Patents Examined by Andrew M Moyer
-
Patent number: 11948280Abstract: A method and system for multi-frame contextual attention are provided. The method includes obtaining a reference frame to be processed, identifying context frames with respect to the reference frame, and producing a refined reference frame by processing the obtained reference frame based on the context frames.Type: GrantFiled: April 28, 2020Date of Patent: April 2, 2024Assignee: Samsung Electronics Co., LtdInventors: Mostafa El-Khamy, Ryan Szeto, Jungwon Lee
-
Patent number: 11948328Abstract: An image-processing device that includes at least one processor which is configured to: set discrete measurement regions in a standard image and a reference image selected from images acquired in a time series manner and calculate motion vectors in the respective measurement regions; set a region of interest in the standard image; select, from the calculated motion vectors, the motion vectors in the set region of interest; estimate a homographic-transformation matrix that represents a motion in the region of interest by using the selected motion vectors; evaluate an error in the homographic-transformation matrix on the basis of the estimated homographic-transformation matrix and the calculated motion vectors; and set, on the basis of the evaluation result for the homographic-transformation matrix estimated on the basis of a first region of interest, a second region of interest a size of which is increased as compared with that of the first region of interest.Type: GrantFiled: July 2, 2021Date of Patent: April 2, 2024Assignee: OLYMPUS CORPORATIONInventors: Kosuke Kajimura, Atsuro Okazawa
-
Patent number: 11922610Abstract: A multi-eye camera system includes multi-eye photographing camera heads 2 and image processing device 3 to which a plurality of the multi-eye photographing camera heads 2 are connected. Each of the multi-eye photographing camera heads 2 is provided with partial image acquiring part 24, parameter memory part 23 and image correcting part 26. Image processing device 3 is provided with input processing part 31, resizing processing part 32 and whole image memory 33.Type: GrantFiled: March 22, 2019Date of Patent: March 5, 2024Assignee: KAMUY Innovation CorporationInventors: Takashi Masuda, Katsuji Matsuura
-
Patent number: 11900567Abstract: An image processing method comprises training an initial network to obtain a first network. The method also comprises training the first network to obtain a second network. The method also comprises training the second network to obtain a third network. The method also comprises processing an original image according to the third network in response to receiving an image processing instruction, to obtain a plurality of target images from the original image.Type: GrantFiled: April 30, 2021Date of Patent: February 13, 2024Assignee: TENCENT TECHNOLOGY (SHENZHEN) COMPANY LIMITEDInventors: Kaihao Zhang, Wenhan Luo, Hongdong Li
-
Patent number: 11887560Abstract: A handheld imaging device has a data receiver that is configured to receive reference encoded image data. The data includes reference code values, which are encoded by an external coding system. The reference code values represent reference gray levels, which are being selected using a reference grayscale display function that is based on perceptual non-linearity of human vision adapted at different light levels to spatial frequencies. The imaging device also has a data converter that is configured to access a code mapping between the reference code values and device-specific code values of the imaging device. The device-specific code values are configured to produce gray levels that are specific to the imaging device. Based on the code mapping, the data converter is configured to transcode the reference encoded image data into device-specific image data, which is encoded with the device-specific code values.Type: GrantFiled: December 7, 2022Date of Patent: January 30, 2024Assignee: Dolby Laboratories Licensing CorporationInventors: Jon Scott Miller, Scott Daly, Mahdi Nezamabadi, Robin Atkins
-
Patent number: 11887273Abstract: A post capture imagery processing system is provided. The system is for use with aerial imagery and includes a server having a processor and a memory and a software application providing instruction to the server to process the captured aerial imagery, such as spherical imagery. The server further includes instructions to geo-rectify the spherical imagery. The geo rectifying of the spherical imagery may include one of use of a third party GIS map to associate corresponding data with the spherical imagery in order to produce a geo-referenced spherical image, or calculate the geo-references by a software application performing particular operations on the server.Type: GrantFiled: November 21, 2022Date of Patent: January 30, 2024Assignee: Aerial Sphere, LLCInventors: Dennis J. Vegh, John C. Femiani, Michael Katic, Anshuman Razdan
