Region Labeling (e.g., Page Description Language) Patents (Class 382/180)
  • Patent number: 10810745
    Abstract: A processor-implemented learning method for an image segmentation includes training first duplicate layers, as duplications of trained first layers of a pre-trained model, so that a second feature extracted from a target image by the trained first duplicate layers is matched to a first feature extracted from a training image by the trained first layers; regularizing the trained first duplicate layers so that a similarity between the first feature and a third feature extracted from the training image by the regularized first duplicate layers meets a threshold; and training second duplicate layers, as duplications of trained second layers of the pre-trained model, to be configured to segment the target image based on the regularized first duplicate layers, the trained second layer being configured to segment the training image.
    Type: Grant
    Filed: August 14, 2018
    Date of Patent: October 20, 2020
    Assignee: Samsung Electronics Co., Ltd.
    Inventors: Nahyup Kang, KeeChang Lee, Inwoo Ha
  • Patent number: 10783368
    Abstract: A method and apparatus for identifying an intersection in an electronic map, and a computer readable medium are provided. An embodiment of the method includes: acquiring boundary information related to road boundaries from an electronic map; determining a topological relationship between the road boundaries in an area having a predetermined size in the electronic map based on the boundary information; and determining a distribution of an intersection in the area based on the topological relationship. The apparatus corresponding to the method, the device implementing the method of the present disclosure, and the computer readable medium are also provided. Through the technical solutions, the intersection may be automatically identified by detecting the road boundaries, which improves the efficiency of producing a high-precision map, and has the advantage of high accurate recall rate, strong universality, or simple method.
    Type: Grant
    Filed: December 21, 2018
    Date of Patent: September 22, 2020
    Assignee: Baidu Online Network Technology (Beijing) Co., Ltd.
    Inventors: Yifei Zhan, Miao Yan, Wang Zhou, Xianpeng Lang, Xiong Duan, Changjie Ma
  • Patent number: 10769794
    Abstract: A method of object detection includes obtaining a set of images depicting overlapping regions of an area containing a plurality of objects. Each image includes input object indicators defined by input bounding boxes, input confidence level values, and object identifiers. The method includes identifying candidate subsets of input object indicators in adjacent images. Each candidate subset has input overlapping bounding boxes in a common frame of reference, and a common object identifier. The method includes adjusting the input confidence levels for each input object indicator in the candidate subsets; selecting clusters of the input object indicators satisfying a minimum input confidence threshold, having a common object identifier, and having a degree of overlap satisfying a predefined threshold; and detecting an object by generating a single output object indicator for each cluster, the output object indicator having an output bounding box, an output confidence level value, and the common object identifier.
    Type: Grant
    Filed: November 22, 2019
    Date of Patent: September 8, 2020
    Assignee: Symbol Technologies, LLC
    Inventors: Joseph Lam, Xinyi Gong
  • Patent number: 10764561
    Abstract: A depth sensing system receives an image pair showing objects in a scene. The system generates binary hashes for pixels in the image pair by performing a random walk. The system matches pixels in the first image to pixels in the second image that depict the same point in the scene by generating cost values representing differences between the binary hashes for pairs of pixels in the images. The system generates a disparity map containing disparity vectors representing coordinate differences between matched pixels in the first and second images. The system generates and outputs a depth map based on the disparity map. The depth map represents the distances between an image acquisition system that acquired the image pair and the objects in the scene.
    Type: Grant
    Filed: April 4, 2017
    Date of Patent: September 1, 2020
    Assignee: COMPOUND EYE INC
    Inventor: Jason Devitt
  • Patent number: 10748035
    Abstract: An active learning system classifies multiple objects in an input image from the set of images with a classification metric indicative of uncertainty of each of the classified object to belong to one or different classes and determines a diversity metric of the input image indicative of diversity of the objects classified in the input image. The active learning system evaluates the diversity metric of the input image and to cause rendering of the input image on a display device based on a result of the evaluation and trains the classifier using the labelled objects of the input image.
    Type: Grant
    Filed: July 5, 2018
    Date of Patent: August 18, 2020
    Assignee: Mitsubishi Electric Research Laboratories, Inc.
    Inventor: Teng-Yok Lee
  • Patent number: 10726200
    Abstract: A computer software that provides the user with the means to import an image of a paper financial document for data extraction. The extracted data automatically populates a financial datasheet and can be synchronized with a company financial record being kept on an external accounting software. The present invention provides the user with the convenience of automatic data input and eliminates the traditional method of individually inputting financial transactions into the accounting software.
