Image Segmentation Using Color Patents (Class 382/164)
-
Patent number: 11379992Abstract: Systems and methods for frame and scene segmentation are disclosed herein. One method includes associating a first primary element from a first frame with a background tag, associating a second primary element from the first frame with a subject tag, generating a background texture using the first primary element, generating a foreground texture using the second primary element, and combining the background texture and the foreground texture into a synthesized frame. The method also includes training a segmentation network using the background tag, the foreground tag, and the synthesized frame.Type: GrantFiled: May 14, 2019Date of Patent: July 5, 2022Assignee: Matterport, Inc.Inventors: Gary Bradski, Prasanna Krishnasamy, Mona Fathollahi, Michael Tetelman
-
Patent number: 11379985Abstract: A computer-implemented method for segmenting an input image, the method comprises: generating a first segmentation of the input image using a first machine learning system, the first segmentation comprising multiple segments; receiving, from a user, at least one indication, wherein each indication corresponds to a particular segment from the multiple segments, and indicates one or more locations of the input image as belonging to that particular segment; constructing, for each segment of the multiple segments having at least one corresponding indication, a respective geodesic distance map, based on the input image and the user indications received for that segment; and generating a second segmentation using a second machine learning system based on the input image and the constructed geodesic distance maps.Type: GrantFiled: June 14, 2018Date of Patent: July 5, 2022Assignee: King's College LondonInventors: Guotai Wang, Tom Vercauteren, Sebastien Ourselin, Wenqi Li, Lucas Fidon
-
Patent number: 11361333Abstract: A system and method for decentralizing data and determining performance of different entities in multiple geographical and categorial markets whereby the system may determine complimentary entities to a user's entities or other existing entities for presentation to the user of overlapping procurements and demographics to gather a deeper understanding into the same results yielded by their competition whereby users will be able to provide more personal experiences for each consumer, as well as achieve pricing discovery, greater brand awareness, and marketing strategy.Type: GrantFiled: August 30, 2021Date of Patent: June 14, 2022Assignee: CONTENT AWARE, LLCInventor: Cameron A. Holmes
-
Patent number: 11341759Abstract: A device may receive a target document. The device may segment the target document into multiple segments. The device may determine, for each segment of the multiple segments, a set of color parameters for a corresponding set of pixels included in that segment. The device may determine, for each segment of the multiple segments, an average color parameter for that segment based on the set of color parameters for the corresponding set of pixels included in that segment. The device may generate a target color profile for the target document based on determining the average color parameter for each segment. The device may compare the target color profile and a model color profile associated with classifying the target document. The device may classify the target document based on comparing the target color profile and the model color profile.Type: GrantFiled: March 31, 2020Date of Patent: May 24, 2022Assignee: Capital One Services, LLCInventor: Thomas Sickert
-
Patent number: 11334086Abstract: Autonomous robots and methods of operating the same are disclosed. An autonomous robot includes a sensor and memory including machine readable instructions. The autonomous robot further includes at least one processor to execute the instructions to generate a velocity costmap associated with an environment in which the robot is located. The processor generates the velocity costmap based on a source image captured by the sensor. The velocity costmap includes velocity information indicative of movement of an obstacle detected in the environment.Type: GrantFiled: September 27, 2017Date of Patent: May 17, 2022Assignee: Intel CorporationInventors: Bin Wang, Jianguo Li, Shoumeng Yan
-
Patent number: 11335081Abstract: A method for automatic extraction of data from a graph, including text area locating and text box classification; locating of coordinate axes, and locating of the positions of hatch marks on the coordinate axes; legend locating and information extraction; extracting corresponding bar or polyline connected components according to legend color, and filtering and classification; determining key points on the X-axis and locating a corresponding X-axis label for each key point; locating key points of the bars and polyline according to the X-axis key points, determining labeled numerical text boxes that correspond to the key points, and identifying the numerical text; calculating a corresponding value for each pixel, and estimating corresponding values of the key points of the bars or polyline; determining a final result according to a difference between the estimated values and the recognized labeled values.Type: GrantFiled: August 19, 2020Date of Patent: May 17, 2022Assignee: HANGZHOU DIANZI UNIVERSITYInventors: Jinliang Yao, Yuelun Yu, Chuang Hu
