Trainable Classifiers Or Pattern Recognizers (e.g., Adaline, Perceptron) Patents (Class 382/159)
  • Patent number: 9824462
    Abstract: A method for detecting an object includes inputting information of a moving object included in a plurality of images and generating a regression tree. In response to input of a new image, the system communicates information of a moving object included in the newly inputted image into the regression tree, and determines a size of a person included in the new image based on a resultant value of the regression tree.
    Type: Grant
    Filed: September 17, 2015
    Date of Patent: November 21, 2017
    Assignee: Samsung Electronics Co., Ltd.
    Inventors: Teshome Mikiyas, Ho-yub Jung
  • Patent number: 9818195
    Abstract: A method for use in estimating a pose of an imaged object comprises identifying candidate elements of an atlas that correspond to pixels in an image of the object, forming pairs of candidate elements, and comparing the distance between the members of each pair and with the distance between the corresponding pixels.
    Type: Grant
    Filed: March 18, 2016
    Date of Patent: November 14, 2017
    Assignee: Kabushiki Kaisha Toshiba
    Inventor: Christopher Zach
  • Patent number: 9811905
    Abstract: A method comprising using at least one hardware processor for computing a patch distinctiveness score for each of multiple patches of a medical image, computing a shape distinctiveness score for each of multiple regions of the medical image, and computing a saliency map of the medical image, by combining the patch distinctiveness score and the shape distinctiveness score.
    Type: Grant
    Filed: February 8, 2017
    Date of Patent: November 7, 2017
    Assignee: International Business Machines Corporation
    Inventors: Sharon Alpert, Pavel Kisilev
  • Patent number: 9813367
    Abstract: Methods and systems for analysis and/or classification of electronic message information so as to capture and identify salient objects exchanged during electronic message passing in order to impute certain information about the object, groups of objects, the message, groups of messages, the parties, communities involved in the message exchange or combinations, thereof.
    Type: Grant
    Filed: July 15, 2011
    Date of Patent: November 7, 2017
    Assignee: Firstwave Technology Pty Ltd
    Inventor: Simon David Ryan
  • Patent number: 9811781
    Abstract: A time-series data prediction device includes an acquisition unit, a prediction model generation unit, and a prediction unit. The acquisition unit acquires a plurality of observation values that continue at predetermined time intervals, as a prediction data, from time-series data of an observation value of a predetermined observation target, and acquires a training data. The prediction model generation unit generates a prediction model to calculate time-series data, which is an observation value predicted based on given time-series data, using the training data. The prediction unit calculates a predicted value of an observation value using the generated prediction model and the prediction data.
    Type: Grant
    Filed: September 29, 2014
    Date of Patent: November 7, 2017
    Assignee: HONDA MOTOR CO., LTD.
    Inventors: Tokitomo Ariyoshi, Takahiro Fukushige
  • Patent number: 9811778
    Abstract: A method, system, and computer program product for social collaboration in probabilistic prediction are provided in the illustrative embodiments. A set of predictions is sent to a user device. A prediction in the set of predictions is a probability of an outcome of an event. The probability is computed using a prediction model trained with training data corresponding to the event. An input is received from the user device. The input comprises a new prediction made at the user device using a new prediction model executing on the user device. A difference is determined between the prediction and the new prediction. The prediction model is revised to produce a revised prediction. A revised difference between the revised prediction and the new prediction is smaller than the difference.
    Type: Grant
    Filed: October 25, 2013
    Date of Patent: November 7, 2017
    Assignee: INTERNATIONAL BUSINESS MACHINES CORPORATION
    Inventors: Aaron Keith Baughman, Brian Marshall O'Connell
  • Patent number: 9805251
    Abstract: Systems and methods of automatic detection of a facial feature are disclosed. Moreover, methods and systems of yaw estimation of a human head based on a geometrical model are also disclosed.
    Type: Grant
    Filed: December 9, 2016
    Date of Patent: October 31, 2017
    Assignee: AMRITA VISHWA VIDYAPEETHAM
    Inventors: Athinarayanan Sankaranarayanan, Kamal Bijlani
  • Patent number: 9798948
    Abstract: Systems and methods for processing images to recognize characters. Such may include locating background in an image which separate lines of characters, and then locating background which separate characters. The inter-character spaces within a line may be used to determine a probable inter-character spacing for the characters. Within each detected line, the character having an inter-character spacing most similar to the probable inter-character spacing may be set as a starting character for classification. Using the probable inter-character spacing and the location of the starting character, the location of a character adjacent to the starting character in a first direction may be determined and the character classified. Such process may repeat in the first direction until a first end of the line is reached. The process may then move to a character adjacent the starting character in a second direction and repeat until a second end of the line is reached.
