, 2020
    Assignee: HRL Laboratories, LLC
    Inventors: Kyungnam Kim, Hyukseong Kwon, Heiko Hoffmann
  • Patent number: 10664706
    Abstract: The present disclosure provides for a method, device, and computer-readable storage medium for performing a method for discerning a vehicle at an access control point. The method including obtaining a video sequence of the access control point; detecting an object of interest from the video sequence; tracking the object from the video sequence to obtain tracked-object data; classifying the object to obtain classified-object data; determining that the object is a vehicle based on the classified-object data; and determining that the vehicle is present in a predetermined detection zone based on the tracked-object data.
    Type: Grant
    Filed: January 10, 2019
    Date of Patent: May 26, 2020
    Inventors: Don Madden, Ethan Shayne
  • Patent number: 10657382
    Abstract: The various embodiments described herein include methods, devices, and systems for providing event alerts. In one aspect, a method includes obtaining a video feed. A frame of the video feed is analyzed at a first resolution to determine whether the frame includes a potential instance of a person. In accordance with the determination that the image includes the potential instance, a region is denoted around the potential instance. The region is analyzed at a second resolution, greater than the first resolution. In accordance with a determination that the region includes the instance of the person. a determination that the frame includes the person is made. An indication of the determination is stored for use in subsequent alert notification processing.
    Type: Grant
    Filed: July 2, 2019
    Date of Patent: May 19, 2020
    Assignee: GOOGLE LLC
    Inventors: Rizwan Ahmed Chaudhry, Navneet Dalal, Jonathan Z. Ben-Meshulam, George Alban Heitz, III
  • Patent number: 10657644
    Abstract: A hybrid detection model may be used for breast cancer detection. A system can identify a region of interest on a received mammogram. The region of interest may have a particular level of grey at each pixel of the image. A morphological and entropy filter may then each be applied to the region of interest. Based on the filters, the system may generate a hybrid result that is the average of a combination of the two filters. The system may then segment the region of interest using a selected clustering algorithm.
    Type: Grant
    Filed: July 19, 2019
    Date of Patent: May 19, 2020
    Assignee: International Business Machines Corporation
    Inventors: Juan Manuel A. Vega, Ramón Oswaldo G. Medina, Rubén Ruelas-Lepe
  • Patent number: 10650271
    Abstract: An image processing apparatus includes a processor configured to generate vertical direction distribution data which is a distribution of distance values to a vertical direction of a distance image from the distance image having distance values according to distance of a road surface in captured images. There is an extraction of a pixel having the highest frequency value for each of the distance values from pixels in a search area. There is a detection of the road surface based on each pixel extracted.
    Type: Grant
    Filed: March 8, 2017
    Date of Patent: May 12, 2020
    Assignee: RICOH COMPANY, LTD.
    Inventors: Otoichi Nakata, Naoki Motohashi, Shinichi Sumiyoshi
  • Patent number: 10643090
    Abstract: The present disclosure relates to the technical field of computer visual detection processing, and provides a method for guiding users to restore Rubik's cubes. The method comprises step 1, acquiring arrangement information of colored blocks of the Rubik's cube; step 2, displaying restoration process of the Rubik's cube.
    Type: Grant
    Filed: December 19, 2017
    Date of Patent: May 5, 2020
    Inventors: Letian Wang, Peng Du, Qingyun Li, Huan Chen, Mingying Gu
  • Patent number: 10631813
    Abstract: A multi-pouch container or carrier for imaging plates which includes multiple pockets which are overlapped, in which multiple relatively small plates can be inserted, sequenced and aligned in a defined overlapping such that simultaneously exposed plates in the pockets form overlapped images which can be stitched together by application software to provide a larger variable sized single contiguous image of a region of interest.
    Type: Grant
    Filed: June 1, 2019
    Date of Patent: April 28, 2020
    Assignee: APIXIA INC
    Inventors: Jerry Huang, Peter Chen, Ta-Ko Huang
  • Patent number: 10631778
    Abstract: A method for patient setup includes: obtaining a first two-dimensional movie having a first plurality of image frames; obtaining a first plurality of two-dimensional images; and displaying the first plurality of two-dimensional images together with the image frames from the first two-dimensional movie in an overlay configuration and in a synchronized manner, wherein the act of displaying is performed during a patient setup procedure. A method for patient setup includes: obtaining a plurality of three-dimensional images; obtaining a first two-dimensional movie having a first plurality of image frames; determining a first plurality of two-dimensional images from the plurality of three-dimensional images; and displaying the first plurality of two-dimensional images together with the image frames from the first two-dimensional movie in an overlay configuration and in a synchronized manner, wherein the act of displaying is performed during a patient setup procedure.