-
Patent number: 11301719Abstract: A semantic segmentation model training method includes: performing, by a semantic segmentation model, image semantic segmentation on at least one unlabeled image to obtain a preliminary semantic segmentation result as the category of the unlabeled image; obtaining, by a convolutional neural network based on the category of the at least one unlabeled image and the category of at least one labeled image, sub-images respectively corresponding to the at least two images and features corresponding to the sub-images, where the at least two images comprise the at least one unlabeled image and the at least one labeled image, and the at least two sub-images carry the categories of the corresponding images; and training the semantic segmentation model on the basis of the categories of the at least two sub-images and feature distances between the at least two sub-images.Type: GrantFiled: December 25, 2019Date of Patent: April 12, 2022Assignee: BEIJING SENSETIME TECHNOLOGY DEVELOPMENT CO., LTD.Inventors: Xiaohang Zhan, Ziwei Liu, Ping Luo, Chen Change Loy, Xiaoou Tang
-
Patent number: 11282279Abstract: A system that enables real-time recognition of a real-space environment in a mixed-reality environment includes a server, a portable display device including a display unit for displaying a virtual object to a user and a photographing unit that photographs a real space, and image acquisition devices that acquire images individually from a plurality of fixed points where it is possible to photograph a region in a predetermined real space, the system making it possible to render the virtual object in a superimposed fashion on a real space or a photographed image of the real space, viewed by the user via the display unit, the server or the display device including a three-dimensional-space-data storage part, a table storage part, a color-information determining part, a color-information updating part, a user-environment determining part, a virtual-illumination-information generating part, and a rendering part.Type: GrantFiled: December 20, 2018Date of Patent: March 22, 2022Assignee: CYGAMES, INC.Inventor: Shuichi Kurabayashi
-
Patent number: 11238285Abstract: Image analysis includes obtaining, from an image signal processor, image processing information corresponding to a previously processed image, obtaining scene classification information for an input image based on the image processing information, generating a processed image by processing the input image based on the scene classification information, and outputting the processed image. The image processing information includes automatic white balance correction information and obtaining the scene classification information includes obtaining the scene classification information based on the automatic white balance correction information.Type: GrantFiled: April 5, 2021Date of Patent: February 1, 2022Assignee: GoPro, Inc.Inventors: Sylvain Leroy, Adrien Cariou, Miguel Marques, Julien Caron, Bruno Faure, Stéphane Belardi
-
Patent number: 11222194Abstract: Systems and methods disclosed herein describe a platform that automatically creates and executes a scoring guide for use in anatomical pathology. The platform can employ a fully-automated workflow for clustering the biological objects of interest and for providing cell-by-cell read-outs of heterogeneous tumor biomarkers based on their stain appearance. The platform can include a module for automatically creating and storing a scoring guide in a training database based on training digital images (240, 250), and an object classification module that executes the scoring guide when presented with new digital images to be scored pursuant to the scoring guide (299).Type: GrantFiled: December 20, 2017Date of Patent: January 11, 2022Assignee: VENTANA MEDICAL SYSTEMS, INC.Inventors: Michael Barnes, Joerg Bredno, Jim F. Martin
-
Patent number: 11200676Abstract: Systems and methods of improving alignment in dense prediction neural networks are disclosed. A method includes identifying, at a computing system, an input data set and a label data set with one or more first parts of the input data set corresponding to a label. The computing system processes the input data set using a neural network to generate a predicted label data set that identifies one or more second parts of the input data set predicted to correspond to the label. The computing system determines an alignment result using the predicted label data set and the label data set and a transformation of the one or more first parts, including a shift, rotation, scaling, and/or deformation, based on the alignment result. The computing system computes a loss score using the transformation, label data and the predicted label data set and updates the neural network based on the loss score.Type: GrantFiled: January 17, 2020Date of Patent: December 14, 2021Assignee: VERILY LIFE SCIENCES LLCInventors: Cheng-Hsun Wu, Ali Behrooz
-