    Type: Grant
    Filed: June 10, 2011
    Date of Patent: July 28, 2020
    Inventor: Benjamin Chou
  • Patent number: 10706188
    Abstract: Systems and methods are provided for implementing a parallel Expectation Minimization algorithm for generalized latent variable models. Item response data that is based on responses to items from multiple respondents is accessed. The item response data includes data for multiple response variables. The item response data is analyzed using a generalized latent variable model, and the analysis includes an application of a Parallel-E Parallel-M (PEPM) algorithm. In a parallel Expectation step of the PEPM algorithm, the respondents are subdivided into N groups of respondents, and computations for the N groups are performed in parallel using the N processor cores. In a parallel Maximization step of the PEPM algorithm, the response variables are subdivided into N groups of response variables, and computations for the N groups of response variables are performed in parallel using the N processor cores.
    Type: Grant
    Filed: November 10, 2016
    Date of Patent: July 7, 2020
    Assignee: Educational Testing Service
    Inventor: Matthias von Davier
  • Patent number: 10635927
    Abstract: Performing semantic segmentation of an image can include processing the image using a plurality of convolutional layers to generate one or more feature maps, providing at least one of the one or more feature maps to multiple segmentation branches, and generating segmentations of the image based on the multiple segmentation branches, including providing feedback to, or generating feedback from, at least one of the multiple segmentation branches in performing segmentation in another of the segmentation branches.
    Type: Grant
    Filed: March 5, 2018
    Date of Patent: April 28, 2020
    Assignee: HONDA MOTOR CO., LTD.
    Inventors: Yi-Ting Chen, Athmanarayanan Lakshmi Narayanan
  • Patent number: 10621453
    Abstract: Disclosed herein is method and system for determining relationship among text segments in signboards for navigating autonomous vehicles. Images of signboards are captured analyzed to determine text segments in the images. Further, relationship among the identified text segments is determined based on relationship among plurality of text regions forming the text segments. Finally, information related to relationship of the text segments is provided to a navigation unit in the autonomous vehicle for facilitating navigation of the autonomous vehicle. In an embodiment, the method of present disclosure helps in eliminating prospective errors in the text identification process due to irregular arrangement of text segments in the signboards, by determining relationship among localized text segments in the images of the signboards.
    Type: Grant
    Filed: January 17, 2018
    Date of Patent: April 14, 2020
    Assignee: Wipro Limited
    Inventors: Balaji Govindaraj, Sujatha Jagannath
  • Patent number: 10614569
    Abstract: A mobile processing device system for patient monitoring data acquisition includes a repository of information. The information associates a particular patient monitoring device type for displaying a particular patient parameter with a particular text label identifying the particular patient parameter. A portable processing device includes an imaging device for acquiring image data representing an image presenting patient parameter data from the particular patient monitoring device type. An image recognition processor uses the information, for analyzing the image data to identify the particular text label identifying the particular patient parameter and a value of the particular patient parameter. An output processor communicates data representing the particular patient parameter and the value to a destination.
    Type: Grant
    Filed: July 28, 2017
    Date of Patent: April 7, 2020
    Assignee: CERNER INNOVATION, INC.
    Inventor: Robert A. Neff
  • Patent number: 10579901
    Abstract: A method of comparing objects in images. A dictionary determined from a plurality of feature vectors formed from a test image and codes formed by applying the dictionary to the feature vectors, is received. The dictionary is based on a modified manifold obtained by determining correspondences for codes using pairwise similarities between codes. Comparison codes are determined for the objects in the images by applying the dictionary to feature vectors of the objects in the images. The objects in the images are compared based on the comparison codes of the objects.
    Type: Grant
    Filed: December 5, 2017
    Date of Patent: March 3, 2020
    Assignee: Canon Kabushiki Kaisha
    Inventors: Getian Ye, Ka Ming Leung
  • Patent number: 10521105
    Abstract: Example apparatus and methods concern establishing, managing, or dis-establishing a primary hover-point for a hover-sensitive input/output interface. One example apparatus includes a proximity detector that detects an object in a hover-space associated with the input/output interface. The apparatus produces characterization data concerning the object. The characterization data may identify where the object is located, how the object is moving, what the object is doing, or other attributes of the object. The apparatus may assign a hover point designation to the object as a function of the characterization data. The apparatus selectively controls input actions associated with the object based on the hover point designation. The apparatus may accept input actions associated with a primary hover point and ignore actions associated with a non-primary hover point.
    Type: Grant
    Filed: June 14, 2018
    Date of Patent: December 31, 2019
    Assignee: Microsoft Technology Licensing, LLC
    Inventors: Lynn Dai, Dan Hwang
  • Patent number: 10521914
    Abstract: A method of object detection includes obtaining a set of images depicting overlapping regions of an area containing a plurality of objects. Each image includes input object indicators defined by input bounding boxes, input confidence level values, and object identifiers. The method includes identifying candidate subsets of input object indicators in adjacent images. Each candidate subset has input overlapping bounding boxes in a common frame of reference, and a common object identifier. The method includes adjusting the input confidence levels for each input object indicator in the candidate subsets; selecting clusters of the input object indicators satisfying a minimum input confidence threshold, having a common object identifier, and having a degree of overlap satisfying a predefined threshold; and detecting an object by generating a single output object indicator for each cluster, the output object indicator having an output bounding box, an output confidence level value, and the common object identifier.
