Statistical Decision Process Patents (Class 382/228)
  • Patent number: 11183174
    Abstract: A processor-implemented method of personalizing a speech recognition model includes: obtaining statistical information of first scaling vectors combined with a base model for speech recognition; obtaining utterance data of a user; and generating a personalized speech recognition model by modifying a second scaling vector combined with the base model based on the utterance data of the user and the statistical information.
    Type: Grant
    Filed: March 13, 2019
    Date of Patent: November 23, 2021
    Assignee: Samsung Electronics Co., Ltd.
    Inventor: Ki Soo Kwon
  • Patent number: 11144986
    Abstract: A computer-implemented method and system may include identifying, by a recommendation engine, a first plurality of images in a first theme that includes a first layout characteristic, determining a first plurality of vector representations of the first plurality of images using an image processing model that includes a layered matrix of nodes with weighted interconnections, identifying at least a second plurality of images in a second theme that includes a second layout characteristic that is different from the first layout characteristic, determining a second plurality of vector representations of the second plurality of images, computing a vector difference between the first plurality of vector representations and the second plurality of vector representations; determining and communicating a recommendation based on the computed vector difference, and mapping received feedback to the weightings of the interconnections to improve a future recommendation.
    Type: Grant
    Filed: July 31, 2019
    Date of Patent: October 12, 2021
    Assignee: Shopify Inc.
    Inventor: Putra Manggala
  • Patent number: 11024004
    Abstract: A system for enhancing a set of object scan images may be provided. The set of object scan images may also here comprise at least a first scan image, located at a distance therefrom a second scan image.
    Type: Grant
    Filed: August 31, 2018
    Date of Patent: June 1, 2021
    Assignee: International Business Machines Corporation
    Inventors: Maria H. Oleszkiewicz, Blazej R. Rutkowski, Wojciech Sobala, Piotr Padkowski
  • Patent number: 11010903
    Abstract: Techniques are described for processing digital video data using one or more machine learning models to determine an outcome of an item placement operation within a fulfillment center environment. Video data is processed using one or more machine learning models to determine an estimated likelihood that an occurrence of a particular fulfillment center operation is depicted within the two or more instances of digital video data. Upon determining that the estimated likelihood exceeds a predefined threshold confidence level, the video data is processed using second one or more machine learning models to determine a bin placement prediction and a confidence value. A data repository for a control system for the fulfillment center environment is updated, based on the bin placement prediction and the confidence value.
    Type: Grant
    Filed: January 30, 2019
    Date of Patent: May 18, 2021
    Assignee: Amazon Technologies, Inc.
    Inventors: Elisha Gallaudet, Timothy Stallman
  • Patent number: 10943103
    Abstract: An information processing apparatus is provided that detects a moving object from an image, and detects a human body from a target region that is an upper part of a region corresponding to the moving object, using a feature of an upper body of a human body, to thereby reduce processing load.
    Type: Grant
    Filed: December 19, 2017
    Date of Patent: March 9, 2021
    Assignee: CANON KABUSHIKI KAISHA
    Inventor: Hisashi Kawabayashi
  • Patent number: 10928831
    Abstract: Described herein are methods of generating learning data to facilitate de-biasing the labeled location of an object of interest within an image. Methods may include: receiving sensor data, where the sensor data is a first image; determining reference corner locations of an object in the first image using image processing; generating observed corner locations of the object in the first image from the determined reference corner locations; generating a bias transformation based, at least in part, on a difference between the reference corner locations and the observed corner locations of the object in the first image; receiving sensor data from another image sensor of a second image; receiving observed corner locations of an object in the second image from a user; and applying the bias transformation to the observed corner locations of the object in the second image to generate de-biased corners for the object in the second image.
    Type: Grant
    Filed: December 5, 2018
    Date of Patent: February 23, 2021
    Assignee: HERE GLOBAL B.V.
    Inventor: Anirudh Viswanathan
  • Patent number: 10867168
    Abstract: An information processing apparatus includes an obtaining unit that obtains image data which is data obtained by digitizing a document including plural areas in one document, a setting unit that sets a determination standard for each of the areas based on a degree of necessity of a confirming operation for a text or a picture included in the area of the image data, and a determination unit that determines a method of the confirming operation for each of the areas based on the determination standard.
    Type: Grant
    Filed: January 29, 2019
    Date of Patent: December 15, 2020
    Assignee: FUJI XEROX CO., LTD.
    Inventor: Shintaro Nishioka
  • Patent number: 10861147
    Abstract: A method for detecting structural defects includes generating a first probabilistic map of a critical location of a structural component based on a physics model of the critical location, capturing data of the critical location, generating a second probabilistic map based on an analysis of the captured data of the critical location, aligning the first probabilistic map and the second probabilistic map, combining the aligned first and second probabilistic maps, and determining if the structural component includes a structural defect based on the combined probabilistic maps.