-
Patent number: 11881013Abstract: A system configuration suitable for carrying out video recognition by machine learning in a video system including an imaging device and a video processing device is provided. One aspect of the video system according to the present invention includes an imaging device and a video processing device. The imaging device includes an imaging unit configured to generate video data by imaging, a front side NN unit including multiple layers of neural networks from an input layer to a predetermined intermediate hidden layer (hereinafter referred to as “intermediate hidden layer”) that are configured to perform recognition processing of the video data; and an imaging side control unit configured to transmit an output of the intermediate hidden layer of the front side NN unit (hereinafter referred to as “intermediate feature quantity”).Type: GrantFiled: March 20, 2019Date of Patent: January 23, 2024Assignee: HITACHI KOKUSAI ELECTRIC INC.Inventor: Takaomi Kanda
-
Patent number: 11871996Abstract: A spine measurement system comprises an optical measurement probe, one or more targets, a fluoroscope, and a remote station. A-P and lateral images of the spine are taken using the fluoroscope and provided to the remote station. The remote station includes computer vision that can identify endplates and pedicle screws in the spine. The computer vision in the remote station is further used to identify vertebra and bone landmarks of the spine. The remote station can generate quantitative measurement data such as Cobb angles and axial rotation of the spine from the fluoroscope images that correspond to the spine deformity. The optical measurement probe can send images of the spine with pedicle screw extenders extending from the pedicle screws to the remote station. The remotes station using computer vision can provide spine metrics in real-time by tracking position of the pedicle screw extenders.Type: GrantFiled: December 11, 2019Date of Patent: January 16, 2024Assignee: Orthosensor, Inc.Inventors: Erik Herrmann, Scott Clegg
-
Patent number: 11861816Abstract: A system and a method for detecting image forgery through a convolutional neural network are capable of detecting image manipulation of compressed and/or color images. The system comprises a manipulated feature pre-processing unit applying an input image to a high-pass filter to enhance features due to image forgery; a manipulated feature extraction unit extracting image manipulated feature information from the image with the enhanced features through a pre-trained convolutional neural network; a feature refining unit refining the extracted image manipulated feature information; and a manipulation classifying unit determining the image forgery based on the image manipulated feature information refined by the feature refining unit.Type: GrantFiled: October 18, 2019Date of Patent: January 2, 2024Assignee: NHN CLOUD CORPORATIONInventors: Hyeon-gi Kim, Rokkyu Lee
-
Patent number: 11853891Abstract: Method and system with federated learning model for health care applications are disclosed. The system for federated learning comprises multiple edge devices of end users, one or more federated learner update repository, and one or more cloud. Each edge device comprises a federated learner model, configured to send tensors to federated learner update repository. Cloud comprises a federated learner model, configured to send tensors to federated learner update repository. Federated learner update repository comprises a back-end configuration, configured to send model updates to edge devices and cloud.Type: GrantFiled: March 11, 2020Date of Patent: December 26, 2023Assignee: SHARECARE AI, INC.Inventors: Walter Adolf De Brouwer, Srivatsa Akshay Sharma, Neerajshyam Rangan Kashyap, Kartik Thakore, Philip Joseph Dow
-
Patent number: 11830194Abstract: Space-based ground-imaging lidar has become increasingly feasible with recent advances in compact fiber lasers and single-photon-sensitive Geiger-mode detector arrays. A challenge with such a system is imperfect pointing knowledge caused by angular jitter, exacerbated by long distances between the satellite and the ground. Unless mitigated, angular jitter blurs the 3D lidar data. Unfortunately, using mechanical isolation, advanced IMUs, star trackers, or auxiliary passive optical sensors to reduce the error in pointing knowledge increases size, weight, power, and/or cost. Here, the 2-axis jitter time series is estimated from the lidar data. Simultaneously, a single-surface model of the ground is estimated as nuisance parameters. Expectation Maximization separates signal and background detections, while maximizing the joint posterior probability density of the jitter and surface states.Type: GrantFiled: April 24, 2020Date of Patent: November 28, 2023Assignee: Massachusetts Institute of TechnologyInventors: Ethan Phelps, Charles Primmerman, Eric Statz
