Patents Examined by Jingge Wu
  • Patent number: 10650243
    Abstract: In the image processing device, the method and the recording medium according to the present invention, the characteristic information extractor extracts the number of times the same subject appears in the still images. The target object detector detects the target subject which appears the number of times not less than the threshold value. The image analysis condition determiner sets the image analysis condition for the portion of the moving image where the target subject is absent to be rougher than that where the target subject is present. The frame image analyzer extracts frame images from the moving image in accordance with the image analysis condition. The frame image output section calculates the evaluation value of the frame images based on the result of the image analysis and output the frame image having the evaluation value not less than the threshold value.
    Type: Grant
    Filed: January 18, 2018
    Date of Patent: May 12, 2020
    Assignee: FUJIFILM Corporation
    Inventor: Kei Yamaji
  • Patent number: 10648921
    Abstract: An image inspection apparatus includes: an imaging section for capturing an image of a workpiece from a certain direction; an illumination section for illuminating the workpiece from different directions at least three times; an illumination controlling section for sequentially turning on the illumination sections one by one; an imaging generating section for driving the imaging section to generate a plurality of images; a normal vector calculating section for calculating a normal vector with respect to the surface of the workpiece at each of pixels by use of a pixel value of each of pixels having a corresponding relation among the plurality of images; and a contour image generating section for performing differential processing in an X-direction and a Y-direction on the calculated normal vector at each of the pixels, to generate a contour image that shows a contour of inclination of the surface of the workpiece.
    Type: Grant
    Filed: October 22, 2018
    Date of Patent: May 12, 2020
    Assignee: Keyence Corporation
    Inventor: Zhuoli Sun
  • Patent number: 10646127
    Abstract: A blood flow analysis device includes a blood vessel region extraction unit that extracts a blood vessel region from a three-dimensional medical image of a subject including a blood vessel, and a blood flow simulation unit that obtains a line resistance by multiplying a resistance value corresponding to a branch blood vessel branching out of the blood vessel region by a surface area of each of preset sections into which the blood vessel region is divided in an extension direction, generates a vascular network model by applying the line resistance to each node of a surface of the blood vessel region, and performs simulation of a blood flow by using the vascular network model.
    Type: Grant
    Filed: February 26, 2018
    Date of Patent: May 12, 2020
    Assignee: FUJIFILM Corporation
    Inventors: Sho Onozawa, Takeshi Unemura, Hirotaka Ito
  • Patent number: 10643328
    Abstract: The present invention is to provide a computer system, and a method and a program for diagnosing an object that is capable to improve the diagnosis accuracy of an object. The computer system acquires a visible light image of an object that is taken by a camera, analyzes the acquired visible light image by comparing the acquired visible light image with a normal visible light image of the object, identifies the type of the object based on the result of the image analysis, identifies an abnormal part of the object based on the result of the image analysis, acquires environment data of the object, and diagnoses the status of the object based on the identified type, the identified abnormal part, and the acquired environment data.
    Type: Grant
    Filed: October 31, 2016
    Date of Patent: May 5, 2020
    Inventor: Shunji Sugaya
  • Patent number: 10645338
    Abstract: A virtual encounter system includes a mannequin coupled to a camera for receiving a video image. The camera sends the video image to a communications network. The virtual encounter system also includes a processor for overlaying a virtual environment over one or more portions of the video image to form a virtual scene and a set of goggles to render the virtual scene.
    Type: Grant
    Filed: April 10, 2018
    Date of Patent: May 5, 2020
    Assignee: Beyond Imagination Inc.
    Inventor: Raymond C. Kurzweil
  • Patent number: 10631730
    Abstract: Methods and systems for suppressing shadowgraphic flow projection artifacts in OCT angiography images of a sample are disclosed. In one example approach, normalized OCT angiography data is analyzed at the level of individual A-scans to classify signals as either flow or projection artifact. This classification information is then used to suppress projection artifacts in the three dimensional OCT angiography dataset.