    Type: Grant
    Filed: January 12, 2016
    Date of Patent: December 8, 2020
    Assignee: Sikorsky Aircraft Corporation
    Inventor: Hongcheng Wang
  • Patent number: 10817756
    Abstract: An information processing apparatus includes a determination, first, second and third receiving, and first and second checking units, and a controller. When a determination result matches a first determination result, the first checking unit outputs a matching determination result as the final output. The second checking unit checks the first determination result against a second. The first and second determination results are respectively received by the first and second receiving units. When the first determination result matches the second, the second checking unit outputs the matching determination result as the final output. When the input information is received, the controller operates the determination, first receiving, and first checking units. When it is determined that the determination result does not match the first determination result, the controller operates selectively one of a unit set and the third receiving unit. The unit set is constituted by the second receiving and checking units.
    Type: Grant
    Filed: October 12, 2018
    Date of Patent: October 27, 2020
    Assignee: FUJI XEROX CO., LTD.
    Inventors: Shunichi Kimura, Masanori Sekino, Ikken So, Satoshi Kubota, Takuya Sakurai, Kumi Fujiwara, Kunikazu Ueno, Yutaka Koshi
  • Patent number: 10803242
    Abstract: Embodiments provide a computer implemented method for identifying and correcting a misspelling in a question answering (QA) system, wherein the QA system is coupled to a document corpus, and the document corpus includes a plurality of documents related to a particular domain. The method includes the following steps: receiving an input question and a plurality of passages, wherein the plurality of passages are extracted from the document corpus by the QA system; providing at least one alternate form for each token extracted from the input question and the plurality of passages; identifying at least one misspelled token; and scoring at least one alternate form of each identified misspelled token.
    Type: Grant
    Filed: October 26, 2018
    Date of Patent: October 13, 2020
    Assignee: International Business Machines Corporation
    Inventors: Stephen A. Boxwell, Keith G. Frost, Stanley J. Vernier, Kyle M. Brake
  • Patent number: 10796452
    Abstract: In one embodiment, a system accesses a probability model associated with an image depicting a body. The probability model includes probability values associated with regions of the image and each probability value represents a probability of the associated region of the image containing a particular body part. The system selects a subset (e.g., 3) of the probability values based on a comparison of the probability values. For each selected probability value, the system identifies surrounding probability values surrounding the selected probability value and computes a probabilistic maximum based on the selected probability value and the surrounding probability values. Each probabilistic maximum is associated with a location within the regions associated with the selected probability value and the surrounding probability values.
    Type: Grant
    Filed: December 31, 2018
    Date of Patent: October 6, 2020
    Assignee: Facebook, Inc.
    Inventors: Peter Vajda, Peizhao Zhang, Matthieu Tony Uyttendaele, Yanghan Wang
  • Patent number: 10789410
    Abstract: To prevent typos in electronic documents, quality checks are performed to identify portions of the document that have been misspelled. Major sources of false positives are words that have been transliterated from other languages into English. The quality control system described herein identifies transliterated words in electronic documents. The system may be trained to evaluate a set of words, and determine likelihoods that individual words in the set of words are transliterated (i.e., they are not native to the language and/or alphabet of the electronic document). In some embodiments, the model may be further configured to identify a language and/or alphabet that the individual word was transliterated from.
    Type: Grant
    Filed: June 26, 2017
    Date of Patent: September 29, 2020
    Assignee: Amazon Technologies, Inc.
    Inventors: Sriraghavendra Ramaswamy, Gururaj Narayanan, Kalidas Yeturu, Nilesh Mahadeo Sah
  • Patent number: 10719115
    Abstract: Methods, systems, and apparatuses are described for isolated word training and detection. Isolated word training devices and systems are provided in which a user may provide a wake-up phrase from 1 to 3 times to train the device or system. A concatenated phoneme model of the user-provided wake-up phrase may be generated based on the provided wake-up phrase and a pre-trained phoneme model database. A word model of the wake-up phrase may be subsequently generated from the concatenated phoneme model and the provided wake-up phrase. Once trained, the user-provided wake-up phrase may be used to unlock the device or system and/or to wake up the device or system from a standby mode of operation. The word model of the user-provided wake-up phrase may be further adapted based on additional provisioning of the wake-up phrase.
