Patents Examined by John Strege
  • Patent number: 9047678
    Abstract: A method of determining change in a state of an object using images of the object, the method including providing a first image and a second image of the object, the first image and the second image being spaced apart in time, performing a plurality of pixel-based change detection algorithms to obtain a plurality of output difference products/images containing change information and pseudo change information, combining the plurality of output difference products to form a hybrid output difference product, and thresholding the output difference product to detect changes in the object.
    Type: Grant
    Filed: April 16, 2012
    Date of Patent: June 2, 2015
    Assignee: Mississippi State University Research and Technology Corporation
    Inventors: Charles G. O'Hara, Anil Cheriyadat, Suyoung Seo, Veeraraghavan Vijayaraj
  • Patent number: 9042654
    Abstract: Embodiments describe an image retrieval apparatus. The image retrieval apparatus includes an unlabelled image selector for selecting one or more unlabelled image(s) from an image database; and a main learner for training in each feedback round of the image retrieval, estimating relevance of images in the image database and a user's intention, and determining retrieval results, wherein the main learner makes use of the unlabelled image(s) selected by the unlabelled image selector in the estimation. In addition, the image retrieval apparatus may also include an active selector for selecting, in each feedback round and according to estimation results of the main learner, one or more unlabelled image(s) from the image database for the user to label.
    Type: Grant
    Filed: April 1, 2013
    Date of Patent: May 26, 2015
    Assignee: FUJITSU LIMITED
    Inventors: Rujie Liu, Takayuki Baba, Susumu Endo, Shuichi Shiitani, Yusuke Uehara, Daiki Masumoto, Shigemi Nagata
  • Patent number: 9042653
    Abstract: A spreadsheet application associates data obtained from a captured image with a spreadsheet. For example, one or more images of physical data may be captured and translated into electronic data that is automatically associated with one or more spreadsheets. The formatting and underlying formulas of the data included within the captured image may be represented within a spreadsheet (e.g. highlighted data remains highlighted within the electronic spreadsheet). The data may also be compared with existing electronic data. For example, differences between the data in the captured image with the data in an existing spreadsheet may be used to update the existing spreadsheet. A display of a captured image may be also be augmented using data that is obtained from the captured image. For example, a chart may be created and displayed using data that is obtained from the captured image.
    Type: Grant
    Filed: January 24, 2011
    Date of Patent: May 26, 2015
    Assignee: Microsoft Technology Licensing, LLC
    Inventors: Amy Lin, Shahar Prish, Sherman Der, John Campbell
  • Patent number: 9036922
    Abstract: An image classification program that is executed by a computer includes a first processing step of calculating feature values on the basis of the pixel densities of images, a second processing step of performing calculating in a space defined by the feature values, and a third processing step of grouping images that correspond to feature values that have been divided by the clustering.
    Type: Grant
    Filed: October 1, 2007
    Date of Patent: May 19, 2015
    Assignee: NIKON CORPORATION
    Inventor: Hiroyuki Abe
  • Patent number: 9036929
    Abstract: A method for scanning and decoding encoded symbols comprises processing low resolution image data from a full field of view and/or high resolution image data from one or more windowed segments of the field of view to provide imaging that is easily adaptable to different types of symbols and varying environmental conditions. The scanning method can be switched between the low resolution mode and the high resolution mode automatically based on whether the low resolution data is sufficiently accurate to decode the symbol.
    Type: Grant
    Filed: September 10, 2012
    Date of Patent: May 19, 2015
    Assignee: Cognex Technology and Investment LLC
    Inventors: Laurens Nunnink, William H. Equitz
  • Patent number: 9031332
    Abstract: The present invention relates to a device for compressing a source image by means of a base layer, coding a base image, and of an enhancement layer, coding a residual existing between the source image and the base image, the source image and the base image being structured as blocks of pixels. According to the invention, such a device is characterized in that it comprises means for generating the enhancement layer of a block as a function of the quality of restitution of this block, determined by comparing the coding of this block in the source image with the coding of this block in the base image, and/or as a function of information coded by this block of pixels.
    Type: Grant
    Filed: August 6, 2004
    Date of Patent: May 12, 2015
    Assignee: Thomson Licensing
    Inventors: Dominique Thoreau, Philippe Guillotel, Philippe Bordes, Edouard Francois
  • Patent number: 9031331
    Abstract: A classification system and method enable improvements to classification with nearest class mean classifiers by computing a comparison measure between a multidimensional representation of a new sample and a respective multidimensional class representation embedded into a space of lower dimensionality than that of the multidimensional representations. The embedding is performed with a projection that has been learned on labeled samples to optimize classification with respect to multidimensional class representations for classes which may be the same or different from those used subsequently for classification. Each multidimensional class representation is computed as a function of a set of multidimensional representations of labeled samples, each labeled with the respective class. A class is assigned to the new sample based on the computed comparison measures.
