Patents Examined by Brian P. Werner
  • Patent number: 9613255
    Abstract: Methods, systems, and computer program products for signature detection. One example of a method includes: acquiring an article defect density map comprising a plurality of sections corresponding to a first resolution level which is indicative of defect numbers for the sections, and determining a distribution representative of the defect numbers or function thereof; determining a threshold in accordance with said distribution, and identifying sections, out of said plurality of sections in the article defect density map, with defect numbers or function thereof above the threshold; and clustering at least part of adjoining identified sections, into one or more signatures, thus detecting said one or more signatures.
    Type: Grant
    Filed: March 30, 2015
    Date of Patent: April 4, 2017
    Assignee: Applied Materials Israel Ltd.
    Inventors: Moshe Amzaleg, Ariel Shkalim, Efrat Rozenman
  • Patent number: 9613406
    Abstract: An approach is provided for reconstruction of dynamic arbitrary specular objects. The approach involves determining time-of-flight data for at least one pixel of at least one time-of-flight sensor configured with at least one retro-reflector, wherein the time-of-flight data includes a first distance from the at least one time-of-flight sensor to at least one point of at least one surface, and a second distance from the at least one point to the at least one retro-reflector. The approach also involves determining other time-of-flight data for one or more neighboring pixels which are neighboring the at least one pixel. The approach further involves determining at least one range distance to the at least one point of the at least one surface by causing, at least in part, a factoring out of the second distance from the time-of-flight data by using the other time-of-flight data. The approach also involves causing, at least in part, a reconstruction of the at least one surface using the at least one range distance.
    Type: Grant
    Filed: March 31, 2015
    Date of Patent: April 4, 2017
    Assignee: Nokia Technologies Oy
    Inventors: Xin Chen, Ruigang Yang
  • Patent number: 9600889
    Abstract: A method of performing depth estimation, the method comprising: estimating, at a higher spatial resolution of images, a global disparity map based on disparity between corresponding pixels of a stereo disparity map obtained from images forming a stereo image set associated with a passive depth estimation technique and an active depth map associated with an active depth estimation technique, wherein the disparity is estimated within a restricted range determined based on a disparity previously estimated between a stereo disparity map and an active depth map corresponding to a lower spatial resolution of the said images and map.
    Type: Grant
    Filed: December 20, 2014
    Date of Patent: March 21, 2017
    Assignee: Thomson Licensing
    Inventors: Guillaume Boisson, Neus Sabater, Valter Drazic
  • Patent number: 9595098
    Abstract: Example methods disclosed herein to perform inventory display auditing include comparing a reference image of an inventory display obtained during a prior audit of the inventory display with a current image captured by a camera to determine whether the current image corresponds to the same inventory display as the reference image. Disclosed example methods also include, when the current image is determined to correspond to the same inventory display as the reference image, identifying a difference region in the current image corresponding to a portion of the inventory display that changed relative to the prior audit. Disclosed example methods further include, while the current image is displayed on a display screen, modifying a characteristic of the current image in the difference region to indicate a change in product inventory in the portion of the inventory display.
    Type: Grant
    Filed: March 30, 2016
    Date of Patent: March 14, 2017
    Assignee: The Nielsen Company (US), LLC
    Inventor: Michael A. Hicks
  • Patent number: 9594940
    Abstract: A method and device for compressing a fingerprint template is provided. The fingerprint template is derived from fingerprint image data captured from a sensor. The compressing process initially surveys the fingerprint template to determine one or more starting points for encoding. Subsequently, the compression process selects a starting point from the one or more starting points and proceeds to encode that starting point. After encoding the starting point, the compression process encodes all contours extending from the selected starting point. The above process is repeated until all starting points and their associated contours have been encoded.