    Type: Grant
    Filed: January 27, 2015
    Date of Patent: July 21, 2020
    Assignee: AVAGO TECHNOLOGIES INTERNATIONAL SALES PTE. LIMITED
    Inventors: Robert W. Zopf, Bengt J. Borgstrom
  • Patent number: 10699170
    Abstract: Disclosed is a method for generating a semantic image labeling model, comprising: forming a first CNN and a second CNN, respectively; randomly initializing the first CNN; inputting a raw image and predetermined label ground truth annotations to the first CNN to iteratively update weights thereof so that a category label probability for the image, which is output from the first CNN, approaches the predetermined label ground truth annotations; randomly initializing the second CNN; inputting the category label probability to the second CNN to correct the input category label probability so as to determine classification errors of the category label probabilities; updating the second CNN by back-propagating the classification errors; concatenating the updated first and second CNNs; classifying each pixel in the raw image into one of general object categories; and back-propagating classification errors through the concatenated CNN to update weights thereof until the classification errors less than a predetermined
    Type: Grant
    Filed: January 8, 2018
    Date of Patent: June 30, 2020
    Assignee: Beijing SenseTime Technology Development Co., Ltd.
    Inventors: Xiaoou Tang, Ziwei Liu, Xiaoxiao Li, Ping Luo, Chen Change Loy
  • Patent number: 10574874
    Abstract: An endoscope apparatus includes a processor comprising hardware, a processor performs a focus control that controls a position of a focus lens to an in-focus position based on an in-focus evaluation value, the focus lens being included in an optical system that forms an image of a captured image that is acquired by an imaging section, and the in-focus evaluation value being calculated from a first area within the captured image, and a change-in-scene detection process that detects whether or not a change in scene has occurred from a second area that includes an area that differs from the first area, wherein the processor is set to a standby state when the position of the focus lens has been controlled to the in-focus position, and resumes the focus control process when a change in scene has been detected when the focus control section is set to the standby state.
    Type: Grant
    Filed: April 14, 2016
    Date of Patent: February 25, 2020
    Assignee: OLYMPUS CORPORATION
    Inventor: Koichiro Yoshino
  • Patent number: 10565351
    Abstract: This disclosure describes a computerized system for managing medical information, the system including at least one computing device configured to receive a plurality of medical documents, determine skip grams comprising tokens for the plurality of medical documents, wherein each of the skip grams comprises one or more tokens of at least one of the medical documents, and populate a trie data structure based on the skip grams. The at least one computing device is further configured to prune the plurality of nodes of the trie based on one or more criteria to produce a pruned trie of nodes, determine rules for associating medical billing codes with the skip grams of the pruned trie of nodes based on pointwise mutual information, and output the determined rules.
    Type: Grant
    Filed: August 22, 2016
    Date of Patent: February 18, 2020
    Assignee: 3M Innovative Properties Company
    Inventors: Michael A. Nossal, Guoli Wang
  • Patent number: 10552968
    Abstract: Dense feature scale detection can be implemented using multiple convolutional neural networks trained on scale data to more accurately and efficiently match pixels between images. An input image can be used to generate multiple scaled images. The multiple scaled images are input into a feature net, which outputs feature data for the multiple scaled images. An attention net is used to generate an attention map from the input image. The attention map assigns emphasis as a soft distribution to different scales based on texture analysis. The feature data and the attention data can be combined through a multiplication process and then summed to generate dense features for comparison.
    Type: Grant
    Filed: September 22, 2017
    Date of Patent: February 4, 2020
    Assignee: Snap Inc.
    Inventors: Shenlong Wang, Linjie Luo, Ning Zhang, Jia Li
  • Patent number: 10540558
    Abstract: An information processing apparatus includes: determination units that make determinations on an input using different methods, and obtain determination results for the input; a first output unit that outputs, when a certain percentage or more of the determination results match, a determination result matched at the certain percentage or more; a second output unit that outputs, when the first output unit does not find a determination result matched at the certain percentage or more, a final determination result for the input; and an accuracy rate calculation unit that calculates, when a determination result obtained by a determination unit of interest among the determination units corresponds to a determination result matched at the certain percentage or more or matches the determination result output by the second output unit, an accuracy rate of the determination unit of interest, regarding that the determination result obtained by the determination unit of interest is correct.
    Type: Grant
    Filed: August 22, 2018
    Date of Patent: January 21, 2020
    Assignee: FUJI XEROX CO., LTD.
    Inventor: Shunichi Kimura
  • Patent number: 10497121
    Abstract: A method includes determining whether a region to be recognized corresponding to a specified feature exists in an image to be processed. The method may also include determining a main subject region containing the main subject of the image to be processed in accordance with a preset feature parameter of the main subject region and the coordinates and size of the region to be recognized when the region to be recognized exists. The method may further include extracting an image of the main subject region as a foreground image for an extraction process of a foreground target, and taking the extracted image as the main subject of the image to be processed.