    Type: Grant
    Filed: February 4, 2019
    Date of Patent: April 28, 2020
    Assignee: Oregon Health & Science University
    Inventors: David Huang, Yali Jia, Miao Zhang
  • Patent number: 10634484
    Abstract: There is provided a fast and a reliable way to inspect a condition of a chain by an arrangement comprising a camera and guide for positioning a chain in front of the camera such that holes of consecutive chain links are visible to the camera. In a method for inspecting a chain connected movable to a hoisting device, the chain is positioned in front of the camera by the guide and the chain is driven past the camera by the hoisting device.
    Type: Grant
    Filed: July 5, 2016
    Date of Patent: April 28, 2020
    Inventors: Jack Hoover, Lasse Eriksson, Teemu Puustinen, Risto Vilppo, Kimmo Rantala, Juhani Kerovuori, Stefan Hecht
  • Patent number: 10638089
    Abstract: A network apparatus system and method of use adapted for viewing, illustrating, consulting and collaborating on medical images and saving images and illustrations in acceptable DICOM format. Specifically, a Consultant Client or a plurality of Consultant Clients having the capability to view still and streaming video medical images, then illustrating over those images with the ability to save those images. In the present invention the illustration images automatically fading and disappearing after a preset predetermined amount of time, without user input. Also a Collaborator Client, a person user or a plurality of Collaborator Clients having the ability to remotely view streaming or still medical images and illustrating over those images. The Collaborator Clients can draw, telestrate and annotate over those images with the ability to save those images and videos, and save the drawings, telestrations and annotations provided these input illustrations are visible when saved.
    Type: Grant
    Filed: December 11, 2017
    Date of Patent: April 28, 2020
    Assignee: Foresight Imaging
    Inventors: Ralf Berger, Steven M. Troppoli, Mark Mariotti
  • Patent number: 10638060
    Abstract: Techniques are provided for color correction of image frames, received from a hybrid RGBIR sensor, using resolution recovery of the red, green, blue (RGB), and infra-red (IR) channels. The color correction is to compensate for IR contamination of the RGB channels. A methodology implementing the techniques according to an embodiment includes a demosaic operation on the RGBIR image frame to generate full resolution red, green, blue, and IR frames. The demosaic operation includes filtering to extract Chroma and Luma components of the RGBIR image to reconstruct the full resolution images. The method also includes calculating an IR weighting factor for the pixels of each of the full resolution RGB frames, and correcting for IR contamination of those pixels by subtracting a scaled value of the corresponding pixels from the full resolution IR frame. The scaled value is based on the IR weighting factor associated with the pixel to be corrected.
    Type: Grant
    Filed: June 28, 2016
    Date of Patent: April 28, 2020
    Inventors: Hsin-I Liu, Yongshen Ni
  • Patent number: 10621727
    Abstract: Systems of the present disclosure allow fields and labels to be identified in a digital image of a form without performing OCR. A digital image of a form can be partitioned into image segments using computer-vision image-segmentation techniques. Features for each image segment can be extracted using computer-vision feature-detection methods. The features extracted from an image segment can be included in an input instance for a machine-learning model. The machine-learning model can assign a classification to the input instance. The classification can associate the input instance with a field type or a label type.
    Type: Grant
    Filed: April 24, 2018
    Date of Patent: April 14, 2020
    Assignee: INTUIT INC.
    Inventors: Richard J. Becker, Pallavika Ramaswamy, Daniel L. Moise, Sheldon Porcina
  • Patent number: 10614551
    Abstract: Image interpolation methods for interpolating a destination pixel of a destination image by using a plurality of source pixels of a source image are provided. The method includes the steps of: selecting the plurality of source pixels from the source image corresponding to the destination pixel; obtaining a gradient direction and a gradient magnitude according to the plurality o source pixels; determining a weight corresponding to each of the plurality of source pixels according to the gradient direction and the gradient magnitude; and performing an interpolation operation according to pixel values of the plurality of source pixels and the weight corresponding to each of the plurality of source pixels to obtain a pixel value of the destination pixel.