    Type: Grant
    Filed: August 11, 2015
    Date of Patent: March 14, 2017
    Assignee: Synaptics Incorporated
    Inventors: Yi-Hsun E. Cheng, Kuntal Sengupta, Anthony Russo
  • Patent number: 9594949
    Abstract: A method, computer readable medium and apparatus for verifying an identity of an individual based upon facial expressions as exhibited in a query video of the individual are disclosed. The method includes receiving a reference video for each one of a plurality of different individuals, wherein a plurality of facial gesture encoders is extracted from at least one frame of the reference video describing one or more facial expressions of each one of the plurality of different individuals, receiving the query video, calculating a similarity score for the reference video for the each one of the plurality of different individuals based on an analysis that compares the plurality of facial gesture encoders of the at least one frame of the reference video for the each one of the plurality of different individuals to a plurality of facial gesture encoders extracted from at least one frame of the query video.
    Type: Grant
    Filed: August 31, 2015
    Date of Patent: March 14, 2017
    Assignee: Xerox Corporation
    Inventors: Matthew Adam Shreve, Jayant Kumar, Qun Li, Edgar A. Bernal, Raja Bala
  • Patent number: 9588957
    Abstract: In object recognition using image information, countermeasures are taken for reducing erroneous judgment caused when the similarity between a visual feature of an object image and a visual feature of dictionary data of another object becomes high in association with a temporal change. In a dictionary update method for a computer for updating dictionary data in which a visual feature of an object which a user desires to recognize is registered: when there are a plurality of data pieces having a visual feature similar to the inputted inquiry image, visual features of a plurality of objects are concluded as being similar and hence the pair of objects are accumulated into a similar object accumulation section; and when the objects accumulated in the similar object accumulation section have reached a condition set forth in advance, data having the visual features of the objects having reached the condition is concluded as requiring update and hence update is recommended to a user having registered the object.
    Type: Grant
    Filed: July 19, 2013
    Date of Patent: March 7, 2017
    Assignee: Biglobe, Inc.
    Inventors: Michitaro Miyata, Kyoji Hirata, Takeshi Kawasaki, Kazuya Furukawa
  • Patent number: 9576179
    Abstract: An authentication apparatus includes a biometric data acquirer configured to acquire fingerprint data and an electrocardiogram (ECG) waveform of a user, and a humidity level acquirer configured to acquire a humidity level of skin of the user. The apparatus further includes a similarity extractor configured to adjust a first similarity between the fingerprint data and reference fingerprint data of a pre-registered user, and a second similarity between the ECG waveform and a reference ECG waveform of the pre-registered user, based on the humidity level, and extract a combined similarity based on the adjusted first similarity and the adjusted second similarity. The apparatus further includes an authenticator configured to authenticate whether the user is the pre-registered user based on the combined similarity.
    Type: Grant
    Filed: March 24, 2015
    Date of Patent: February 21, 2017
    Assignee: Samsung Electronics Co., Ltd.
    Inventors: Chisung Bae, Sangjoon Kim
  • Patent number: 9576196
    Abstract: A system to recognize text or symbols contained in a captured image using machine learning models leverages context information about the image to improve accuracy. Contextual information is determined for the entire image, or spatial regions of the images, and is provided to a machine learning model when a determination is made as to whether a region does or does not contain text or symbols. Associating features related to the larger context with features extracted from regions potentially containing text or symbolic content provides an incremental improvement of results obtained using machine learning techniques.
    Type: Grant
    Filed: August 20, 2014
    Date of Patent: February 21, 2017
    Assignee: AMAZON TECHNOLOGIES, INC.
    Inventor: Pradeep Natarajan
  • Patent number: 9575996
    Abstract: A emotion image recommendation system and method are disclosed, where a recommended emotion image is provided in replace with a selected emotion image as conventionally used by a user, thereby an emotion image corresponding to an emotional word used by the user may be provided, so that the efficacy of a reduced frequency for locating the appropriate emotion image for the user may be achieved.