    Type: Grant
    Filed: April 27, 2017
    Date of Patent: December 3, 2019
    Assignee: ALIBABA GROUP HOLDING LIMITED
    Inventors: Shan An, Hongming Zhang, Bin Liu, Guocheng Chen
  • Patent number: 10459561
    Abstract: Aspects of the disclosure are related to a method for determining a touch pressure level on a touchscreen, comprising: detecting a touch event by the touchscreen; obtaining data relating to features associated with the touch event comprising a capacitance value, a touch area, and/or a touch duration; and determining a touch pressure level based on one or more of the features.
    Type: Grant
    Filed: July 9, 2015
    Date of Patent: October 29, 2019
    Assignee: QUALCOMM Incorporated
    Inventors: Marc Bosch Ruiz, Galit Hofree, Robert Scott Tartz, Robyn Teresa Oliver, Jonathan Karl Kies, Virginia Walker Keating
  • Patent number: 10445910
    Abstract: A generating apparatus according to an embodiment of the present application includes an obtaining unit and a generating unit. The obtaining unit obtains intermediate images in an intermediate layer of a neural network that recognizes the quantity of objects contained in an image. The generating unit generates a combination image by combining together the intermediate images obtained by the obtaining unit. For example, the generating unit generates the combination image by combining together the intermediate images to which a weight has been applied on the basis of a change in the recognition of the quantity of objects corresponding to a process applied to each of the intermediate images.
    Type: Grant
    Filed: March 10, 2017
    Date of Patent: October 15, 2019
    Assignee: YAHOO JAPAN CORPORATION
    Inventors: Tomohiro Tanaka, Naoaki Yamashita
  • Patent number: 10368104
    Abstract: Systems and methods for transmission and display of synchronized physical and visual data for three-dimensional display are disclosed. Physical and visual data may be encoded and interlaced to enable synchronized transmission of the data in efficient manners. A single data transport stream may be utilized to transmit both physical and visual data over a communication channel, allowing physical and visual data to be efficiently co-joined on a same surface at a receiving end to provide a realistic, true three-dimensional representation.
    Type: Grant
    Filed: April 1, 2015
    Date of Patent: July 30, 2019
    Assignee: Rockwell Collins, Inc.
    Inventor: James P. Mitchell
  • Patent number: 10322801
    Abstract: The delivery of a package using an unmanned aerial vehicle (UAV) may include performing a surveillance action at a property of an authorized party. The surveillance action may include imaging the property, and may be performed before or after the package is delivered. Surveillance data such as image data gathered in a surveillance action may be modified to remove or obscure adjacent properties not associated with the property of the authorized party. Based on surveillance data, the probability or confidence value of a surveillance event may be determined and a surveillance alert may be generated. Surveillance actions may be scheduled by a central controller to ensure that the surveillance action may be performed without exhausting the resources, such as battery power or time, of the UAV. Surveillance actions may be received as an interrupt and may have an associated priority level.
    Type: Grant
    Filed: June 12, 2015
    Date of Patent: June 18, 2019
    Assignee: Amazon Technologies, Inc.
    Inventor: Kalidas Yeturu
  • Patent number: 10127220
    Abstract: Systems and processes for language identification from short strings are provided. In accordance with one example, a method includes, at a first electronic device with one or more processors and memory, receiving user input including an n-gram and determining a similarity between a representation of the n-gram and a representation of a first language. The representation of the first language is based on an occurrence of each of a plurality of n-grams in the first language and an occurrence of each of the plurality of n-grams in a second language. The method further includes determining whether the similarity between the representation of the n-gram and the representation of the first language satisfies a threshold.
    Type: Grant
    Filed: September 3, 2015
    Date of Patent: November 13, 2018
    Assignee: Apple Inc.
    Inventors: Jerome R. Bellegarda, Bishal Barman
  • Patent number: 10122973
    Abstract: The present disclosure relates to a fluoroscopic inspection system for automatic classification and recognition of cargoes. The system includes: an image data acquiring unit, configured to perform scanning and imaging for a container by using an X-ray scanning device to acquire a scanned image; an image segmenting unit, configured to segment the scanned image into small regions each having similar gray scales and texture features; a feature extracting unit, configured to extract features of the small regions; a training unit, configured to generate a classifier according to annotated images; and a classification and recognition unit, configured to recognize the small regions by using the classifier according to the extracted features, to obtain a probability of each small region pertaining to a certain category of cargoes, and merge small regions to obtain large regions each representing a category.
    Type: Grant
    Filed: December 23, 2014
    Date of Patent: November 6, 2018
    Assignee: NUCTECH COMPANY LIMITED
    Inventors: Ziran Zhao, Yaohong Liu, Jianping Gu, Qiang Li, Jian Zhang
  • Patent number: 10096127
    Abstract: It is inter alia disclosed to classify samples of a set of samples of an image regarding whether or not a respective sample comprises optical information on a specific object type based on classification information. Therein, the classification information comprises a plurality of classifiers, each classifier being associated exclusively with one color channel of a plurality of color channels.