Patent number: 11184644Abstract: A weighed run-length encoding and decoding method and related devices and encoded bitstream. The encoded bitstream can comprise one or more of the following: a skip command packed into a nybble, the skip command indicating how many transparent pixels which are inserted into the bitstream, wherein there is up to a maximum number of transparent pixels; a solid command packed into a nybble, the solid command indicating how many solid pixels should be inserted into the decoded bitstream, wherein there are up to the maximum number of solid pixels; and a quote command packed into a nybble, the quote command indicating how many quoted pixels should be inserted into the decoded bitstream, wherein there are up to the maximum number of quoted pixels.Type: GrantFiled: June 25, 2019Date of Patent: November 23, 2021Assignee: Moddable, Inc.Inventor: J. Peter Hoddie
-
Patent number: 11176715Abstract: There is provided a system and method for color representation generation. In an aspect, the method includes: receiving three base colors; receiving a patchwork parameter; and generating a color representation having each of the three base colors at a vertex of a triangular face, the triangular face having a color distribution therein, the color distribution discretized into discrete portions, the amount of discretization based on the patchwork parameter, each discrete portion having an interpolated color determined to be a combination of the base colors at respective coordinates of such discrete portion. In further aspects, one or more color representations are generated based on an input image and can be used to modify colors of a reconstructed image.Type: GrantFiled: November 24, 2020Date of Patent: November 16, 2021Assignee: THE GOVERNING COUNCIL OF THE UNIVERSITY OF TORONTOInventors: Maria Shugrina, Amlan Kar, Sanja Fidler, Karan Singh
-
Patent number: 11176402Abstract: Provided are an artificial intelligence (AI) system for simulating functions such as recognition, determination, and so forth of the human brain by using a mechanical learning algorithm like deep learning, or the like, and an application thereof.Type: GrantFiled: May 8, 2018Date of Patent: November 16, 2021Inventors: Deok-ho Kim, Won-woo Lee, Jae-woong Lee, Bong-hoon Park, Gun-Ill Lee, Ji-won Jeong, Teresa Ka Ki Ng
-
Patent number: 11169085Abstract: Non-transitory computer-readable media, spectroradiometer systems, and methods for calibrating a spectroradiometer. In one embodiment, a non-transitory computer-readable medium includes instructions that, when executed by an electronic processor, cause the electronic processor to perform a set of operations. The set of operations includes receiving spectral data regarding an object-of-interest that is captured by a handheld spectroradiometer, detecting a characteristic of the object-of-interest by performing a spectral analysis on the spectral data that is received, and controlling a display to display the characteristic of the object-of-interest.Type: GrantFiled: February 18, 2020Date of Patent: November 9, 2021Assignee: MICHIGAN TECHNOLOGICAL UNIVERSITYInventors: Robert Shuchman, Michael Sayers, Reid Sawtell, Karl Bosse
-
Patent number: 11172096Abstract: An image processing apparatus includes a detection unit, a determination unit, a decision unit, and a processing unit. The detection unit detects an isolated pixel included in an image. The determination unit determines whether the isolated pixel detected by the detection unit changes brighter or darker by edge enhancement processing. The decision unit decides an edge enhancement amount of the isolated pixel based on a determination result of the determination unit. The processing unit performs the edge enhancement processing on the isolated pixel, based on the edge enhancement amount decided by the decision unit.Type: GrantFiled: May 26, 2020Date of Patent: November 9, 2021Assignee: Canon Kabushiki KaishaInventor: Hirokazu Tamura
-
Patent number: 11158091Abstract: The present disclosure includes systems and methods for color transfer. The method includes receiving a target image, and determining dominant source colors. The method further includes transforming the target image into a color model including a target luminance component and a target color information component. Additionally, the method includes segmenting the target image into a plurality of target segments based on the target color information component or the target luminance component and extracting dominant target colors from the target image by extracting information for at least one of the dominant target colors from each target segment of the plurality of target segments. Further, the method includes generating a color mapping relationship between the dominant target colors and the dominant source colors, and creating a recolored target image using the color mapping relationship.Type: GrantFiled: September 7, 2017Date of Patent: October 26, 2021Assignees: Trustees of Tufts College, THE BOARD OF REGENTS OF THE UNIVERSITY OF TEXAS SYSTEMInventors: Karen Panetta, Long Bao, Sos Agaian
-
Patent number: 11151307Abstract: An annotation corresponding to a first range of text of a first document may be received. Based on the annotation, comparisons may be performed between a text string that comprises the first range of text and a group of text of a second document at different positions in the group of text. Based on the comparisons, similarity scores between the text string and the group of text may be determined at the different positions in the group of text. A position for the annotation in the group of text may be selected based on the similarity scores at the different positions. The annotation may be associated with a second range of text in the group of text that corresponds to the position.Type: GrantFiled: November 13, 2018Date of Patent: October 19, 2021Assignee: Adobe Inc.Inventors: Mudit Rastogi, Prakhar Mehrotra, Souvik Sinha Deb, Gaurav Tarlok Kakkar