    Type: Grant
    Filed: September 7, 2017
    Date of Patent: December 31, 2019
    Assignee: Symbol Technologies, LLC
    Inventors: Joseph Lam, Xinyi Gong
  • Patent number: 10521734
    Abstract: A computing device predicts an event or classifies an observation. A trained labeling model is executed with unlabeled observations to define a label distribution probability matrix used to select a label for each observation. Unique combinations of observations selected from the unlabeled observations are defined. A marginal distribution value is computed from the label distribution probability matrix. A joint distribution value is computed between observations included in each combination. A mutual information value is computed for each combination as a combination of the marginal distribution value and the joint distribution value computed for the respective combination. A predefined number of observation vector combinations is selected from the combinations that have highest values for the computed mutual information value. Labeled observation vectors are updated to include each observation vector included in the selected observation vector combinations with a respective obtained label.
    Type: Grant
    Filed: May 7, 2019
    Date of Patent: December 31, 2019
    Assignee: SAS Institute Inc.
    Inventors: Xu Chen, Jorge Manuel Gomes da Silva
  • Patent number: 10510147
    Abstract: An image processing device includes: a plurality of label data generation units which generate label data in which a predetermined label is assigned to each of a plurality of pixels in each of a plurality of divided images; a plurality of label integration information generation units which correspond to the respective label data generation units and generate label integration information representing association of labels included in the label data in order to integrate label data generated by a corresponding label data generation unit and label data generated by another label data generation unit; a plurality of label integration units which generate integrated label data in which respective pieces of label data corresponding to neighboring divided images are integrated; and a label integration processing control unit which distributes the label data to the respective label integration units such that computational loads to integrate the label data are equalized.
    Type: Grant
    Filed: November 9, 2017
    Date of Patent: December 17, 2019
    Assignee: OLYMPUS CORPORATION
    Inventors: Dai Ito, Tatsuki Tadenuma
  • Patent number: 10489686
    Abstract: An object detection system for an autonomous vehicle processes sensor data, including one or more images, obtained for a road segment on which the autonomous vehicle is being driven. The object detection system compares the images to three-dimensional (3D) environment data for the road segment to determine pixels in the images that correspond to objects not previously identified in the 3D environment data. The object detection system then analyzes the pixels to classify the objects not previously identified in the 3D environment data.
    Type: Grant
    Filed: March 3, 2017
    Date of Patent: November 26, 2019
    Assignee: UATC, LLC
    Inventor: Carlos Vallespi-Gonzalez
  • Patent number: 10482583
    Abstract: Implementations relate to generating and displaying blur in images. In some implementations, a method includes generating a plurality of mipmap images based on an input image, including applying a blur to a respective plurality of pixels derived from the input image for each mipmap image. In some examples, the blur is at least partially based on depth data for the image. Parameter data is obtained that indicates an output focal plane depth for an output focal plane of an output image and an output focal range in front of the output focal plane. Output pixel values of the output image are generated, including determining blurred pixel values based on one or more of the mipmap images selected based on the output focal plane depth and the output focal range. The blurred pixel values are based on particular pixels associated with a depth outside the output focal range.
    Type: Grant
    Filed: May 10, 2018
    Date of Patent: November 19, 2019
    Assignee: Google LLC
    Inventor: Austin Suszek
  • Patent number: 10460219
    Abstract: In an example, a method is described that includes dividing an input image into a plurality of strips, where each strip is smaller than a whole of the input image. A plurality of binary images is then generated, where each of the binary images corresponds to one of the strips. Connected component labeling is performed on the binary images, one binary image at a time. An object map for the input image is then generated based on the results of the connected component labeling.
    Type: Grant
    Filed: March 31, 2016
    Date of Patent: October 29, 2019
    Assignees: Hewlett-Packard Development Company, L.P., Purdue Research Foundation
    Inventors: Brent Michael Bradburn, Zuguang Xiao, Mengqi Gao, Lu Wang, Jan P. Allebach
  • Patent number: 10445854
    Abstract: An image processing device includes a label memory to store a first label assigned to a pixel in a previous line of a processing target image subjected to an assignment process prior to the assignment process on a processing target line of the processing target image, a label area memory to store identification information identifying a label area including pixels assigned with an identical label, and circuitry to, if at least one target pixel in the processing target line is an effective pixel, perform the assignment process on the target pixel based on at least one label that includes the first label and is assigned to at least one peripheral pixel peripheral to the target pixel, and to perform an updating process of updating the identification information based on a position of the target pixel and the label assigned to the target pixel.