    Type: Grant
    Filed: October 22, 2012
    Date of Patent: October 9, 2018
    Assignee: Nokia Technologies Oy
    Inventors: Jiangwei Li, Lei Xu, Yikai Fang, Kongqiao Wang
  • Patent number: 10043254
    Abstract: In an example embodiment, an image transformation is automatically performed on a digital image to improve perceived professionalism of a subject of the image. A machine learning algorithm is utilized to generate a professionalism score for the digital image, the utilizing a machine learning algorithm comprising: a training mode where a plurality of sample images with labeled professionalism scores are used to train a classification function in a model that produces as professionalism score as output; an analysis mode where the model is used to generate a professionalism score for the digital image. Then the professionalism score is used as an input to a continuous variable optimization algorithm to determine an optimum version of the digital image from a plurality of possible versions of the digital image on which one or more image transformations have been performed, using the classification function.
    Type: Grant
    Filed: April 14, 2016
    Date of Patent: August 7, 2018
    Assignee: Microsoft Technology Licensing, LLC
    Inventors: Uri Merhav, Dan Shacham
  • Patent number: 10026015
    Abstract: A method includes acquiring image data from for a given time step. The image data represents at least one of an instrument and a target within an image space that includes patient anatomy. A given belief of system state is estimated to include an estimate of instrument state and target state determined from corresponding measurement and motion models computed based on the acquired image data and control input parameters for the given time step. An expected belief of system state is estimated for a subsequent time step based on the estimated given belief of system state and based on the measurement models and the motion models computed over a plurality of different control inputs parameters. An information metric is computed for each of the expected beliefs of system state and control input parameters are selected for the subsequent time step based on the computed information metrics.
    Type: Grant
    Filed: April 1, 2015
    Date of Patent: July 17, 2018
    Assignees: Case Western Reserve University, The Cleveland Clinic Foundation
    Inventors: Murat Cenk Cavusoglu, Mark Renfrew, Zhuofu Bai, Taoming Liu, Nathaniel Lombard Poirot, Jeffrey Ustin
  • Patent number: 9930250
    Abstract: One image processing method has at least the following steps: receiving an image input in a device, wherein the image input is composed of at least one source image; receiving image selection information; regarding a source image included in the image input, checking the image selection information to determine whether the source image is selected or skipped; and performing an object oriented image processing operation upon each selected source image. Another image processing method has at least the following steps: receiving an image input in a device, wherein the image input is composed of at least one source image; receiving algorithm selection information; and regarding a source image included in the image input, checking the algorithm selection information to determine a selected image processing algorithm from a plurality of different image processing algorithms, and performing an image processing operation upon the source image based on the selected image processing algorithm.
    Type: Grant
    Filed: September 15, 2013
    Date of Patent: March 27, 2018
    Assignee: MEDIATEK INC.
    Inventors: Cheng-Tsai Ho, Ding-Yun Chen, Chi-Cheng Ju
  • Patent number: 9832540
    Abstract: A method and apparatus for encoding a source media program is disclosed. In one embodiment, the method comprises the steps of transcoding the source media program to produce a constant video quality transcoded version of the media program, determining if a portion of constant video quality version of the transcoded media program fails to satisfy a constraint, if the portion of the transcoded media program fails to satisfy the constraint, adjusting at least one transcoding parameter of a set of transcoding parameters, transcoding the portion of the source media program according to the adjusted at least one transcoding parameter to produce a constant bit rate version of the portion of the transcoded media program, and substituting the constant bit rate version of the portion of the transcoded media program for the constant video quality portion of the transcoded media program.
    Type: Grant
    Filed: December 15, 2010
    Date of Patent: November 28, 2017
    Assignee: HULU, LLC
    Inventors: Yupeng Liao, Kang Kang, Zhibing Wang
  • Patent number: 9805025
    Abstract: Embodiments relate to a system and a method for identifying, from contractual documents, (i) standard exact clauses matching clause examples and (ii) non-standard clauses semantically related to but not matching the clause examples. A standard feature data set comprising standard exact clauses matching clause examples is obtained. In addition, a mirror feature data set comprising semantically related clauses of the clause examples is obtained using semantic language analysis, where the mirror feature data set encompasses the standard feature data set. Non-standard clauses are obtained by extracting a difference between the mirror feature data set and the standard exact feature data set.