-
Patent number: 11151374Abstract: A method of generating at least one digital signature for at least one surface of a material element, referred to as an “examined surface.” The method is performed by a data processor system, and may include, for each examined surface: obtaining an “offset” image having at least a portion showing the examined surface; obtaining at least one zone of interest in the offset image by putting into correspondence the offset image with at least one predetermined “equivalent” zone that is included in a model image having at least a portion representing a “reference” surface; obtaining a “registered” image by registering the offset image with the model image by applying a homography transformation to the offset image, wherein the transformation is obtained from the at least one zone of interest and the at least one equivalent zone; and generating the digital signature, which characterizes the structure of the examined surface, from the registered image.Type: GrantFiled: August 30, 2017Date of Patent: October 19, 2021Assignee: ARJO SOLUTIONSInventors: Fanny Cugnet, Hubert Glasson, Hervé Guiot Du Doignon
-
Patent number: 11120606Abstract: An image processing method is disclosed, comprising: receiving a plurality of images of an object; and, for each image, determining a brightness value for each point on the object that is visible, and generating a tile grid for the image. Each point that is visible in an image corresponds to one tile in the tile grid. For each tile pair where at least one point is visible in both tiles of the tile pair, a brightness offset value is determined. For each image, an offset matrix is generated. Values of the offset matrix comprise brightness offsets to apply to each tile of a tile grid for the image to normalize the brightness of the image with the other images. A plurality of reconstructed images is generated based on performing, for each image, transformation operations on the given image based on the corresponding offset matrix to generate a reconstructed image.Type: GrantFiled: April 24, 2020Date of Patent: September 14, 2021Assignee: Electronic Arts Inc.Inventor: Mathieu Lamarre
-
Patent number: 11120297Abstract: An input image that includes a target area may be received. A first segment for extracting the target area from the input image may be generated using a first extracting model. A second segment for extracting the target area from the input image may be generated using a second extracting model. The first segment is compared to the second segment to determine a combined segment of at least the target area.Type: GrantFiled: November 30, 2018Date of Patent: September 14, 2021Assignee: International Business Machines CorporationInventors: Koichi Takahashi, Hiroki Nakano, Masahiro Okawa
-
Patent number: 11115597Abstract: The present disclosure provides a mobile terminal having a plurality of artificial intelligence (AI) agents. More particularly, a mobile terminal according to the present disclosure includes: a camera, a touch screen configured to display screen information, and a controller configured to activate the camera in response to a request of a user, and store an image received through the camera. The controller extracts a search keyword from a search command when the search command for searching image information is input via a first AI agent, retrieves at least one piece of image information corresponding to the search keyword from pre-stored image information through a second AI agent, and controls the touch screen such that a search result is displayed on an execution screen of the first AI agent.Type: GrantFiled: June 3, 2019Date of Patent: September 7, 2021Assignee: LG ELECTRONICS INC.Inventors: Jaeryong Kang, Mingyoung Kam, Jongseol Kim, Myoseop Sim, Hyunju Oh, Bomi Jang, Jaeki Cho, Dongki Cheon, Jahee Hur
-
Patent number: 11113490Abstract: Evaluating a symbol on an object can include acquiring a first image of the object, including the symbol. A second image can be derived from the first image based upon determining a saturation threshold for the second image and possibly scaling of pixel values to a reduced bit-depth for the second image.Type: GrantFiled: August 24, 2020Date of Patent: September 7, 2021Assignee: Cognex CorporationInventor: Glenn Spitz
-
Patent number: 11109065Abstract: Compressing a frame of video includes receiving a frame of a video, identifying a three dimensional (3D) object in the frame, matching the 3D object to a stored 3D object, compressing the frame of the video using a color prediction scheme based on the 3D object and the stored 3D object, and storing the compressed frame with metadata, the metadata identifying the 3D object, indicating a position of the 3D object in the frame of the video and indicating an orientation of the 3D object in the frame of the video.Type: GrantFiled: September 26, 2018Date of Patent: August 31, 2021Assignee: GOOGLE LLCInventors: Michael Hemmer, Ameesh Makadia
-