-
Patent number: 11823442Abstract: Described herein are systems and methods that search videos and other media content to identify items, objects, faces, or other entities within the media content. Detectors identify objects within media content by, for instance, detecting a predetermined set of visual features corresponding to the objects. Detectors configured to identify an object can be trained using a machine learned model (e.g., a convolutional neural network) as applied to a set of example media content items that include the object. The systems and methods describe techniques that search videos and media content to determine the presence of unknown objects, generate novel detectors trained to identify the unknown objects, and apply the novel detectors to historical media content to identify previous appearances of the unknown objects.Type: GrantFiled: March 4, 2020Date of Patent: November 21, 2023Assignee: Matroid, Inc.Inventors: Reza Zadeh, Ryan Wong, John Goddard, Jiahang Li, Steven Chen, Xiaoyun Yang
-
Patent number: 11816823Abstract: A computer-implemented method for processing images may include obtaining at least one image for analyzing; inputting the at least one image to at least one of a plurality of image plugins; analyzing the at least one image via the at least one of the plurality of image plugins; determining metadata related to the at least one image based on the at least one of the plurality of image plugins; filtering the at least one image based on one or more rule sets to generate at least one filtered image; sorting the at least one filtered image to generate at least one sorted image; displaying the at least one sorted image based on an organizational sequence of a webpage; displaying navigation controls via the webpage; and displaying the at least one sorted image according to a user interaction with the navigation controls on the webpage.Type: GrantFiled: October 26, 2020Date of Patent: November 14, 2023Assignee: Capital One Services, LLCInventors: Geoffrey Dagley, Qiaochu Tang, Micah Price, Avid Ghamsari, Staevan Duckworth
-
Patent number: 11816838Abstract: An image processing apparatus (10) is disclosed that comprises a processor arrangement (16) adapted to receive image data corresponding to a region of interest (1) of a patients cardiovascular system, said image data comprising a temporal sequence (15) of intravascular ultrasound images acquired (150) at different phases of at least one cardiac cycle of said patient, said intravascular ultrasound images imaging overlapping volumes of the patient's cardiovascular system; implement a spatial reordering process of said temporal sequence of intravascular ultrasound images by evaluating the image data to select at least one spatial reference (6, Vref) associated with said temporal sequence of intravascular ultrasound images; estimating a distance to the at least one spatial reference for each of the intravascular ultrasound images of said temporal sequence; and reordering said temporal sequence of intravascular ultrasound images into a spatial sequence of intravascular ultrasound images based on the estimated distType: GrantFiled: September 25, 2019Date of Patent: November 14, 2023Assignee: KONINKLIJKE PHILIPS N.V.Inventors: Vincent Maurice André Auvray, Raoul Florent, Antoine Collet-Billon, Benoit Jean-Dominique Bertrand Maurice Mory
-
Patent number: 11790492Abstract: There is provided a method and a system for customized image denoising with interpretability. A deep neural network (NN) is trained to denoise an image on a training dataset including pairs of noisy and corresponding clean images acquired from an imaging apparatus, where during the training a structured covariance score (SCS) indicative of a performance of the deep NN in recovering content of corresponding clean images relative to the denoised image is determined based on sparse conditional correlations. A test noisy image is received and denoised by the deep NN. A user feedback score indicative of user satisfaction of the denoising is obtained. A quality parameter is obtained based on the SCS and a quality metric indicative of denoised image quality is obtained from a pretrained NN, and compared with the user feedback score. If the SCS is above the user feedback score, the deep NN is provided for denoising.Type: GrantFiled: December 17, 2020Date of Patent: October 17, 2023Assignee: THALES SAInventors: Ola Ahmad, Freddy Lecue
-