    Type: Grant
    Filed: October 1, 2015
    Date of Patent: April 28, 2020
    Assignee: Varian Medical Systems International AG
    Inventors: David Kleiner, Patrik Kunz, Michael Huber
  • Patent number: 10621720
    Abstract: A computer-implemented method for performing deformable registration between Magnetic Resonance (MR) and Ultrasound (US) images include receiving an MR volume depicting an organ and segmenting the organ from the MR volume to yield a first 3D point representation of the organ in MR coordinates. Additionally, a US volume depicting an organ is received and the organ is segmented from the US volume to yield a second 3D point representation of the organ in US coordinates. Next, a plurality of point correspondences between the first 3D point representation and the second 3D point representation are determined. Then, a biomechanical model is applied to register the MR volume to the US volume. The plurality of point correspondences are used as displacement boundary conditions for the biomechanical model.
    Type: Grant
    Filed: April 27, 2017
    Date of Patent: April 14, 2020
    Assignee: Siemens Healthcare GmbH
    Inventors: Thomas Pheiffer, Ankur Kapoor, Jin-hyeong Park, Ali Kamen
  • Patent number: 10621594
    Abstract: Apparatuses and methods associated with multi-level authentication are disclosed herein. In embodiments, a method includes authenticating a physical object of a plurality of physical objects that together form an aggregate physical object; storing in a database system relationship information reflecting a relationship between the aggregate physical object and the plurality of physical objects; attempting to authenticate a target physical object; responsive to matching the target physical object to the aggregate physical object based on the attempt to authenticate the physical target: identifying in the database system a database record corresponding to the aggregate physical object; storing in the database record authentication data reflecting the match between the target physical object and the aggregate physical object; and storing an indication of a re-authentication of the physical object in the database system based on the relationship information. Other embodiments may be disclosed or claimed.
    Type: Grant
    Filed: February 17, 2017
    Date of Patent: April 14, 2020
    Assignee: ALITHEON, INC.
    Inventors: Scot E. Land, David Keesu Kim, David Justin Ross, Mark Tocci, Robert Saxon Ross
  • Patent number: 10599926
    Abstract: Pixel color values representing an image of a portion of a field are received where each pixel color value has a respective position within the image. A processor identifies groups of the received pixel color values as possibly representing a Nitrogen-deficient plant leaf. For each group of pixel color values, the processor converts the pixel color values into feature values that describe a shape and the processor uses the feature values describing the shape to determine whether the group of pixel color values represents a Nitrogen-deficient leaf of a plant. The processor stores in memory an indication that the portion of the field is deficient in Nitrogen based on the groups of pixel color values determined to represent a respective Nitrogen-deficient leaf.
    Type: Grant
    Filed: December 15, 2016
    Date of Patent: March 24, 2020
    Assignee: Regents of the University of Minnesota
    Inventors: Nikolaos Papanikolopoulos, Vassilios Morellas, Dimitris Zermas, David Mulla, Michael Bazakos, Daniel Kaiser
  • Patent number: 10586350
    Abstract: In one embodiment, a system accesses pose probability models for predetermined parts of a body depicted in an image. Each of the pose probability models is configured for determining a probability of the associated predetermined body part being at a location in the image. The system determines a candidate pose that is defined by a set of coordinates representing candidate locations of the predetermined body parts. The system further determines a first probability score for the candidate pose based on the pose probability models and the set of coordinates of the candidate pose. A pose representation is generated for the candidate pose using a transformation model and the candidate pose. The system determines a second probability score for the pose representation based on a pose-representation probability model. The system selects the candidate pose to represent a pose of the body based on at least the first and second probability scores.
    Type: Grant
    Filed: May 4, 2018
    Date of Patent: March 10, 2020
    Assignee: Facebook, Inc.
    Inventors: Peter Vajda, Peizhao Zhang, Fei Yang, Yanghan Wang