    Type: Grant
    Filed: July 31, 2015
    Date of Patent: October 24, 2017
    Assignee: DATALOGIC IP TECH, S.R.L.
    Inventors: Francesco Deppieri, Andrea Gambini
  • Patent number: 9798956
    Abstract: A method for recognizing a target object in an image, and an apparatus, where the method includes extracting feature data from an image, and transforming the extracted feature data into a uniform expression, performing automatic clustering for features in the image according to the feature data in the uniform expression and a historical clustering result, grouping a target object of a known class included in an automatic clustering result into the corresponding known class, in order to recognize a target object of the known class in the image, and training a classifier in a machine learning manner, for a target object of an unknown class included in the automatic clustering result, in order to recognize a target object of an unknown class in the image. In the embodiments of the present invention, recognizing a target object of an unknown class can be implemented.
    Type: Grant
    Filed: June 27, 2016
    Date of Patent: October 24, 2017
    Assignee: HUAWEI TECHNOLOGIES CO., LTD.
    Inventors: Zhuolin Jiang, Yangqiu Song, Qiang Yang
  • Patent number: 9792529
    Abstract: A sensor data processing system and method is described. Contemplated systems and methods derive a first recognition trait of an object from a first data set that represents the object in a first environmental state. A second recognition trait of the object is then derived from a second data set that represents the object in a second environmental state. The sensor data processing systems and methods then identifies a mapping of elements of the first and second recognition traits in a new representation space. The mapping of elements satisfies a variance criterion for corresponding elements, which allows the mapping to be used for object recognition. The sensor data processing systems and methods described herein provide new object recognition techniques that are computationally efficient and can be performed in real-time by the mobile phone technology that is currently available.
    Type: Grant
    Filed: August 24, 2016
    Date of Patent: October 17, 2017
    Assignee: Nant Holdings IP, LLC
    Inventors: Kamil Wnuk, Jeremi Sudol, Bing Song, Matheen Siddiqui, David McKinnon
  • Patent number: 9786030
    Abstract: Implementations generally relate to providing focal length adjustments in photos. In some implementations, a method includes determining a face model of a face in a photo. The method further includes simulating a change in the focal length based on the face model.
    Type: Grant
    Filed: June 16, 2014
    Date of Patent: October 10, 2017
    Assignee: Google Inc.
    Inventors: Thomas Binder, Martin Christopher Böhme
  • Patent number: 9779329
    Abstract: It aims to accurately identify a class concerning classification of an object for each area of an image. A plurality of subsets are created from a plurality of learning images, and an area discriminator for performing area discrimination of the learning images is learned for each subset created. Then, area discrimination of a plurality of learning evaluation images is performed by using the plurality of learned area discriminators. Further, an integrated identifier is learned based on an area discrimination result performed to the plurality of learning evaluation images and scene information previously associated with each of the plurality of learning evaluation images.
    Type: Grant
    Filed: November 13, 2015
    Date of Patent: October 3, 2017
    Assignee: CANON KABUSHIKI KAISHA
    Inventor: Takayuki Saruta
  • Patent number: 9774780
    Abstract: Approaches are described which enable a user of a computing device (e.g., mobile phone, tablet computer) to capture a first image using the front-facing camera and then use that first image to guide the capturing of a second image (e.g., a higher quality image) using the rear-facing camera of the device. When the user turns the device around, the first image is used as a model template to guide the user in moving and/or rotating the device to align the rear-facing camera for capturing the second image.
    Type: Grant
    Filed: August 10, 2015
    Date of Patent: September 26, 2017
    Assignee: AMAZON TECHNOLOGIES, INC.
    Inventors: Leo Benedict Baldwin, Philip Scott McClendon
  • Patent number: 9767496
    Abstract: A method and system is provided for notifying an addressee about a mail piece having an address. The system captures an image of the mail piece and determines a communication channel for the addressee using the address from the mail piece. The system provides the addressee with the image via the communication channel.