    Type: Grant
    Filed: July 13, 2015
    Date of Patent: October 31, 2017
    Assignee: Seal Software Limited
    Inventor: Kevin Gidney
  • Patent number: 9799110
    Abstract: An abnormality detection method of detecting abnormality of a blast furnace from tuyere images shot by cameras installed in vicinities of a plurality of tuyeres of the blast furnace includes: collecting, in a time-series manner, representative brightness vectors defined by representative brightnesses determined based on brightness values of respective pixels for each of the tuyeres image previously shot by the cameras at a same time; extracting a principal component vector by performing principal component analysis on the representative brightness vectors collected in the time-series manner; calculating, as an evaluation value, a length of a normal line drawn in a direction of the principal component vector from the representative brightness vector collected from the tuyere images shot by the cameras at the same time during an operation; and detecting the abnormality of the blast furnace by comparing the evaluation value with a predetermined threshold.
    Type: Grant
    Filed: June 18, 2014
    Date of Patent: October 24, 2017
    Assignee: JFE STEEL CORPORATION
    Inventors: Kazuro Tsuda, Takehide Hirata, Naoshi Yamahira, Hidekazu Abe
  • Patent number: 9792359
    Abstract: A method and system of providing training information for training a categorizer includes receiving a query relating to at least one category and identifying at least one case within a data set that matches the query. The method and system receives one of a first indication that the identified at least one case belongs to the category, and a second indication that the identified at least one case does not belong to the category. Training information is modified based on receiving one of the first indication and second indication.
    Type: Grant
    Filed: April 29, 2005
    Date of Patent: October 17, 2017
    Assignee: ENTIT SOFTWARE LLC
    Inventors: Evan R. Kirshenbaum, Henri J. Suermondt, George H. Forman, Stephane Chiocchetti
  • Patent number: 9773185
    Abstract: An image processing apparatus includes a corresponding region connecting unit that connects regions that depict the same target between a series of images captured in time series, thereby sets at least one connected region; a connected region feature data calculation unit that calculates feature data of the connected region; a digest index value calculation unit that calculates a digest index value corresponding to a degree at which the target depicted in the series of images is aggregated in each image of the series of images, based on the feature data; and a digest image detector that detects a digest image based on the digest index value.
    Type: Grant
    Filed: July 18, 2012
    Date of Patent: September 26, 2017
    Assignee: OLYMPUS CORPORATION
    Inventor: Yamato Kanda
  • Patent number: 9773159
    Abstract: At least one example embodiment discloses an image feature extracting method. The method includes determining a probabilistic model based on pixel values of pixels in a kernel, determining image feature information of a current pixel of the pixels in the kernel and determining whether to change the image feature information of the current pixel based on a random value and a probability value of the current pixel, the probability value being based on the probabilistic model.
    Type: Grant
    Filed: July 29, 2015
    Date of Patent: September 26, 2017
    Assignee: SAMSUNG ELECTRONICS CO., LTD.
    Inventors: Changkyo Lee, Byungin Yoo, Namjoon Kim, Jaejoon Han
  • Patent number: 9727800
    Abstract: Disclosed are a system, apparatus, and method for detecting objects. Input image frames may be received and distinct regions created within each frame. Descriptors may be extracted from the regions according to their associated probability. Extracted descriptors may be matched to reference descriptors. Votes or confidence is cast for particular regions according to region properties. The region properties may be determined from center voting methods based on vector intersection to other vectors or intersections with a region. The probability of selecting particular regions can increase with each vote or increase in confidence for a region. In response to updating probabilities, additional regions may be selected and additional descriptors may be extracted. Additional voting iterations can update the probability of selecting a next region. An object pose may be estimated in response to meeting one or more thresholds.
    Type: Grant
    Filed: September 25, 2015
    Date of Patent: August 8, 2017
    Assignee: QUALCOMM Incorporated
    Inventors: Qi Pan, Romain Tallonneau, Emilio Maggio, Stefan Walk
  • Patent number: 9710457
    Abstract: A computer-implemented apparatus and method for performing patent portfolio analysis. The patent portfolio analysis apparatus and method clusters a group of patents based upon one or more techniques. The clustering techniques include linguistic clustering techniques (e.g., eigenvector analysis), claim meaning, and patent classification techniques. Different aspects of the clusters are analyzed, including financial, claim breadth, and assignee patent comparisons. Moreover, patents and/or their clusters are linked to the Internet in order to determine what products might be covered by the claims of the patents or whether materials on the Internet might render patent claims invalid.
    Type: Grant
    Filed: January 9, 2012
    Date of Patent: July 18, 2017
    Inventors: Gregory A. Stobbs, John V. Biernacki
  • Patent number: 9699447
    Abstract: A system and method for calibrating a dynamic digital imaging system for the detection of defects in a moving product stream. The system has an elevated platform above a conveying unit for receiving a reference color tile. The elevated platform allows for the passage of products to be inspected on the below conveying unit surface such that calibration and re-calibration processes during image capturing may be accomplished on a continuous basis without interruption of the product stream.