Patent number: 11100655Abstract: An image processing apparatus includes a detection unit configured to detect a first region in which a foreground object is present with respect to a plurality of captured images, a retention unit configured to retain a first background image, a generation unit configured to generate a second background image based on portions of each of the plurality of captured images which are not detected as a first region, and an output unit configured to select one of the first background image and the second background image based on a property of the second background image, and configured to output, based on the selected background image and the first region, an image in which the foreground object is obscured.Type: GrantFiled: January 29, 2019Date of Patent: August 24, 2021Assignee: Canon Kabushiki KaishaInventor: Taichi Matsui
-
Patent number: 11070786Abstract: An immersive experience system is provided. The immersive experience system has a processor that determines a position of a first head-mounted display. Further, the processor determines a position of a second head-mounted display. The processor also generates a first image for a first immersive experience corresponding to the position of the first head-mounted display. Moreover, the process encodes the first image into a first infrared spectrum illumination having a first wavelength. In addition, the processor generates a second image for a second immersive experience corresponding to the position of the second head-mounted display. Finally, the processor encodes the second image into a second infrared spectrum illumination having a second wavelength. The first wavelength is distinct from the second wavelength.Type: GrantFiled: May 2, 2019Date of Patent: July 20, 2021Assignee: DISNEY ENTERPRISES, INC.Inventors: Steven Chapman, Joseph Popp, Alice Taylor, Joseph Hager
-
Patent number: 11070705Abstract: A method includes obtaining a dynamic range representation of an image including multiple color channels. The dynamic range representation includes dynamic range information of pixels of the image in the color channels. The method further includes decomposing the dynamic range representation into a smooth portion, a color portion, and a detail portion. The smooth portion includes low frequency information of the pixels in the color channels. The color portion includes color intensity information of the pixels in the color channels. The detail portion includes high frequency information of the pixels in the color channels. The method also includes adjusting the smooth portion, the color portion, and the detail portion, respectively, to obtain an adjusted smooth portion, an adjusted color portion, and an adjusted detail portion, and combining the adjusted smooth portion, the adjusted color portion, and the adjusted detail portion to generate an output image.Type: GrantFiled: November 27, 2019Date of Patent: July 20, 2021Assignee: SZ DJI TECHNOLOGY CO., LTD.Inventor: Zisheng Cao
-
Patent number: 11062480Abstract: A method for processing an image may include: collecting an ambient light spectrum while capturing an image (10); determining RGB tristimulus values corresponding to ambient environment of the captured image based on the collected ambient light spectrum and a color filter absorption curve (11); determining skin color coordinates based on the collected ambient light spectrum and a spectral absorption curve of a skin color model (12); and determining RGB values in the skin color model based on the RGB tristimulus values and the skin color coordinates (13).Type: GrantFiled: December 14, 2018Date of Patent: July 13, 2021Assignee: BOE TECHNOLOGY GROUP CO., LTD.Inventor: Xingxing Zhao
-
Patent number: 11037357Abstract: A technique for compressing an original image is disclosed. According to the technique, an original image is obtained and a delta-encoded image is generated based on the original image. Next, a segregated image is generated based on the delta-encoded image and then the segregated image is compressed to produce a compressed image. The segregated image is generated because the segregated image may be compressed more efficiently than the original image and the delta image.Type: GrantFiled: December 6, 2019Date of Patent: June 15, 2021Assignees: Advanced Micro Devices, Inc., ATI Technologies ULCInventors: Ruijin Wu, Skyler Jonathon Saleh, Christopher J. Brennan, Kei Ming Kwong, Anthony Hung-Cheong Chan
-
Patent number: 11032442Abstract: An image processing apparatus of an embodiment includes an image processing unit to convert a color image to a monochrome image. The color image comprises pixels with a plurality of color components. The image processing unit is configured to generate a histogram from the monochrome image showing a color intensity gradation in the monochrome image by pixel frequency. A processor is configured to obtain a first threshold value based on the histogram, determine for each color component of each pixel in the color image whether or not each color component of the pixel is light based on the first threshold value, and generate a corrected color image by removing a background coloring from the color image by correcting each pixel for which all the color components are determined to be light.Type: GrantFiled: March 15, 2019Date of Patent: June 8, 2021Assignee: TOSHIBA TEC KABUSHIKI KAISHAInventor: Keiko Kanamori
-