    Type: Grant
    Filed: March 21, 2012
    Date of Patent: September 19, 2017
    Assignee: UNITED STATES POSTAL SERVICE
    Inventor: Michael J. Amato
  • Patent number: 9760763
    Abstract: Disclosed are a system and a method for detecting feature points of a face. The system for detecting feature points of a face include: a training unit configured to divide each of a plurality of facial images into segments of a predetermined size, and generate an eigen-codebook representing a correlation between a feature vector of each of the divided segments and location of a predetermined feature point in the facial image; and a detection unit configured to detect a facial area from a target image, divide the facial area into segments of a predetermined size, detect a feature vector of each of the divided segments of the facial area, and detect a location of a feature point in the facial area by using the detected feature vector and the eigen-codebook.
    Type: Grant
    Filed: January 6, 2016
    Date of Patent: September 12, 2017
    Assignee: RESEARCH COOPERATION FOUNDATION OF YEUNGHAM UNIVERSITY
    Inventor: Hyun Chul Choi
  • Patent number: 9760784
    Abstract: Although persons on board in a driver's seat and a passenger seat can be detected, it is difficult to detect persons on board in a back seat and thus the number of passengers cannot be accurately measured. Image obtaining means 100 photographs the inside of a vehicle from the outside of the vehicle (a shoulder, etc.). View determining means 102 estimates the way in which the vehicle in the image photographed by the image obtaining means 100 is viewed, and outputs a view determination result. Person detecting means 101 performs front-face, side-face, and angled-face detection on the image photographed by the image obtaining means 100, to detect persons and outputs a person detection result. In-vehicle position estimating means 103 determines at which positions in the vehicle the detected persons are present, using the person detection information and the view determination result information, and outputs in-vehicle position estimation information.
    Type: Grant
    Filed: October 10, 2013
    Date of Patent: September 12, 2017
    Assignee: NEC CORPORATION
    Inventor: Ryoma Oami
  • Patent number: 9754259
    Abstract: Systems, methods, and non-transitory computer-readable media can acquire historical data including a plurality of features associated with known legitimate activities and with known illegitimate activities. A machine learning technique can be applied to the historical data to gain information about the plurality of features associated with the known legitimate activities and with the known illegitimate activities. A decision tree can be generated based on at least a portion of the information about the plurality of features. A node in the decision tree that satisfies specified precision criteria can be identified. A rule can be created based on the node. One or more illegitimate activities can be identified based on the rule.
    Type: Grant
    Filed: May 18, 2016
    Date of Patent: September 5, 2017
    Assignee: Facebook, Inc.
    Inventor: Daniel Adam Jenson
  • Patent number: 9747556
    Abstract: An automated Web portal template generation method includes parsing, via a parser subsystem, a number of Webpages of a first Website from which a Web portal template to be customized is to be accessed. The method further includes producing an entity feature set for the first Website based on a result of the parsing and processing the entity feature set for the first Website via a classifier subsystem to produce a set of data that represents, for each of a plurality of entities, a respective probability of the entity belonging to a respective one of a plurality of classes. The method additionally includes performing, by a color matching subsystem, color matching on the set of data produced by the classifier subsystem to generate a number of proposed color combinations for a proposed customization of the Web portal template.
    Type: Grant
    Filed: August 20, 2014
    Date of Patent: August 29, 2017
    Assignee: Vertafore, Inc.
    Inventors: Sara Garrison, Aleksey Sinyagin
  • Patent number: 9742983
    Abstract: An image capturing apparatus comprises a detection unit configured to detect a subject, a focus adjusting unit configured to adjust a position of a focusing lens of an imaging optical system so as to focus the imaging optical system on the subject, and a setting unit configured to set a focus detection area, which is an area within an image capturing screen where the focus adjusting unit detects focus, based on detection results of the detection unit, wherein the setting unit sets the focus detection area based on the position of the face of the subject in a case where the subject is moving, and sets the focus detection area based on the position of the eye of the subject in a case where the subject is still.
    Type: Grant
    Filed: February 1, 2016
    Date of Patent: August 22, 2017
    Assignee: CANON KABUSHIKI KAISHA
    Inventor: Naoki Iwasaki
  • Patent number: 9740924
    Abstract: In order to classify the presented pose of a human hand, a feature distribution map of the pose is compared to reference feature distribution maps. To generate a feature distribution map, each contour point of a hand is analyzed to determine a corresponding feature set. The feature set of a contour point includes a distance feature and an angle feature of the contour point in relation to one of its neighboring contour points. The feature sets generated from an observed pose are compared to feature sets of reference poses to determine which of the reference poses most closely matches the presented pose.