    Type: Grant
    Filed: November 14, 2017
    Date of Patent: April 7, 2020
    Inventor: Feifei Yang
  • Patent number: 10616467
    Abstract: An imaging device including a pixel matrix and a processor is provided. The pixel matrix includes a plurality of phase detection pixels and a plurality of regular pixels. The processor performs autofocusing according to pixel data of the phase detection pixels, and determines an operating resolution of the regular pixels according to autofocused pixel data of the phase detection pixels, wherein the phase detection pixels are always-on pixels and the regular pixels are selectively turned on after the autofocusing is accomplished.
    Type: Grant
    Filed: October 30, 2018
    Date of Patent: April 7, 2020
    Inventors: Guo-Zhen Wang, Ren-Hau Gu, En-Feng Hsu
  • Patent number: 10614306
    Abstract: A method, system and computer program product are disclosed for tagging a resource. In one embodiment, the method comprises receiving a given number of unique electronic tags for tagging a specified resource; for a harvested one of the specified resources, generating image data representing an image of the harvested resource, and selecting one of the electronic tags for the harvested resource; and sending the image data and data identifying the selected electronic tag to a specified entity to register the harvested resource. In an embodiment, the given number of unique physical tags are generated for the specified resource, and the generated image data include data representing one of the physical tags. In an embodiment, a mobile computing device is used to receive the electronic tags, to generate the image data, and to transmit the image data to the specified entity.
    Type: Grant
    Filed: November 29, 2018
    Date of Patent: April 7, 2020
    Assignee: International Business Machines Corporation
    Inventors: David R. Burns, Gregory J. Boss, Andrew R. Jones, Kevin C. McConnell
  • Patent number: 10614593
    Abstract: The present invention relates to an image processing device and method enabling noise removal to be performed according to images and bit rates. A low-pass filter setting unit 93 sets, from filter coefficients stored in a built-in filter coefficient memory 94, a filter coefficient corresponding to intra prediction mode information and a quantization parameter. A neighboring image setting unit 81 uses the filter coefficient set by the low-pass filter setting unit 93 to subject neighboring pixel values of a current block from frame memory 72 to filtering processing. A prediction image generating unit 82 performs intra prediction using the neighboring pixel values subjected to filtering processing, from the neighboring image setting unit 81, and generates a prediction image. The present invention can be applied to an image encoding device which encodes with the H.264/AVC format, for example.
    Type: Grant
    Filed: September 18, 2017
    Date of Patent: April 7, 2020
    Assignee: VELOS MEDIA, LLC
    Inventor: Kazushi Sato
  • Patent number: 10607120
    Abstract: Disclosed are a training method and apparatus for a CNN model, which belong to the field of image recognition. The method comprises: performing a convolution operation, maximal pooling operation and horizontal pooling operation on training images, respectively, to obtain second feature images; determining feature vectors according to the second feature images; processing the feature vectors to obtain category probability vectors; according to the category probability vectors and an initial category, calculating a category error; based on the category error, adjusting model parameters; based on the adjusted model parameters, continuing the model parameters adjusting process, and using the model parameters when the number of iteration times reaches a pre-set number of times as the model parameters for the well-trained CNN model. After the convolution operation and maximal pooling operation on the training images on each level of convolution layer, a horizontal pooling operation is performed.
    Type: Grant
    Filed: March 30, 2018
    Date of Patent: March 31, 2020
    Inventors: Xiang Bai, Feiyue Huang, Xiaowei Guo, Cong Yao, Baoguang Shi
  • Patent number: 10600197
    Abstract: A method of recognizing an object, and a device therefor are provided. The method includes obtaining first sensing data from a first sensor that senses the object, obtaining second sensing data from a second sensor that senses the object, obtaining a first object recognition reliability for the object and a second object recognition reliability for the object respectively based on the first sensing data and the second sensing data, based on the first object recognition reliability and the second object recognition reliability, matching object information of the object recognized using the second sensing data to the first sensing data, and storing the matched object information in a database of one of the first sensor and the second sensor.