Patent number: 11010876Abstract: An image processing system includes: a weight determination circuitry configured to determine a band containing important information from among a group of images, which are acquired by a plurality of sensors, and to express a degree of importance of the band as a weight; a calculation circuitry configured to calculate, using the weight, an amount calculated based on a gradient of an image based on a gradient of each image, that is calculated based on the group of images, in order to restrict a gradient of an output image; and an image optimization circuitry configured to compose the output image using the amount calculated based on the gradient of the image.Type: GrantFiled: June 10, 2016Date of Patent: May 18, 2021Assignees: NEC Corporation, TOKYO INSTITUTE OF TECHNOLOGYInventors: Takashi Shibata, Masayuki Tanaka, Masatoshi Okutomi
-
Patent number: 11012589Abstract: An image processing apparatus includes a first generation unit and a second generation unit. The first generation unit generates reduced original image data for forming a reduced image of an original image to be formed on the basis of print data. The second generation unit generates background-emphasized reduced image data for forming a reduced image in which a reduced background region is emphasized in the reduced image, and the reduced background region corresponds to a background region of an object included in the original image.Type: GrantFiled: November 9, 2017Date of Patent: May 18, 2021Assignee: FUJI XEROX CO., LTD.Inventor: Masaki Watanabe
-
Patent number: 10991078Abstract: Received image data is enhanced to create enhanced image data using image processing to remove artifacts and to retrieve information associated with a desired target output. Image segmentation is performed on useable enhanced image data to created segmented image data by partitioning the enhanced image data into coherent regions with respect to a particular image-based criterion. Useable segmented image data and auxiliary data is pre-processing for input into a neural network as pre-processed data. The pre-processed data is divided into training, validation, and testing data subsets. A neural network architecture is determined to process the pre-processed data and the determined neural network architecture is executed using the pre-processed data. Output of the determined neural network is post-processed as post-processed data. The post-processed data is compared to a known value range associated with the post-processed data to determine if the post-processed data satisfies a desired output result.Type: GrantFiled: September 14, 2018Date of Patent: April 27, 2021Assignee: Saudi Arabian Oil CompanyInventor: Mustafa Hakimuddin
-
Patent number: 10991103Abstract: A method for extracting a person region includes detecting a face from a raw image input in real time; configuring a search range based on the detected face; extracting an outline of a person region by performing preprocessing on an image within the search range; configuring watershed markers on a background and a foreground of the image within the search range based on the extracted outline; performing a watershed transform based on the configured watershed markers; and separating the person region from the raw image using a result of the watershed transform.Type: GrantFiled: October 18, 2019Date of Patent: April 27, 2021Assignee: ELECTRONICS AND TELECOMMUNICATIONS RESEARCH INSTITUTEInventors: Won Seok Chae, Hyeon Jin Kim, Soo Myung Park, Ah Reum Oh, Ha Lim Choi
-
Patent number: 10977802Abstract: Techniques and systems are provided for segmenting one or more frames. For example, incremental optical flow maps can be determined between adjacent frames of a plurality of frames. Using the incremental optical flow maps, a cumulative optical flow map can be determined between a first frame of the plurality of frames and a last frame of the plurality of frames. A segmentation mask can be determined using the first frame of the plurality of frames. Foreground pixels of the segmentation mask for the last frame of the plurality of frames can then be adjusted relative to corresponding foreground pixels for the first frame. The foreground pixels can be adjusted using the cumulative optical flow map between the first frame and the last frame of the plurality of frames.Type: GrantFiled: August 29, 2018Date of Patent: April 13, 2021Assignee: QUALCOMM IncorporatedInventors: Sunid Wilson, Ramin Rezaiifar
-
Patent number: 10964288Abstract: A method for selecting a dominant color among a set of digital content. The dominant color is used to set the visual aspects of a user interface through which the user views the digital content, enabling a visually appealing application, rather than one reliant on neutral and possibly uncomplimentary colors.Type: GrantFiled: June 26, 2019Date of Patent: March 30, 2021Assignee: Western Digital Technologies, Inc.Inventors: Thomas Burlington Walker, Laurent Baumann
-