    Type: Grant
    Filed: March 31, 2015
    Date of Patent: February 21, 2017
    Assignees: INVENTEC (PUDONG) TECHNOLOGY CORPORATION, INVENTEC CORPORATION
    Inventor: Ti Ku Yao
  • Patent number: 9574883
    Abstract: Systems and methods are provided for generating maps with semantic labels. A computing device can determine a first map that includes features located at first positions and semantic labels located at semantic positions, and determine a second map that includes at least some of the features located at second positions. The computing device can identify a first region with fixed features located at first positions and corresponding equivalent second positions. The computing device can identify a second region with moved features located at first positions and corresponding non-equivalent second positions. The computing device can determine one or more transformations between first positions and second positions. The computing device can assign the semantic labels to the second map at second semantic positions, where the second semantic positions are the same in the first region, and where the second semantic positions in the second region are based on the transformation(s).
    Type: Grant
    Filed: March 24, 2015
    Date of Patent: February 21, 2017
    Assignee: X Development LLC
    Inventors: Kevin William Watts, Julian Mac Neille Mason, Peter Elving Anderson-Sprecher
  • Patent number: 9569087
    Abstract: A method of operating an electronic device is provided. The method includes detecting a plurality of fingerprint drags after a touch on a portion of the electronic device, wherein each of the plurality of fingerprint drags is a consecutive drag from a previous drag and each of the plurality of fingerprint drags has a different orientation from each other; and identifying a fingerprint based on a plurality of images acquired from the plurality of fingerprint drags, wherein each of the plurality of images is compared with corresponding portion of a reference image.
    Type: Grant
    Filed: March 26, 2015
    Date of Patent: February 14, 2017
    Assignee: Samsung Electronics Co., Ltd
    Inventors: Jeong-Ho Cho, Jeong-Min Park, Do-Hyoung Chung, Ki-Hong Min, Dong-Han Lee
  • Patent number: 9569700
    Abstract: A system that identifies attributes of an item depicted in an image using artificial intelligence is provided. For example, the system may use one or more deep belief networks (DBNs) or convolution neural networks (CNNs) trained to analyze images and identify attributes in items depicted in the images. A first artificial intelligence module may analyze an image to determine a type of item depicted in the image. The system may then select a second artificial intelligence module that is associated with the type of item and use the second artificial intelligence module to identify attributes in the item depicted in the image. Identified attributes, if associated with a confidence level over a threshold value, may be provided to a user. The user may provide feedback on the accuracy of the identified attributes, which can be used to further train the first and/or second artificial intelligence modules.
    Type: Grant
    Filed: December 17, 2014
    Date of Patent: February 14, 2017
    Assignee: Amazon Technologies, Inc.
    Inventors: Anthony Alexander Santos, Adam James Finkelstein, David Akira Gingrich, David Michael Hurley, Siqi Zhao
  • Patent number: 9558389
    Abstract: Embodiments of a system and method for detecting a palm of a hand using an image are generally described herein. A method for detecting a palm of a hand may include determining a plurality of cumulative distance values from pixels in a set of image pixels to pixels in a plurality of finger template matches, wherein the set of image pixels includes a set of image pixels on detected edges in the image. The method may include selecting a set of finger template matches from the plurality of finger template matches, corresponding to a set of smallest cumulative distance values from the plurality of cumulative distance values. The method may include refining the set of finger template matches to create refined finger templates, bundling refined finger templates to create a bundle of refined finger templates, and verifying that the bundle of refined finger templates includes a valid palm detection.
    Type: Grant
    Filed: March 24, 2015
    Date of Patent: January 31, 2017
    Assignee: Intel Corporation
    Inventor: Michael Kounavis
  • Patent number: 9552535
    Abstract: Apparatus, methods, and articles of manufacture for obtaining examples that break a visual expression classifier at user devices such as tablets, smartphones, personal computers, and cameras. The examples are sent from the user devices to a server. The server may use the examples to update the classifier, and then distribute the updated classifier code and/or updated classifier parameters to the user devices. The users of the devices may be incentivized to provide the examples that break the classifier, for example, by monetary reward, access to updated versions of the classifier, public ranking or recognition of the user, a self-rewarding game. The examples may be evaluated using a pipeline of untrained crowdsourcing providers and trained experts. The examples may contain user images and/or depersonalized information extracted from the user images.