    Type: Grant
    Filed: December 13, 2017
    Date of Patent: March 24, 2020
    Assignee: Samsung Electronics Co., Ltd.
    Inventors: Hyun-seok Hong, Sahng-gyu Park, Bo-seok Moon, Hae-in Chun
  • Patent number: 10592991
    Abstract: Methods for assessing a condition of property for insurance purposes include determining a concentration of a molecular constituent at the insured property. The molecular constituent may be a byproduct or residual product of anthropogenic fire accelerants or anthropogenic sources of ignition or explosion. In some embodiments, the concentration of the molecular constituent at the insured property is determined using spectral imaging technology. Radiative transfer computer models may be used to determine the concentration of the molecular constituent based on spectral images.
    Type: Grant
    Filed: February 28, 2018
    Date of Patent: March 17, 2020
    Assignee: Hartford Fire Insurance Company
    Inventor: Patrick D. Brown
  • Patent number: 10593022
    Abstract: In one embodiment, a medical image processing apparatus which analyzes blood flow dynamics in a predetermined region of a subject, the blood flow dynamics being generated from medical images obtained by imaging the predetermined region in time sequence over a plurality of time phases. The medical image processing apparatus includes memory circuitry configured to store a program; and processing circuitry configured to correct pixel values of a second medical image according to an amount of deformation of the second medical image when the second medical image is aligned with a first medical image by executing the program read out from the memory circuitry, the first medical image and the second medical image being among the medical images in the plurality of time phases.
    Type: Grant
    Filed: April 25, 2017
    Date of Patent: March 17, 2020
    Assignee: Canon Medical Systems Corporation
    Inventors: Shintaro Funabasama, Yasuko Fujisawa
  • Patent number: 10586308
    Abstract: Techniques for removal of obstructions in a digital image scene are described, in which target and source digital images that exhibit parallax, one to another, are obtained that were captured together by an image capture device at a similar point in time using two different lenses of the image capture device. A foreground obstruction is identified based on displacement in apparent position of objects in the target and source digital images. The foreground obstruction is removed from the target digital image, such as by generating an obstruction mask that represents the location of the foreground obstruction and copying pixels from the source digital image to the target digital image based on the locations identified in the obstruction mask. The target digital image with the obstruction removed is output to a user interface or service provider system, for example.
    Type: Grant
    Filed: May 9, 2017
    Date of Patent: March 10, 2020
    Assignee: Adobe Inc.
    Inventors: Krishna Singh Karki, Subham Gupta, Poonam Bhalla, Ajay Bedi
  • Patent number: 10580615
    Abstract: Disclosed are a system and method, wherein, during manufacturing of integrated circuit chips on a semiconductor wafer, an in-line optical inspection is performed to acquire a two-dimensional (2D) image of an area of the semiconductor wafer and to confirm and classify a defect in the area. The 2D image is then converted into a virtual three-dimensional (3D) image. To ensure that the 3D image is accurate, techniques are employed to determine the topography of the surface shown in the 2D image based on material-specific image intensity information and, optionally, to filter out any edge effects that result in anomalies within the 3D image. The resulting 3D image is usable for performing an in-line failure analysis to determine a root cause of a defect. Such an in-line failure analysis can be performed significantly faster than any off-line failure analysis and, thus, allows for essentially real-time advanced process control (APC).
    Type: Grant
    Filed: March 6, 2018
    Date of Patent: March 3, 2020
    Inventors: Kok Hin Teo, Jay A. Mody, Jeffrey B. Riendeau, Philip V. Kaszuba, Jian Qiu