Patent number: 10955505Abstract: A magnetic resonance imaging apparatus according to an embodiment includes sequence controlling circuitry and processing circuitry. The sequence controlling circuitry is configured to acquire k-space data by executing a pulse sequence while performing undersampling. The processing circuitry is configured to generate an output target image by generating a folded image by applying a Fourier transform to the k-space data and further unfolding the folded image by performing a process that uses a regularization term. The processing circuitry applies a weight to the regularization term on the basis of whether or not each of the pixels in the output target image is included in an observation target region.Type: GrantFiled: November 15, 2016Date of Patent: March 23, 2021Assignee: CANON MEDICAL SYSTEMS CORPORATIONInventors: Kanako Saito, Hidenori Takeshima
-
Patent number: 10943158Abstract: A method performed by a mobile terminal may include receiving an image that includes text, translating the text into another language and superimposing and displaying the translated text over the received image.Type: GrantFiled: July 19, 2019Date of Patent: March 9, 2021Assignee: Sony CorporationInventors: Hakan Mats Englund, Simon Lessing
-
Patent number: 10930005Abstract: A method and apparatus for determining a building/urban profile image location includes extracting a building/urban profile from an obtained image and comparing the image to a database of building/urban profiles. The database may be created by obtaining point cloud data sets scanned building/urban profiles, converting the obtained point cloud data sets to a corresponding 3D surface model, and creating the database of stored building/urban profiles viewed from multiple locations within each of the 3D surface models. By comparing the extracted building/urban profile image with the stored building/urban profiles contained within the database and finding a match the location from which the extracted building/urban profile image was taken can be determined.Type: GrantFiled: August 31, 2018Date of Patent: February 23, 2021Assignee: ObjectVideo Labs, LLCInventors: Gang Qian, Huiping Li, Yunxian Zhou Zhou, David Conger, Allison Beach
-
Patent number: 10915788Abstract: Disclosed herein are system, method, and computer program product embodiments for optical character recognition using end-to-end deep learning. In an embodiment, an optical character recognition system may train a neural network to identify characters of pixel images and to assign index values to the characters. The neural network may also be trained to identify groups of characters and to generate bounding boxes to group these characters. The optical character recognition system may then analyze documents to identify character information based on the pixel data and produce a segmentation mask and one or more bounding box masks. The optical character recognition system may supply these masks as an output or may combine the masks to generate a version of the received document having optically recognized characters.Type: GrantFiled: September 6, 2018Date of Patent: February 9, 2021Assignee: SAP SEInventors: Johannes Hoehne, Anoop Raveendra Katti, Christian Reisswig
-
Patent number: 10916333Abstract: A regression model is generated to map observation records of a first dimensionality to a second dimensionality. Using a set of transformed records obtained from the first regression model, a Gaussian mixture model of the distribution of observation records of the second dimensionality is trained. Using a Gaussian distribution obtained from the Gaussian mixture model, a recommended modification of a proposed training set of a classifier is obtained.Type: GrantFiled: June 26, 2017Date of Patent: February 9, 2021Assignee: Amazon Technologies, Inc.Inventor: Kalidas Yeturu
-
Patent number: 10902607Abstract: A method of instance segmentation, comprising, defining a bounding box, depth mapping the bounding box, sorting the depth map, estimating stability due to noise of the sorted depth map, detecting low frequency jumps of the sorted depth map based on the estimated stability, detecting high frequency jumps of the sorted depth map, comparing the detected low frequency jumps and the detected high frequency jumps, detecting at least one of a largest object and a closest object based on the comparison and pixel masking the at least one of the largest object and the closest object.Type: GrantFiled: December 6, 2019Date of Patent: January 26, 2021Assignee: Black Sesame International Holding LimitedInventors: Jeff Lee, Guan Wang
-
Patent number: 10896323Abstract: The present disclosure relates to an image processing method and device, a computer readable storage medium, and an electronic device. The method includes: acquiring an image to be processed; detecting a foreground target of the image, and recognizing the foreground target, to obtain a foreground classification result; detecting a background area of the image, and recognizing the background area, to obtain a background classification result; and obtaining a classification label of the image based on the foreground classification result and the background classification result.Type: GrantFiled: March 8, 2019Date of Patent: January 19, 2021Assignee: GUANGDONG OPPO MOBILE TELECOMMUNICATIONS CORP., LTD.Inventor: Yan Chen
-