    Type: Grant
    Filed: March 14, 2013
    Date of Patent: July 4, 2017
    Inventors: Rick Wendell Bajema, Wilfred Marcellien Bourg, Jr., Scott Fagan, Sonchai Lange, Kerwin Bradley, David Ray Warren
  • Patent number: 9639930
    Abstract: Provided are an apparatus for processing a medical image, and a method of processing a medical image by using the apparatus. The apparatus includes: an image obtainer that receives a first image obtained by photographing an object; and an image processor that calculates a scaling factor for adjusting a magnification of the first image based on statistic information corresponding to the object and generates a second image to be displayed by normalizing a size of the object by applying the calculated scaling factor to the first image.
    Type: Grant
    Filed: September 22, 2014
    Date of Patent: May 2, 2017
    Assignee: SAMSUNG MEDISON CO., LTD.
    Inventor: Jun-sang Yoo
  • Patent number: 9607228
    Abstract: The method and apparatus segment parts from a main target image, represent each part as a vertex in a spanning tree, use a detector to generate a confidence map of a location of each part in a succeeding video frame and apply scale change to detector sliding windows centered about each pixel in the part location image. In the succeeding video frame, the target location is sampled and a tracking probability is generated for each part bounding box, with the tracking probability having the maximum value being selected as the location of the target in the succeeding video frame.
    Type: Grant
    Filed: February 14, 2014
    Date of Patent: March 28, 2017
    Assignee: Toyota Motor Engineering & Manufacturing North America, Inc.
    Inventors: Xue Mei, Danil Prokhorov
  • Patent number: 9571844
    Abstract: An analyzing unit sets a threshold on the basis of a target compression ratio of a frame. The compression unit compresses a block having an attribute value larger than or equal to a threshold by applying a first compression format and a block having an attribute value smaller than the threshold by applying a second compression format having a higher compression ratio than the first compression format.
    Type: Grant
    Filed: December 18, 2014
    Date of Patent: February 14, 2017
    Assignee: MegaChips Corporation
    Inventor: Handa Chen
  • Patent number: 9530070
    Abstract: A method for parsing the text of a complex graphical image comprises obtaining a series of blocks of text from a complex graphical image. Those blocks of text are used to generate location scores, size scores and length scores. Each of the scores is weighted and linearly summed. The highest resulting sum is identified as the most likely to be a desired text block.
    Type: Grant
    Filed: April 29, 2015
    Date of Patent: December 27, 2016
    Assignee: Procore Technologies, Inc.
    Inventors: Andrew Lee Maltun, Michael Anthony Stock, II, Jake Sanders
  • Patent number: 9405976
    Abstract: The present disclosure relates to systems and methods for classifying videos based on video content. For a given video file including a plurality of frames, a subset of frames is extracted for processing. Frames that are too dark, blurry, or otherwise poor classification candidates are discarded from the subset. Generally, material classification scores that describe type of material content likely included in each frame are calculated for the remaining frames in the subset. The material classification scores are used to generate material arrangement vectors that represent the spatial arrangement of material content in each frame. The material arrangement vectors are subsequently classified to generate a scene classification score vector for each frame. The scene classification results are averaged (or otherwise processed) across all frames in the subset to associate the video file with one or more predefined scene categories related to overall types of scene content of the video file.
    Type: Grant
    Filed: August 18, 2015
    Date of Patent: August 2, 2016
    Assignee: TiVo Inc.
    Inventors: Heather Dunlop, Matthew Berry
  • Patent number: 9378434
    Abstract: An image analysis method includes acquiring fluorescent images of frames in time-series. Each fluorescent image comprises pixels in which pixel data are acquired in the time-series. The method further includes setting analysis areas to the acquired fluorescent images, calculating a classification value of the analysis areas, classifying the images into one or more groups on the basis of the classification value, calculating an average image of the analysis area every group, subtracting the average image from each image of the analysis area every group to calculate a new image of the analysis area, and calculating a correlation value on the basis of the new images every group.
    Type: Grant
    Filed: April 25, 2013
    Date of Patent: June 28, 2016
    Assignee: OLYMPUS CORPORATION
    Inventor: Akemi Suzuki
  • Patent number: 9330338
    Abstract: An image analysis method includes acquiring fluorescent images of frames in time-series. Each fluorescent image comprises pixels in which pixel data are acquired in the time-series. The method further includes setting analysis areas to the fluorescent images, selecting the fluorescent images of two or more frames to be used in analysis, extracting data pairs each comprising two pixels in which acquisition time intervals are the same in the analysis area of each of the selected fluorescent images, and performing product sum calculation of each of the data pairs for all of the selected images to calculate a correlation value.