    Type: Grant
    Filed: September 26, 2014
    Date of Patent: August 22, 2017
    Assignee: Amazon Technologies, Inc.
    Inventors: Sowmya Gopalan, Manika Puri
  • Patent number: 9740925
    Abstract: A method and system for the spotting of keywords in a handwritten document, the method comprising the steps of inputting an image of the handwritten document, performing word segmentation on the image to obtain segmented words, performing word matching, and outputting the spotted keywords. The word matching itself consisting in the substeps of performing character segmentation on the segmented words, performing character recognition on the segmented characters, performing distance computations on the recognized characters using a Generalized Hidden Markov Model with ergodic topology to identify words based on character models and performing nonkeyword rejection using a classifier based on a combination of Gaussian Mixture Models, Hidden Markov Models and Support Vector Machines.
    Type: Grant
    Filed: November 19, 2013
    Date of Patent: August 22, 2017
    Assignee: IMDS AMERICA INC.
    Inventors: Mehdi Haji, Dominique Ponson
  • Patent number: 9734401
    Abstract: Technologies are generally described for detection and tracking of item features. In some examples, features of an object may be initially found through detection of one or more edges and one or more corners of the object from a first perspective. In addition, determination may be made whether the detected edges and corners of the object are also detectable from one or more other perspectives. In response to an affirmative determination, the detected edges and of the object may be marked as features to be tracked, for example, in subsequent frames of a camera feed. The perspectives may correspond to distributed locations in a substantially umbrella-shaped formation centered over the object. In other examples, lighting conditions of an environment where the object is being tracked may be programmatically controlled.
    Type: Grant
    Filed: August 8, 2014
    Date of Patent: August 15, 2017
    Assignee: RoboticVISIONTech, Inc.
    Inventor: Shawn Hunt
  • Patent number: 9723174
    Abstract: An image processing apparatus is provided. The image processing apparatus includes a data receiver, an image specifier, a replacement specifier, and a data generator. The data receiver receives data. The image specifier specifies a first image contained in the data. The replacement specifier specifies replacement data for the first image on the basis of a characteristic of the data. The data generator generates data in which the first image has been replaced with an image represented by the replacement data.
    Type: Grant
    Filed: April 29, 2016
    Date of Patent: August 1, 2017
    Assignee: FUJI XEROX CO., LTD.
    Inventor: Kohshiro Inomata
  • Patent number: 9710898
    Abstract: Embodiments of the present invention provide systems, methods, and computer storage media directed at image synthesis utilizing an active mask. In one embodiment, input is received that identifies a target region within an image that is to be synthesized. A patch synthesis technique can then be performed to synthesize the target region based on portions of a source region that are identified by the patch synthesis technique. In embodiments, the patch synthesis technique includes, for at least one iteration, generating an active mask that indicates one or more portions of the target region as inactive. This active mask can be utilized by at least one process of the patch synthesis technique to ignore the one or more portions indicated as inactive by the active mask for the at least one iteration of the patch synthesis technique. Other embodiments may be described and/or claimed.
    Type: Grant
    Filed: November 18, 2015
    Date of Patent: July 18, 2017
    Assignee: Adobe Systems Incorporated
    Inventors: Sohrab Amirghodsi, Elya Shechtman, Daniel Robert Goldman, Aliakbar Darabi
  • Patent number: 9710769
    Abstract: The disclosed embodiments illustrate methods and systems for crowdsourcing a task. The method includes extracting one or more components from at least one task based on a task category associated with the at least one task. Each component corresponds to a part of the at least one task, that requires an input from a crowdworker. Further, the one or more components are categorized in a set of first categories, each of which includes similar components, and a second category that includes dissimilar components. The categorization is based on a comparison between content of each of component, and the task category. Thereafter, a representative component from at least one category from the set of first categories, and at least one component from the second category are combined to create the task. The task, so created, is crowdsourced to one or more crowdworkers.