Patent number: 11762751Abstract: Disclosed are a computer-implemented method, a system, and a computer program product for system-level tunable parameter identification. Performance characteristic data for an application to be tuned can be obtained by one or more processing units. At least one system-level tunable parameter for the application to be tuned can be identified by one or more processing units based on the obtained performance characteristic data for the application to be tuned and a pattern between training performance characteristic data and a set of training system-level parameter-related correlation coefficients. The set of training system-level parameter-related correlation coefficients can be respective correlation coefficients of system-level tunable parameters with respect to at least one performance metric.Type: GrantFiled: September 17, 2019Date of Patent: September 19, 2023Assignee: International Business Machines CorporationInventors: Li Cao, Guang Cheng Li, Rong Yan, Qi Ming Teng, Yubo Li, Cheng Fang Wang
-
Patent number: 11741591Abstract: Provided are a method and an apparatus for detecting a thickness of paper, and a storage medium. The method includes: acquiring a first thickness image of a target paper; extracting at least one first thickness feature and at least one second thickness feature from the first thickness image in the first thickness image; performing average processing on gray values of multiple pixel points in the at least one first thickness feature to acquire a first gray value, and the average processing is performed on gray values of the multiple pixel points in the at least one second thickness feature to acquire a second gray value; when the first gray value is smaller than the second gray value of multiple pixel points in at least one second thickness feature, determining that a thickness of a region, corresponding to the at least one second thickness feature, on the target paper has changed.Type: GrantFiled: May 17, 2019Date of Patent: August 29, 2023Assignee: WEIHAI HUALING OPTO-ELECTRONICS CO., LTD.Inventors: Li Jiang, Rongxin Song, Xiumei Qi, Jie Xian
-
Patent number: 11734795Abstract: A processing device and method for a panoramic image includes converting the panoramic image to a cubemap format; under the cubemap format, calculating a depth information image of the panoramic image, and performing a seamless processing on the depth information image. The format of the depth information image is converted to the original format from the cubemap format to obtain the depth information image corresponding to the panoramic image.Type: GrantFiled: December 28, 2020Date of Patent: August 22, 2023Assignee: INDUSTRIAL TECHNOLOGY RESEARCH INSTITUTEInventors: Yan-Jen Su, Ruen-Rone Lee, Feng-Sheng Lin
-
Patent number: 11714881Abstract: A method of improving image quality of a stream of input images is described. The stream of input images, including a current input image, is received. One or more target objects, including a first target object, are identified spatio-temporally within the stream of input images. The one or more target objects are tracked spatio-temporally within the stream of input images. The current input image is segmented into i) a foreground including the first target object, and ii) a background. The foreground is processed to have improved image quality in the current input image. Processing of the foreground further comprises processing the first target object using a same processing technique as for a prior input image of the stream of input images based on the tracking of the first target object. The background is processed differently from the foreground. An output image is generated by merging the foreground with the background.Type: GrantFiled: May 27, 2021Date of Patent: August 1, 2023Assignee: Microsoft Technology Licensing, LLCInventors: Eric Chris Wolfgang Sommerlade, Sunando Sengupta, Alexandros Neophytou
-
Patent number: 11715214Abstract: Methods and apparatus for determining a trajectory of a axisymmetric object in 3-D physical space using a digital camera which records 2-D image data are described. In particular, based upon i) a characteristic length of the axisymmetric object, ii) a physical position of the camera determined from sensors associated with the camera (e.g., accelerometers) and iii) captured 2-D digital images from the camera including a time at which each image is generated relative to one another, a position, a velocity vector and an acceleration vector can be determined in three dimensional physical space for axisymmetric object objects as a function of time. In one embodiment, the method and apparatus can be applied to determine the trajectories of objects in games which utilize axisymmetric object objects, such as basketball, baseball, bowling, golf, soccer, rugby or football.Type: GrantFiled: July 24, 2020Date of Patent: August 1, 2023Assignee: Pillar Vision, Inc.Inventors: Alan W. Marty, John Carter