    Type: Grant
    Filed: July 30, 2012
    Date of Patent: May 12, 2015
    Assignee: Xerox Corporation
    Inventors: Thomas Mensink, Jakob Verbeek, Gabriela Csurka, Florent Perronnin
  • Patent number: 9031358
    Abstract: Aspects of the present disclosure provide improved image resizing techniques including seam craving and seam lining that consider image content in addition to geometric constraints. In one aspect of the disclosure, the energy for each of the pixels of an image is determined in accordance with an energy function. A line energy is determined for each straight line of the image, and the line energy corresponds to a sum of the energy of the pixels in each straight line. A plurality of starting points are identified for a plurality of seams corresponding to the straight lines having the lowest line energy. Further, the paths for each of the seams are determined. The image may be resized by removing the seams from the image.
    Type: Grant
    Filed: March 15, 2013
    Date of Patent: May 12, 2015
    Assignee: QUALCOMM Incorporated
    Inventors: PhaniKumar K. Bhamidipati, Vijayalakshmi R. Raveendran
  • Patent number: 9031325
    Abstract: A system for automatically extracting or isolating structures or areas of interest (e.g., built-up structures such as buildings, houses, shelters, tents; agricultural areas; etc.) from HR/VHR overhead imagery data by way of making as little as a single pass through a hierarchical data structure of input image components (where pixels are grouped into components based on any appropriate definition or measure of dissimilarity between adjacent pixels of the input image) to identify candidate components (e.g., possible structures of interest) free of necessarily having to re-iterate the same operator configured with different threshold parameters for a plurality of values.
    Type: Grant
    Filed: August 29, 2013
    Date of Patent: May 12, 2015
    Assignee: DigitalGlobe, Inc.
    Inventor: Georgios Ouzounis
  • Patent number: 9025860
    Abstract: A document that includes a representation of a two-dimensional (2-D) image may be obtained. A selection indicator indicating a selection of at least a portion of the 2-D image may be obtained. A match correspondence may be determined between the selected portion of the 2-D image and a three-dimensional (3-D) image object stored in an object database, the match correspondence based on a web crawler analysis result. A 3-D rendering of the 3-D image object that corresponds to the selected portion of the 2-D image may be initiated.
    Type: Grant
    Filed: August 6, 2012
    Date of Patent: May 5, 2015
    Assignee: Microsoft Technology Licensing, LLC
    Inventors: Geoffrey G. Zweig, Eric J. Stollnitz, Richard Szeliski, Sudipta Sinha, Johannes Kopf
  • Patent number: 9025859
    Abstract: The disclosure is directed to creating an inertial sensor aided depth map of a scene. An embodiment of the disclosure captures at least a first image and a second image during movement of a device caused by a user while framing or recording the scene, compensates for rotation between the first image and the second image, calculates an amount of translation of the device between the first image and the second image, calculates a pixel shift of a plurality of key points of the first image and the second image, and estimates a depth to one or more of the plurality of key points of the first image and the second image.
    Type: Grant
    Filed: July 30, 2012
    Date of Patent: May 5, 2015
    Assignee: Qualcomm Incorporated
    Inventors: Subramaniam Venkatraman, Szepo Robert Hung, Ruben M. Velarde, Carlos Manuel Puig
  • Patent number: 9025843
    Abstract: A method and apparatus for reconstruction of a region of interest for an object is provided. The reconstruction of the object may be based on chords which may fill a part, all, or more than all of the region of interest. Using chords for reconstruction may allow for reducing data acquired and/or processing for reconstructing a substantially exact image of the ROI. Moreover, various methodologies may be used in reconstructing the image, such as backprojection-filtration, and modified filtration backprojection.
    Type: Grant
    Filed: February 20, 2012
    Date of Patent: May 5, 2015
    Assignee: The University of Chicago
    Inventors: Xiaochuan Pan, Yu Zou, Lifeng Yu, Chien-Min Kao, Martin King, Maryellen Giger, Dan Xia, Howard Halpern, Charles Pelizzari, Emil Y. Sidky, Seungryong Cho
  • Patent number: 9025835
    Abstract: A computer implemented method for modifying a digital image comprises padding two or more face regions in the image and digitally defining at least one combined padded region that includes the two or more face regions. At least one border of the combined padded region is collinear with a border of one of the individual regions. Each of the borders of the at least one combined padded region is selected so as to be collinear with at least one border of one of the individual padded regions.