    Type: Grant
    Filed: April 1, 2014
    Date of Patent: July 18, 2017
    Assignee: Conduent Business Services, LLC
    Inventors: Saraschandra Karanam, Chithralekha Balamurugan, Neha Gupta, Nischal Murthy Piratla
  • Patent number: 9696900
    Abstract: Disclosed is a digital photo book creation and management system based on a client-server technology, wherein the client device comprises a touch screen. The system comprises receiving one or more source (or virtual location) of one or more images, via a client user interface and downloading one or more images from the source into the server and client databases. Further, the metadata is extracted from the downloaded images or from the source of the images and stored within the server and client databases. Further, the downloaded images are rated using gesture recognition and the rating is saved as photo's metadata. The images are compiled into a digital photo book that can be edited from the client device based on a criteria performed on the photo's metadata. Further, the compiled digital photo book can be shared with users within the network and can be signed by the visiting users.
    Type: Grant
    Filed: July 10, 2014
    Date of Patent: July 4, 2017
    Assignee: AtomKnows LLC
    Inventor: Jane P. Hong
  • Patent number: 9697584
    Abstract: An apparatus for multi-stage super-resolution is described herein. The apparatus includes a personalized dictionary, a plurality of super-resolution stages, and a reference merger. Each of the plurality of super-resolution stages correspond to at least one personalized dictionary, and the personalized dictionary is applied to an input image that is sparse-coded to generate a reconstructed image. The reference merger is to merge the reconstructed image and the input image to generate a super-resolved image for each stage.
    Type: Grant
    Filed: December 26, 2015
    Date of Patent: July 4, 2017
    Assignee: Intel Corporation
    Inventors: Chit Kwan Lin, Marcus Z. Comiter, Harnek Gulati, Gautham N. Chinya
  • Patent number: 9684831
    Abstract: A method of recognizing an object of interest in an image includes extracting a first set of features from within the image. Each extracted feature in the first set of features is then categorized as either blob-like or edge-like. A second set of features is then taken from the first set, where a number of the edge-like features to include in the second set of features is based on a relative number of edge-like features to blob-like features included in the first set of extracted features. An object of interest within the image is detected according to the second set of features.
    Type: Grant
    Filed: February 18, 2015
    Date of Patent: June 20, 2017
    Assignee: QUALCOMM Incorporated
    Inventors: Raghuraman Krishnamoorthi, Bojan Vrcelj
  • Patent number: 9684838
    Abstract: Methods, apparatuses and systems directed to pattern identification and pattern recognition. In some particular implementations, the invention provides a flexible pattern recognition platform including pattern recognition engines that can be dynamically adjusted to implement specific pattern recognition configurations for individual pattern recognition applications. In some implementations, the present invention also provides for a partition configuration where knowledge elements can be grouped and pattern recognition operations can be individually configured and arranged to allow for multi-level pattern recognition schemes.
    Type: Grant
    Filed: May 17, 2012
    Date of Patent: June 20, 2017
    Assignee: Rokio, Inc.
    Inventor: Jeffrey Brian Adams
  • Patent number: 9665768
    Abstract: Process, and related apparatus, that exploits psycho-physiological aspects involved in generation and perception of handwriting for directly inferring from the trace on the paper (or any other means on which the author writes by hand) the interpretation of writing, i.e. the sequence of characters that the trace is intended to represent.
    Type: Grant
    Filed: January 11, 2014
    Date of Patent: May 30, 2017
    Assignee: Natural Intelligent Technologies s.r.l.
    Inventors: Angelo Marcelli, Adolfo Santoro, Claudio De Stefano, Antonio Parziale, Rosa Senatore
  • Patent number: 9661334
    Abstract: A method for constructing an epitome from an image divided into non overlapping blocks is disclosed. The method comprises: determining, for each block, similar patches in the image, a similar patch being a patch with similar content; constructing at least one epitome chart for the picture from the similar patches; wherein determining, for each block, similar patches in the image comprises: a) determining, for each current block, similar blocks in the image, a similar block being a block with content similar to the content of the current block; b) determining, for one current block and for the similar blocks determined for the current block, similar patches in the image, a similar patch being a patch with content similar to the content of the current block; c) repeating step b) for a next current block for which no similar patch is determined until at least one similar patch is determined for each block in the image.