    Type: Grant
    Filed: February 11, 2014
    Date of Patent: January 24, 2017
    Assignee: Emotient, Inc.
    Inventors: Javier Movellan, Marian Stewart Bartlett, Ian Fasel, Gwen Ford Littlewort, Joshua Susskind, Jacob Whitehill
  • Patent number: 9536129
    Abstract: A sensor for detecting fingerprints is provided having first and second substrates, a two-dimensional array of sensing elements formed on the first substrate, and a plurality of thin-film transistors or TFTs for controlling the sensing elements at pixel locations along the array. Each of the sensing elements detects one of electrical signals (e.g., capacitance, resistance, or impedance), temperature, or light via one of the first or second substrates representative of one or more fingerprints. The top of the second substrate or the bottom of the first substrate may provide a platen upon which one or more fingers can be disposed. The sensor may be utilized in a fingerprint scanner having one or more processors driving sensing elements or reading from sensing elements analog signals representative of one or more fingerprints, and generating an image representative of the one or more fingerprints from the analog signals.
    Type: Grant
    Filed: January 21, 2016
    Date of Patent: January 3, 2017
    Assignee: CROSS MATCH TECHNOLOGIES, INC.
    Inventors: John F. Carver, Daniel H. Raguin
  • Patent number: 9536178
    Abstract: An object recognition system and method is provided which uses automated algorithmically determined negative training. Negative training with respect to a particular object classifier allows for more streamlined and efficient targeted negative training, enabling time and cost savings while simultaneously improving the accuracy of recognition based on the targeted negative training. According to certain additional aspects, embodiments of the invention relate to a system and method for structuring an object recognition engine to maximize recognition accuracy and emulate human visual cortex.
    Type: Grant
    Filed: August 19, 2014
    Date of Patent: January 3, 2017
    Assignee: VUFIND, INC.
    Inventor: Moataz Ali Rashad Mohamed
  • Patent number: 9530200
    Abstract: A method and a system for inspection of a patterned structure are provided. In various embodiments, the method for inspection of a patterned structure includes transferring the patterned structure into a microscope. The method further includes acquiring a top-view image of the patterned structure by the microscope. The method further includes transferring the patterned structure out of the microscope and exporting the top-view image to an image analysis processor. The method further includes measuring a difference between a contour of the top-view image and a predetermined layout of the patterned structure by the image analysis processor.
    Type: Grant
    Filed: June 20, 2014
    Date of Patent: December 27, 2016
    Assignee: TAIWAN SEMICONDUCTOR MANUFACTURING CO., LTD.
    Inventors: Wen-Chang Hsueh, Chia-Jen Chen, Hsin-Chang Lee
  • Patent number: 9531952
    Abstract: Aspects of this disclosure relate to generating a composite image of an image of and another image that has a wider field of view. After an image is selected, the visual features in the image may be identified. Several images, such as panoramas, which have wider fields of view than an image captured by a camera, may be selected according to a comparison of the visual features in the image and the visual features of the larger images. The image may be aligned with each of the larger images, and at least one of these smaller-larger image pairs may be generated as a composite image.
    Type: Grant
    Filed: March 27, 2015
    Date of Patent: December 27, 2016
    Assignee: Google Inc.
    Inventors: Keith Noah Snavely, Pierre Georgel, Steven Maxwell Seitz
  • Patent number: 9530212
    Abstract: An image frame conversion method and a video frame conversion method and apparatus are provided. The video frame conversion method includes: extracting a longest line segment of a 2D video frame; determining a first depth image of the 2D video frame according to a direction of the longest line segment; determining motion information of the 2D video frame according to a video frame adjacent to the 2D video frame; determining a second depth image of the 2D video frame according to the motion information; and generating, according to the first depth image, the second depth image, and the 2D video frame, a 3D video frame corresponding to the 2D video frame.
    Type: Grant
    Filed: December 17, 2014
    Date of Patent: December 27, 2016
    Assignee: Huawei Technologies Co., Ltd.
    Inventors: Mingchen Han, Cheolkon Jung, Lei Wang, Biao Hou