    Type: Grant
    Filed: May 4, 2017
    Date of Patent: October 15, 2019
    Assignee: Ricoh Company, Ltd.
    Inventors: Shohichi Satoh, Dai Kurabayashi, Deyong Si
  • Patent number: 10409895
    Abstract: Embodiments of the present disclosure relate to a document layout system that automatically optimizes dynamic documents (i.e., documents having dynamic content). In particular, the document layout system intelligently populates and arranges content items, including dynamic content items, into a dynamic document. Further, upon receiving dynamic content updates that change the dimensions (e.g., expand or shrink) of one or more dynamic content items within a dynamic document, the document layout system automatically modifies the arrangement of the content items to resolve any position conflicts detected between content items within the dynamic document. Further, the document layout system provides a number of features that assist users in organizing and arranging dynamic documents.
    Type: Grant
    Filed: October 17, 2017
    Date of Patent: September 10, 2019
    Assignee: Qualtrics, LLC
    Inventor: Owen Hancock
  • Patent number: 10402629
    Abstract: Methods, systems, and apparatus, including computer programs encoded on a computer storage medium, for facial recognition using fractal features are disclosed. In one aspect, a method includes the actions of accessing data encoding a facial image, the facial image including a face. The actions further include generating a hierarchical graphical model of the face in the facial image, the hierarchical graphical model including more than nodes from at least one level, each level approximating the face within a contour. The actions further include applying a bank of filters to the face at a particular node, each filter spanning more than one scale and at least one direction. The actions further include analyzing filter responses from the bank of filters applied at each direction to obtain a similarity measure that consolidates filter responses from filters applied at more than one scale. The actions further include generating a vector representation.
    Type: Grant
    Filed: October 5, 2018
    Date of Patent: September 3, 2019
    Assignee: MorphoTrust USA, LLC
    Inventor: Alexandru Bogdan
  • Patent number: 10345912
    Abstract: A control method for an electronic device comprising a sensor unit having a first detection range and a second detection range measured from the electronic device, and the second detection range being smaller than the first detection range. The method includes: detecting a 3D movement track of an operation body when the relative distance between the operation body and the electronic device is smaller than the first detection range, the 3D movement track being a continuous movement track having one end between the first and second detection range and the other end equal to or smaller than the second detection range; generating a status change signal when the operation body reaches close to or moves away from the second detection range of the sensor unit; executing a corresponding control command according to the 3D movement track and the status change signal.
    Type: Grant
    Filed: March 7, 2012
    Date of Patent: July 9, 2019
    Assignees: LENOVO (BEIJING) CO., LTD., BEIJING LENOVO SOFTWARE LTD.
    Inventors: Junfeng Liu, Qianying Wang, Yuan Gao
  • Patent number: 10325372
    Abstract: Techniques for providing an accurate auto-crop feature for images captured by an image capture device may be described herein. For example, one or more image masks for a color image captured by an image capture device may be received by a computer system. Metadata about the color image that identifies portions of the image as foreground and the color image itself may also be received by the computer system. Further, a representation of a user and a floor region associated with a user may be extracted from the color image using the one or more image masks and the metadata. A first area of the color image may be cropped with respect to the extracted representation of the user and the floor region associated with the user to generate a second area of the color image. In embodiments, a third area of the color image may be obscured based on the received metadata.
    Type: Grant
    Filed: December 20, 2016
    Date of Patent: June 18, 2019
    Assignee: Amazon Technologies, Inc.
    Inventors: Amit Kumar Agrawal, Alexander Adrian Hugh Davidson, Prakash Ramu
  • Patent number: 10318572
    Abstract: A system, method, and media are provided for generating a structured labeling graphical user interface. The user interface receives user input that associates multimedia content with categories. The user input may include user-supplied tags that further define the category for the multimedia content. The user-supplied tags are rendered proximate to the categories. In turn, a database logs user events to store, among other things, the categories, the user-supplied tags, time associated with completing the user-supplied tags, and time for associating multimedia content with the categories or tags.