    Type: Grant
    Filed: January 28, 2015
    Date of Patent: May 23, 2017
    Assignee: THOMSON LICENSING
    Inventors: Martin Alain, Dominique Thoreau, Philippe Guillotel, Christine Guillemot
  • Patent number: 9659236
    Abstract: A system and method for training multiple pattern recognition and registration models commences with a first pattern model. The model is trained from multiple images. Composite models can be used to improve robustness or model small differences in appearance of a target region. Composite models combine data from noisy training images showing instances of underlying patterns to build a single model. A pattern recognition and registration model is generated that spans the entire range of appearances of the target pattern in the set of training images. The set of pattern models can be implemented as either separate instances of pattern finding models or as a pattern multi-model. The underlying models can be standard pattern finding models or pattern finding composite models, or a combination of both.
    Type: Grant
    Filed: November 25, 2015
    Date of Patent: May 23, 2017
    Assignee: Cognex Corporation
    Inventors: Simon Barker, David J. Michael
  • Patent number: 9659218
    Abstract: Implementations disclose predicting video start times for maximizing user engagement. A method includes applying a machine-learned model to audio-visual content features of segments of a target content item, the machine-learned model trained based on user interaction signals and audio-visual content features of a training set of content item segments, calculating, based on applying the machine-learned model, a salience score for each of the segments of the target content item, and selecting, based on the calculated salience scores, one of the segments of the target content item as a starting point for playback of the target content item.
    Type: Grant
    Filed: April 29, 2015
    Date of Patent: May 23, 2017
    Assignee: Google Inc.
    Inventors: Sanketh Shetty, Apostol Natsev, Balakrishnan Varadarajan, Tomas Izo
  • Patent number: 9633250
    Abstract: A method for face alignment operates on a face image and a set of intitial landmark locations by first aligning globally the initial locations to a set of landmark locations of a face with a prototype shape to obtain global alignment parameters, and then warping the initial locations and the image from a coordinate frame of the image to a coordinate frame of the prototype shape according to the global alignment parameters to obtain warped landmark locations and a warped face image. Features are extracted from the warped face image at the warped landmark locations, and a regression function is applied to the features to obtain updated landmark locations in the coordinate frame of the prototype shape. Finally, the updated landmark locations in the coordinate frame of the prototype shape are warped to the coordinate frame of the image, to obtain updated landmark locations.
    Type: Grant
    Filed: September 21, 2015
    Date of Patent: April 25, 2017
    Assignee: Mitsubishi Electric Research Laboratories, Inc.
    Inventors: Oncel Tuzel, Tim Marks, Salil Tambe
  • Patent number: 9626598
    Abstract: A method is provided for recognition of a ceiling portion, a vertical object portion and a ground portion in an image of indoor scene executed in an electronic system. The image is divided into a plurality of pixel sets. Expected values of each pixel sets with a ceiling distribution function, a vertical object distribution function and a ground distribution function are calculated. The expected values of each pixel set in the ceiling distribution function, the vertical object distribution function and the ground distribution function are compared to determine whether each pixel set belongs to a ceiling object, a vertical object or a ground object.
    Type: Grant
    Filed: April 13, 2015
    Date of Patent: April 18, 2017
    Assignee: NANJING YUANJUE INFORMATION AND TECHNOLOGY
    Inventors: Liming Zheng, Bingbing Cui
  • Patent number: 9619735
    Abstract: An approach is provided in which a knowledge manager processes an image using a convolutional neural network. The knowledge manager generates a pixel-level heat map of the image that includes multiple decision points corresponding to multiple pixels of the image. The knowledge manager analyzes the pixel-level heat map and detects sets of decision points that correspond to target objects. In turn, the knowledge manager marks regions of the heat map corresponding to the detected sets of per-pixel decision points, each of the regions indicating a location of the target objects.
    Type: Grant
    Filed: June 8, 2016
    Date of Patent: April 11, 2017
    Assignee: International Business Machines Corporation
    Inventors: Nicholas A. Lineback, Michael S. Ranzinger
  • Patent number: 9621741
    Abstract: Techniques are disclosed to adaptively control an image sensor and image signal processor (ISP), collectively known as an image pipeline, such that a minimum amount of power is consumed by an electronic device while producing images that can be accurately analyzed by a computer vision application. The techniques disclosed herein can be implemented in various electronic devices capable of capturing and processing image data, such as, for example, smart phones, wearable computers, laptops, tablets, and other mobile computing or imaging systems. In an embodiment, the techniques disclosed herein are implemented in a system-on-chip device.