    Type: Grant
    Filed: April 25, 2013
    Date of Patent: May 3, 2016
    Assignee: OLYMPUS CORPORATION
    Inventor: Akemi Suzuki
  • Patent number: 9286527
    Abstract: Techniques are provided for segmenting an input by cut point classification and training a cut classifier. A method may include receiving, by a computerized text recognition system, an input in a script. A heuristic may be applied to the input to insert multiple cut points. For each of the cut points, a probability may be generated and the probability may indicate a likelihood that the cut point is correct. Multiple segments of the input may be selected, and the segments may be defined by cut points having a probability over a threshold. Next, the segments of the input may be provided to a character recognizer. Additionally, a method may include training a cut classifier using a machine learning technique, based on multiple text training examples, to determine the correctness of a cut point in an input.
    Type: Grant
    Filed: February 20, 2014
    Date of Patent: March 15, 2016
    Assignee: Google Inc.
    Inventors: Li-Lun Wang, Thomas Deselaers, Henry Allan Rowley
  • Patent number: 9213910
    Abstract: Methods and systems for achieving accurate segmentation of characters with respect to a license plate image utilizing a reinforcement learning approach. A vehicle image can be captured by an image capturing unit and processed utilizing an ALPR (Automatic License Plate Recognition) unit. The reinforcement learning (RL) approach can be configured to initialize a segmentation agent with a starting location. A proper segmentation path (cuts) from top to bottom and from a darker to lighter area in a cropped license plate image can be identified by the segmentation agent during a training phase. Rewards can be provided based on a number of good and bad moves. The association between a current state and a sensory input with a preferred action can be learned by the segmentation agent at the end of the training phase.
    Type: Grant
    Filed: January 21, 2014
    Date of Patent: December 15, 2015
    Assignee: Xerox Corporation
    Inventors: Aaron Michael Burry, Farnaz Abtahi
  • Patent number: 9158813
    Abstract: A structured query may specify attribute values for attributes. An estimate of the number of items that will match the structured query if it is applied to a structured database is determined. If the estimated number of items is below a threshold, the structured query may be relaxed to form new candidate structured queries. The number of candidate queries may be determined based on a desired running time. Each of the candidate structured queries may be determined by changing one or more attribute values of the attributes of the structured query. Estimates of the number of items each of the candidate structured queries will match is determined, and the candidate structured query that has the highest matching estimation is used to query the database. The matching results may be output.
    Type: Grant
    Filed: June 9, 2010
    Date of Patent: October 13, 2015
    Assignee: MICROSOFT TECHNOLOGY LICENSING, LLC
    Inventors: Alexandros Ntoulas, Sreenivas Gollapudi, Samuel Ieong, Stelios Paparizos, John Christopher Shafer
  • Patent number: 9111169
    Abstract: A method and system for identifying one or more features represented in a plurality of sensor acquired data sets is described. The method and apparatus is particularly useful in automatic license plate recognition applications, where the sensor acquired data sets are data obtained from one or more digital cameras. This is achieved by determining a first probability of the identity of the one or more features (e.g., alphanumeric characters) from a first one of the data sets; determining a second probability of the identity of the one or more features from a second one of the data sets; and, using data fusion techniques, fusing the determined first and second probabilities to provide a fused probability. This fused probability is used to identify the one or more features from data sets.
    Type: Grant
    Filed: September 5, 2012
    Date of Patent: August 18, 2015
    Assignee: Sensen Networks Pty Ltd
    Inventor: Subhash Challa
  • Patent number: 9036923
    Abstract: Provided are an age estimation apparatus, an age estimation method, and an age estimation program capable of obtaining a recognition result closely matching the result perceived by human. An age estimation apparatus 10 for estimating an age of a person on image data includes a dimension compressor 11 for applying dimension compression to the image data to output low dimensional data; and an identification device 12 for estimating an age of a person on the basis of a learning result using a feature amount contained in the low dimensional data, wherein a parameter used for the dimension compression by the dimension compressor 11 and the feature amount used for age estimation by the identification device 12 are set on the basis of a result of an evaluation of a generalization capability using a weighting function that shows a degree of seriousness of an age estimation error for every age, and learning of the identification device 12 is performed on the basis of the weighting function.
    Type: Grant
    Filed: April 14, 2010
    Date of Patent: May 19, 2015
    Assignees: NEC Solution Innovators, Ltd., TOKYO INSTITUTE OF TECHNOLOGY
    Inventors: Kazuya Ueki, Masashi Sugiyama, Yasuyuki Ihara