    Type: Grant
    Filed: February 10, 2014
    Date of Patent: June 11, 2019
    Assignee: MICROSOFT TECHNOLOGY LICENSING, LLC
    Inventors: Todd Kulesza, Denis Charles, Rich Caruana, Saleema Amin Amershi, Danyel Aharon Fisher
  • Patent number: 10297028
    Abstract: The image data analytics for computation configuration method and computer system with computation accessibility is provided. The computation accessibility and configuration on image data processing are implemented by a plurality of computers having a plurality of processors and a plurality of data storages, the image data analytics for computation accessibility and configuration includes the following steps: inputting an original image by an input device and initializing the original image; defining a plurality of tiles equally dividing the original image into a same size of a regular shape; transferring the plurality of image regions to form a graph having vertices and edges; cutting the graph into a plurality of sub-graphs; arranging the plurality of sub-graphs to the plurality of processors or cores to conduct parallel processing simultaneously for analyzing the image data; and storing a plurality of processing results respectively in the plurality of data storages.
    Type: Grant
    Filed: July 10, 2017
    Date of Patent: May 21, 2019
    Assignee: National Cheng Kung University
    Inventors: Gwo Giun Lee, Shih-Yu Hung, Tai-Ping Wang
  • Patent number: 10249044
    Abstract: In one embodiment, a method includes detecting one or more objects in an image, generating at least one mask for each of the detected objects, wherein each of the masks is defined by a perimeter, classifying the detected objects, receiving gesture input in relation to the image, determining whether one or more locations associated with the gesture input correlate with any of the masks, and providing feedback regarding the image in response to the gesture input. Each of the masks may include data identifying the corresponding detected object, and the perimeter of each mask may correspond to a perimeter of the corresponding detected object. The perimeter of the corresponding detected object may separate the detected object from one or more portions of the image that are distinct from the detected object.
    Type: Grant
    Filed: December 30, 2016
    Date of Patent: April 2, 2019
    Assignee: Facebook, Inc.
    Inventors: Vincent Charles Cheung, Connie Yeewei Ho, Balmanohar Paluri
  • Patent number: 10217027
    Abstract: A recognition training apparatus according to the present invention includes a generation unit configured to generate relevance between a specific domain and a candidate of a recognition target based on ontology information expressing a concept structure of the specific domain, a selection unit configured to select the recognition target from the candidate of the recognition target based on the relevance generated by the generation unit, and a training unit configured to train a recognizer using training data regarding the recognition target selected by the selection unit.
    Type: Grant
    Filed: January 13, 2017
    Date of Patent: February 26, 2019
    Assignee: Canon Kabushiki Kaisha
    Inventors: Hirotaka Hachiya, Masakazu Matsugu
  • Patent number: 10204395
    Abstract: Virtual reality is provided by a mobile VR device. VR scenes are displayed by serving an application's image requests from a local cache of pre-rendered image frames. As the device user moves, image frames that match a VR scene viewed from a current pose (3D position and 3D orientation of the device translated to the virtual environment) are retrieved from the cache and displayed to the user as a VR scene. The cache may include a static cache including static image frames and a dynamic object cache including dynamic object image frames. The static cache may be indexed by the device current pose. The dynamic object cache may be indexed by the dynamic object's animation stage, orientation, and relative distance from the device user's current pose for a given time or movement-based trigger. Static image frames and dynamic object image frames may be composited, and displayed as a scene.
    Type: Grant
    Filed: October 19, 2016
    Date of Patent: February 12, 2019
    Assignee: Microsoft Technology Licensing, LLC
    Inventors: Eduardo Cuervo, Kevin Boos, David Chu
  • Patent number: 10186061
    Abstract: An approach to facilitating proximate-tile-type-based map generation is provided. Tile type information that indicates tile types for map tiles of a map to be generated may be received. One or more proximate tile types may be determined for individual ones of the map tiles based on the tile types. The proximate tile types may include one or more of the tile types of other ones of the map tiles that are proximate the individual map tiles on the map. Level information that indicates levels for the individual map tiles may be determined based on the proximate tile types of the individual map tiles. The map may be generated based on the tile type information and the level information. In some implementations, the tile type information may be received as a graphic representation having colors that indicate the tile types.
    Type: Grant
    Filed: January 11, 2013
    Date of Patent: January 22, 2019
    Assignee: Electronic Arts Inc.
    Inventor: Nate Boxer
  • Patent number: 10178149
    Abstract: Design analysis for framework assessment is described. A file including a designed image may be obtained from a storage device, and a guide structure may be extracted from the file. The guide structure may include multiple guide lines, and the guide lines may be analyzed to determine column-related data, such as at least relative widths or positional orders, a total number of columns, groups of column widths, a combination thereof, and so forth. From multiple frameworks, a framework may be ascertained that substantially matches the guide structure based at least partially on a comparison of column-related data to each framework of the multiple frameworks. Ascertainment of a framework may be indicated to an end-user. The ascertained framework may be applied to facilitate development of a coded functional version of at least a visual design of the image.
    Type: Grant
    Filed: May 14, 2015
    Date of Patent: January 8, 2019
    Assignee: Adobe Inc.