Patent number: 10878540Abstract: Systems and methods can automatically detect bad contrast ratios in a rendered frame of a game application. A frame in the game application can be divided into a plurality of pixel regions with each pixel regions having multiple pixels. The systems and methods can calculate luminance of the pixels in the pixel region and calculate a contrast ratio for the pixel region based on the luminance of the pixels in the pixel region. The contrast ratio of the pixel region can be used to determine whether it is sufficient to meet a threshold contrast ratio. The color of the pixel region can be automatically changed to a predefined color to indicate that contrast ratio is sufficient to meet the threshold contrast ratio.Type: GrantFiled: August 15, 2017Date of Patent: December 29, 2020Assignee: ELECTRONIC ARTS INC.Inventor: Karen Elaine Stevens
-
Patent number: 10824907Abstract: The disclosure relates to systems and methods for image processing. A trained deep learning model may be determined. An input image may be acquired. A processing result may be generated by processing the input image based on the trained deep learning model. The trained deep learning model may be obtained according to a process including: acquiring a preliminary deep learning model; acquiring a sample image; generating a plurality of sub-images based on the sample image; and training the preliminary deep learning model based on the plurality of sub-images to obtain the trained deep learning model.Type: GrantFiled: June 29, 2018Date of Patent: November 3, 2020Assignee: SHANGHAI UNITED IMAGING HEALTHCARE CO., LTD.Inventor: Chuanfeng Lyu
-
Patent number: 10812767Abstract: The present disclosure provides a white balance processing method. The white balance processing method includes processing an image to determine whether a light source is contained in a scene corresponding to the image; when the light source is contained in the scene, processing the image according to a first predetermined algorithm to acquire color temperature of the light source in the scene and performing a white balance processing on the image according to the color temperature of the light source; and when no light source is contained in the scene, performing the white balance processing on the image according to a second predetermined algorithm. The present disclosure further provides a white balance processing device, an electronic device and a non-transitory computer readable storage medium.Type: GrantFiled: September 12, 2018Date of Patent: October 20, 2020Assignee: GUANGDONG OPPO MOBILE TELECOMMUNICATIONS CORP., LTD.Inventor: Huichao Wang
-
Patent number: 10796145Abstract: A method and apparatus for separating a text and figure of a document image are provided. The method of separating the text and the figure of the document image includes acquiring a document image, dividing the document image into a plurality of regions of interest, acquiring a feature vector by using a two-dimensional (2D) histogram by resizing the regions of interest and extracting a connection component of the regions of interest, acquiring a transformation vector of the feature vector by using a kernel, obtaining a cluster center of the transformation vector, and performing clustering on the cluster center to acquire a supercluster, and classifying the supercluster into one of a text class and a figure class, based on the number of superclusters.Type: GrantFiled: June 28, 2018Date of Patent: October 6, 2020Assignee: Samsung Electronics Co., Ltd.Inventor: Valery Valerievich Anisimovskiy
-
Patent number: 10791341Abstract: A video decoder can be configured to generate a first bin stream by, for at least one bin of the first bin stream, determining a first probability estimation for the at least one bin of the first bin stream based on a first set of parameters and generate a second bin stream by, for at least one bin of the second bin stream, determining a second probability estimation for the at least one bin of the second bin stream based on a second set of parameters that are different parameters than the first set of parameters.Type: GrantFiled: October 9, 2018Date of Patent: September 29, 2020Assignee: Qualcomm IncorporatedInventors: Amir Said, Hilmi Enes Egilmez, Marta Karczewicz, Vadim Seregin, Li Zhang, Xin Zhao
-
Patent number: 10791304Abstract: The functional effect of having additional color-sensing cone types in the human eye is implemented by an encoding of some spectral information differently for the left and right eyes. This different encoding for identical features seen by the left and right eyes is interpreted as a perceptively different feature by the human brain, allowing additional spectral information to be conveyed through the limited tristimulus sensitivity of the human eye.Type: GrantFiled: August 3, 2017Date of Patent: September 29, 2020Assignee: Wisconsin Alumni Research FoundationInventors: Mikhail A. Kats, Bradley Gundlach
-
Patent number: 10713456Abstract: Signal detection and recognition employees coordinated illumination and capture of images under to facilitate extraction of a signal of interest. Pulsed illumination of different colors facilitates extraction of signals from color channels, as well as improved signal to noise ratio by combining signals of different color channels. The successive pulsing of different color illumination appears white to the user, yet facilitates signal detection, even for lower cost monochrome sensors, as in barcode scanning and other automatic identification equipment.Type: GrantFiled: March 4, 2019Date of Patent: July 14, 2020Assignee: Digimarc CorporationInventors: Jacob L. Boles, Alastair M. Reed, John D. Lord