    Type: Grant
    Filed: October 28, 2011
    Date of Patent: May 5, 2015
    Assignee: Intellectual Ventures Fund 83 LLC
    Inventor: Raymond William Ptucha
  • Patent number: 9020226
    Abstract: [Problem] To separate biological patterns more efficiently and effectively. [Solution] An image processing apparatus according to the present invention is characterized by comprising: an image obtaining means for obtaining an image in which a first biological pattern and a second biological pattern having different textures are superimposed; a normalizing means for normalizing density of the image on the basis of an average and a standard deviation of pixel values inside a local region in the image using a parameter in accordance with difference in texture between the first biological pattern and the second biological pattern to thereby separate the first biological pattern and the second biological pattern in the image; and a pattern collating means for collating a known pattern for collation with each of the first biological pattern and the second biological pattern separated by the normalizing means.
    Type: Grant
    Filed: August 2, 2011
    Date of Patent: April 28, 2015
    Assignee: NEC Corporation
    Inventor: Toshio Kamei
  • Patent number: 9020268
    Abstract: Annotation of a sequence of digitized images in multimedia data is aided by a computer analyzing the multimedia data to identify one or more objects and assigning each object to a respective role. The role assignment is determined by processing context information representing a model of the multimedia data.
    Type: Grant
    Filed: November 3, 2010
    Date of Patent: April 28, 2015
    Assignee: Siemens Aktiengsellschaft
    Inventors: Andreas Hutter, Thomas Riegel, Tobias Schwarze
  • Patent number: 9008450
    Abstract: A system and method for determining a preferred motion vector is provided, which may be used to estimate motion in a series of video image frames. The system may include a video encoder having a motion vector selector. The motion vector selector may include a first generator to generate a cost associated with motion vectors in a first direction, a selector to set a cross over point corresponding to a lowest cost motion vector in the first direction, a direction indicator to sample a set of motion vectors about the cross over point to select a preferred direction, and a second generator to generate a cost associated with the motion vectors in the preferred direction. The method may include determining a final motion vector having the lowest cost within a search window.
    Type: Grant
    Filed: June 27, 2013
    Date of Patent: April 14, 2015
    Assignee: Imagination Technologies Limited
    Inventor: Saif Choudhary
  • Patent number: 9008465
    Abstract: An index is provided that holds information about each image content item in a collection of items, For each image content item, a first information item identifying the image content item and its location on a network, and at least one of (i) a second information item identifying a signature value of an object in the image content, or (ii) identification of a recognized object in the image content.
    Type: Grant
    Filed: September 14, 2012
    Date of Patent: April 14, 2015
    Assignee: Google Inc.
    Inventors: Salih Burak Gokturk, Munjal Shah, Azhar Khan
  • Patent number: 9008431
    Abstract: In a character string extraction method, a character portion, a rim portion, a character frame, and a character string frame are set, a feature value of each image in the character portion and the rim portion is calculated for each character frame, a character string frame evaluation value is calculated based on the feature value for the character string frame, a position of the character string frame is moved on the paper sheet image, and the image in the character portion is extracted by using the character string frame at a position at which the character string frame evaluation value reaches a maximum.
    Type: Grant
    Filed: March 4, 2011
    Date of Patent: April 14, 2015
    Assignee: Glory Ltd.
    Inventors: Masanori Goto, Toru Yonezawa, Motoko Kuroiwa
  • Patent number: 9008435
    Abstract: Embodiments enable searching of portions of objects in images, including programmatically analyzing each image in a collection in order to determine image data that, for individual images in the collection, represents one or more visual characteristics of a portion of an object shown in that image. A user is enabled to specify one or more search criteria that includes image data, and a search result may be determined based on one or more images in the collection that show a corresponding object that has a portion that satisfies a threshold. The threshold is defined at least in part by the one or more search criteria.
    Type: Grant
    Filed: September 14, 2012
    Date of Patent: April 14, 2015
    Assignee: Google Inc.
    Inventors: Salih Burak Gokturk, Baris Sumengen, Diem Vu, Navneet Dalal, Danny Yang, Xiaofan Lin, Azhar Khan, Munjal Shah, Dragomir Anguelov, Lorenzo Torresani, Vincent Vanhoucke
  • Patent number: 9008432
    Abstract: Information is arranged within a predetermined shape. The predetermined shape is captured in an image taken by a tangible image-capturing element of a device which captures images at the direction of a user. The device includes a tangible processor that analyzes the image captured by the image-capturing element to determine whether a predetermined shape is present in the image. The tangible processor determines an alignment of the predetermined shape when the predetermined shape is present in the image, and reads data within the predetermined shape in accordance with the determined alignment. The device automatically forwards a request for content over a communications network in accordance with the data read within the predetermined shape. The device receives the content over the network and displays the content.
    Type: Grant
    Filed: December 10, 2010
    Date of Patent: April 14, 2015
    Assignee: Qyoo, LLC.
    Inventors: Jeffrey Berthiaume, John Ellis Stafira, Mark Alexander Stafford