    Inventor: Jonathan Joshua Pimento
  • Patent number: 10176253
    Abstract: According to some embodiments of the present invention there is provided a computerized method for labeling a cluster of text documents. The method comprises receiving a document cluster and producing automatically multiple document sub-clusters determined by randomly changing some documents. The method applies multiple cluster labeling algorithms on the cluster and on each sub-cluster, to generate ordered lists. The method comprises generating a ranked label list for each cluster labeling algorithm by computing automatically label values, one for each cluster label in the lists of the selected algorithm, and re-ranking the ordered list. The method combines the re-ranked label lists using a label fusing algorithm to produce a fused label list.
    Type: Grant
    Filed: January 28, 2015
    Date of Patent: January 8, 2019
    Assignee: International Business Machines Corporation
    Inventors: Shay Hummel, Haggai Roitman, Michal Shmueli-Scheuer
  • Patent number: 10147191
    Abstract: A panoramic video camera includes a lens having a horizontal FOV portion of 360° about an optical axis. The panoramic video camera includes an image sensor, a processor, and a storage medium. The processor executes programmatic instructions to analyze digital image files to locate an object and positions thereof to produce object position data. Executed programmatic instructions also generate a plurality of zoom levels for the respective digital image files based on the object position data. The zoom levels enable a playback device to visually show the object at varying magnifications during playback of a video segment derived from the digital image files. Zoom levels may alter image capture settings of the panoramic video camera, may alter playback device playback settings, or both.
    Type: Grant
    Filed: July 26, 2016
    Date of Patent: December 4, 2018
    Assignee: 360FLY, INC.
    Inventors: Claudio S. Ribeiro, Michael J. Harmon, Felippe M. Bicudo
  • Patent number: 10140510
    Abstract: Computer program products for discriminating hand and machine print from each other, and from signatures, are disclosed and include program code readable and/or executable by a processor to: receive an image, determine a color depth of the image; reducing the color depth of non-bi-tonal images to generate a bi-tonal representation of the image; identify a set of one or more graphical line candidates in either the bi-tonal image or the bi-tonal representation, the graphical line candidates including true graphical lines and/or false positives; discriminate any of the true graphical lines from any of the false positives; remove the true graphical lines from the bi-tonal image or the bi-tonal representation without removing the false positives to generate a component map comprising connected components and excluding graphical lines; identify one or more of the connected components in the component map; and output and/or display and indicator of each of the connected components.
    Type: Grant
    Filed: March 2, 2018
    Date of Patent: November 27, 2018
    Assignee: KOFAX, INC.
    Inventors: Alexander Shustorovich, Christopher W. Thrasher, Anthony Macciola, Jan W. Amtrup
  • Patent number: 10142540
    Abstract: A first panoramic video camera of the multi-camera imaging system captures images to produce captured panoramic images. At least some panoramic images include images of an object being tracked by the multi-camera imaging system. The first panoramic video camera generates a set of digital image files based on the captured panoramic images. Each digital image file of the set of digital image files is time sequenced to reflect a time at which a panoramic image was captured. The first panoramic video camera receives over a, first wireless communication network, a control message relating to tracking the object. Responsive to the control message, the first panoramic video camera transmits at least a portion of the set of digital image files over a second wireless communication network, which supports higher data rate transmission than does the first wireless communication network.
    Type: Grant
    Filed: July 26, 2016
    Date of Patent: November 27, 2018
    Assignee: 360FLY, INC.
    Inventors: Claudio S. Ribeiro, Michael J. Harmon, Felippe M. Bicudo
  • Patent number: 10123052
    Abstract: A data processing method comprising receiving a digital image file comprising a plurality of digital channel datasets corresponding to a plurality of image channels, wherein the digital image file includes an image region and a non-image region; determining a boundary for the image region; for each channel in the plurality of channels, generating a buffer region by using the digital channel datasets, identifying a plurality of pixels along the boundary and for each pixel of the plurality of pixels, copying the pixel one or more times outward into the non-image region in digital memory; and storing a modified digital image file that includes the buffer region.
    Type: Grant
    Filed: November 18, 2016
    Date of Patent: November 6, 2018
    Assignee: Mapbox, Inc.
    Inventor: Damon Burgett
  • Patent number: 10108708
    Abstract: The invention provides an apparatus, system and method for the comparison between songs to analyze how similar each song is to the next using a new system of classification. A new and improved method and apparatus for the weighting of these comparable variables between songs in order to find the best objective criteria for finding a likeness between songs is disclosed. The time complexity associated with the computation of the optimal path for smoothing a playlist of songs is substantially reduced using the system of classification.
    Type: Grant
    Filed: March 31, 2016
    Date of Patent: October 23, 2018
    Assignee: SPOTIFY AB
    Inventors: Brendan O'Driscoll, Craig Watson, Aidan Sliney, George Boyle, Clodagh McCarthy Luddy
  • Patent number: 9934444
    Abstract: According to an embodiment, an image processing apparatus includes an interface that receives a first image and a second image. A processor extracts a first character string appearing on the first image, and a second character string appearing on the second image. The processor determines that both the first character string and the second character string correspond to a photographed subject in the first image and second image. The processor cuts out each of the characters constituting the first character string, in units of a character. The processor cuts out each of the characters constituting the second character string, in units of a character. The processor calculates a first similarity degree group and a second similarity degree group. The processor recognizes a character string of the subject, based on the first similarity degree group and the second similarity degree group.
    Type: Grant
    Filed: January 30, 2017
    Date of Patent: April 3, 2018
    Assignee: TOSHIBA TEC KABUSHIKI KAISHA
    Inventors: Masaaki Yasunaga, Kazuki Taira
  • Patent number: 9830710
    Abstract: A method includes receiving an image comprising a human head from an imaging device. Further, the method includes grouping one or more pixels of the image into one or more super pixels. Furthermore, the method includes generating a polar coordinate grid for the image, wherein the polar coordinate grid comprises one or more grid points. Additionally, the method includes labeling the one or more grid points as one of a face region, a hair region, and a background region based on tiered structure constraint. The method also includes labeling the one or more super pixels as one of the face region, the hair region, and the background region based on the one or more labels of the one or more grid points. Further, the method includes transmitting a notification comprising the one or more labels of the one or more super pixels.
    Type: Grant
    Filed: December 16, 2015
    Date of Patent: November 28, 2017
    Assignee: General Electric Company
    Inventors: Peter Henry Tu, Jixu Chen, Yang Yu
  • Patent number: 9818035
    Abstract: According to an embodiment, an image processing apparatus includes an interface that receives a first image and a second image. A processor extracts a first character string appearing on the first image, and a second character string appearing on the second image. The processor determines that both the first character string and the second character string correspond to a photographed subject in the first image and second image. The processor cuts out each of the characters constituting the first character string, in units of a character. The processor cuts out each of the characters constituting the second character string, in units of a character. The processor calculates a first similarity degree group and a second similarity degree group. The processor recognizes a character string of the subject, based on the first similarity degree group and the second similarity degree group.
    Type: Grant
    Filed: January 30, 2017
    Date of Patent: November 14, 2017
    Assignee: TOSHIBA TEC KABUSHIKI KAISHA
    Inventors: Masaaki Yasunaga, Kazuki Taira
  • Patent number: 9800891
    Abstract: A method for reducing visual artifacts in a frame of a digital video signal, which is coded by blocks and then decoded, includes defining a block type is according to the coding method for a block selected according to a predetermined set of coding types. In the method filtering is performed to reduce visual artifacts due to a block boundary The filtering performed depends on block types of the frame in the environment of the block boundary.
    Type: Grant
    Filed: January 19, 2001
    Date of Patent: October 24, 2017
    Assignee: Nokia Technologies Oy
    Inventors: Ossi Kalevo, Emre Aksu, Marta Karczewicz
  • Patent number: 9792694
    Abstract: A method for segmenting a target image includes receiving the target image of an anatomical structure, registering a plurality of atlases to the target image, each of the atlases including an image and a plurality of labels corresponding to portions of the image, selecting a plurality of registered atlases, transferring the labels of selected registered atlases to the target image, combining the labels that are transferred to the target image using a fusion of a discriminative model and a generative model, and outputting a segmentation of the target image isolating the anatomical structure, wherein a segmentation of the target image is displayed.
    Type: Grant
    Filed: September 26, 2016
    Date of Patent: October 17, 2017
    Assignee: INTERNATIONAL BUSINESS MACHINES CORPORATION
    Inventors: Rahil Garnavi, Xi Liang, Suman Sedai
  • Patent number: 9779331
    Abstract: Provided is a method and system of tracking partially occluded objects using an elastic deformation model. According to an exemplary method and system, partially occluded vehicles are detected and tracked in a scene including side-by-side drive-thru lanes.
    Type: Grant
    Filed: February 26, 2015
    Date of Patent: October 3, 2017
    Assignee: Conduent Business Services, LLC
    Inventors: Orhan Bulan, Yusuf Oguzhan Artan, Aaron Michael Burry
  • Patent number: 9740692
    Abstract: Disclosed are systems, computer-readable mediums, and methods for creating a flexible structure description. To create the flexible structure description an image of a document of a particular document type that contains a table is received. An entry describing an item in the table is received. Title elements within the document are searched for based upon the entry. Data fields and anchor elements are detected for the entry. A flexible structure description for the particular document type is generated that includes a set of search elements for each data field in the image of the document and the title elements. The flexible structure description is matched against the image. Data from the image is extracted based upon the matching of the flexible structure description against the image.
    Type: Grant
    Filed: November 5, 2014
    Date of Patent: August 22, 2017
    Assignee: ABBYY Development LLC
    Inventors: Sergei Golubev, Irene Filimonova, Sergey Zlobin
  • Patent number: 9679226
    Abstract: An image processing system automatically segments and labels an image using a hierarchical classification model. A global classification model determines initial labels for an image based on features of the image. A label-based descriptor is generated based on the initial labels. A local classification model is then selected from a plurality of learned local classification model based on the label-based descriptor. The local classification model is applied to the features of the input image to determined refined labels. The refined labels are stored in association with the input image.
    Type: Grant
    Filed: March 27, 2015
    Date of Patent: June 13, 2017
    Assignee: Google Inc.
    Inventors: Qixing Huang, Mei Han, Bo Wu, Sergey Ioffe
  • Patent number: 9646203
    Abstract: A method, apparatus and computer program product are provided for generating map data based on construction designs. In the context of a method, the method includes receiving a construction design, performing optical character recognition to detect characters within the construction design, extracting design features within the construction design, and generating map data based on the detected characters, and the extracted design features.
    Type: Grant
    Filed: July 18, 2014
    Date of Patent: May 9, 2017
    Assignee: HERE Global B.V.
    Inventor: Xin Chen
  • Patent number: 9639558
    Abstract: An image is built. The building of an image includes selecting an image to be built, performing a hashing function on the image to produce a checksum, querying a registry for an existing image including the same checksum, and if no existing image including the same checksum is found, acquiring a parent image for the image to be built, building the image to produce a built image, adding the checksum for the image to the built image, and storing the built image in the registry.
    Type: Grant
    Filed: September 17, 2015
    Date of Patent: May 2, 2017
    Assignee: INTERNATIONAL BUSINESS MACHINES CORPORATION
    Inventor: Benjamin P. Alton
  • Patent number: 9613267
    Abstract: This disclosure provides an exemplary method and system for extracting structured label and value pairwise textual data from a textual document. According to an exemplary method, initially a layout analysis is performed resulting in one or more alternatives for grouping and ordering the textual elements of interest. Next, textual elements are tagged as including a label term, a value term or a label and value term. Finally, a sequence-based method is applied to the tagged elements to generate one or more sequence listings representative of the label and value pairwise data structure(s) and label:value pairwise data is extracted.
    Type: Grant
    Filed: September 3, 2014
    Date of Patent: April 4, 2017
    Assignee: Xerox Corporation
    Inventors: Hervé Déjean, Thierry Lehoux, Eric H. Cheminot
  • Patent number: 9501722
    Abstract: In an image classifying apparatus, an image is designated as a designated image from among plural candidate images, the designated image being one of images to be contained in at least one group, attribute information is obtained from the designated image, selecting conditions of images are decided based on the attribute information obtained from the designated image, at least one image is selected as a selected image to be contained in the group, from among the plural candidate images based on the selecting conditions, and both the designated image and the selected images are classified into the group.
    Type: Grant
    Filed: July 2, 2013
    Date of Patent: November 22, 2016
    Assignee: CASIO COMPUTER CO., LTD.
    Inventor: Yoshiki Ishige
  • Patent number: 9449384
    Abstract: A method registers a source image with a target image, wherein the images are deformable, by first measuring dissimilarity between the source image and the target image. The dissimilarity minimized using a discrete energy function. At multiple scales, multi-scale Markov random field registration is applied to the source and target images to determine a deformation vector field. Then, the target image is warped according to the deformation field vector to obtain a warped target mage registered to the source image.
    Type: Grant
    Filed: April 30, 2014
    Date of Patent: September 20, 2016
    Assignee: Mitsubishi Electric Research Laboratories, Inc.
    Inventors: Jeroen van Baar, Dohyung Seo
  • Patent number: 9411445
    Abstract: A processing system for an input device includes a sensor module and a determination module. The sensor module includes sensor circuitry coupled to transmitter electrodes and receiver electrodes. The sensor module is configured to transmit transmitter signals with the transmitter electrode and receive resulting signals with the receiver electrodes. The determination module is configured to obtain a set of measurements for an input object detected in a sensing region of the input device from a capacitive image, and perform, based on the set of measurements, a multistage classification to obtain a classification result identifying a type of the input object. wherein the capacitive image is generated based on the resulting signals.
    Type: Grant
    Filed: June 27, 2013
    Date of Patent: August 9, 2016
    Assignee: Synaptics Incorporated
    Inventors: Chia-yun Kuan, Shwetank Kumar