    Type: Grant
    Filed: December 10, 2014
    Date of Patent: April 11, 2017
    Assignee: INTEL Corporation
    Inventors: Jorge E. Caviedes, Sin Lin Wu, Artem Zinevich, Frank Ghenassia, Dmitri Bitouk
  • Patent number: 9619753
    Abstract: A data analysis system includes a modeling unit, a feature-extraction unit, a processing unit and an output unit. The modeling unit creates a prediction model by a machine learning algorithm according to training data. The feature-extraction unit extracts a plurality of fragment of feature data of input data, and classifies the feature data into a plurality of groups. The processing unit obtains a probability of the input data corresponding to the prediction model by the machine learning algorithm according to the feature of one of the groups, and determines the probability. When the probability is less than a predetermined value, the processing unit uses another feature data corresponding to another group which is not used to renew the probability of the input data corresponding to the prediction model through the machine learning algorithm. When the probability is greater than or equal to the predetermined value, the processing unit classifies the input data. The output unit outputs a classification result.
    Type: Grant
    Filed: December 30, 2014
    Date of Patent: April 11, 2017
    Assignee: Winbond Electronics Corp.
    Inventors: Hsi-Chia Chang, Yen-Chin Liao
  • Patent number: 9613405
    Abstract: Techniques for enhancing an image using pixel-specific processing are disclosed. An image can be enhanced by updating certain pixels through patch aggregation. Neighboring pixels of a selected pixel are identified. Respective patch values for patches containing the selected pixel are determined. Patch values provide update information for updating the respective pixels in the patch. Relevant patch values for the selected pixel are identified by identifying associated patches of the pixel. Information from the relevant patch values of the selected pixel may be obtained. Using this information, pixel-specific processing may be performed to determine an updated pixel value for the selected pixel or for neighboring pixels of the selected pixel. Pixel-specific processes may be executed for each of the selected or neighboring pixels. These pixel-specific processes can be executed in parallel.
    Type: Grant
    Filed: June 21, 2016
    Date of Patent: April 4, 2017
    Assignee: Adobe Systems Incorporated
    Inventors: Chih-Yao Hsieh, Zhe Lin
  • Patent number: 9602728
    Abstract: A method, which is performed by an electronic device, for adjusting at least one image capturing parameter in a preview mode is disclosed. The method may include capturing a preview image of a scene including at least one text object based on a set of image capturing parameters. The method may also identify a plurality of text regions in the preview image. From the plurality of text regions, a target focus region may be selected. Based on the target focus region, the at least one image capturing parameter may be adjusted.
    Type: Grant
    Filed: June 9, 2014
    Date of Patent: March 21, 2017
    Assignee: QUALCOMM Incorporated
    Inventors: Hyun-Mook Cho, Duck-Hoon Kim, Te-Won Lee, Ananthapadmanabhan Kandhadai, Pengjun Huang
  • Patent number: 9594952
    Abstract: A character recognition apparatus includes a storage and a processor that executes a method including generating input stroke data from input coordinate data corresponding to an input operation, generating stroke combination candidates and evaluation data corresponding to the stroke combination candidates from the input stroke data, the stroke candidates including target verification strokes combined to constitute target verification characters, identifying a first stroke combination from the stroke combination candidates based on appearance probability of the target verification strokes, appearance probability data of each stroke stored in the storage, and the evaluation data, and outputting a character corresponding to the input operation based on the first stroke combination. The appearance probability indicates a probability in which the target verification strokes appear in each of the target verification characters.
    Type: Grant
    Filed: December 7, 2015
    Date of Patent: March 14, 2017
    Assignee: FUJITSU LIMITED
    Inventors: Yugo Matsuda, Masahiro Kataoka, Yasuhiro Tsuyuki
  • Patent number: 9589356
    Abstract: The invention proposes a method for determining a number (13) of transfer objects which are moving from a first subregion (8) of an observed region (5) into a second subregion (9) of the observed region (5), wherein a succession of images of the observed region (5) is recorded which identify objects (1; 2; 3; 4) and determine positions (1b, 1c; 2a, 2b, 2c; 3a, 3b, 3c; 4a, 4b) for the objects (1; 2; 3; 4), respectively, the objects (1; 2; 3; 4) are each associated either with the first subregion (8) or with the second subregion (9) on the basis of the positions (1b, 1c; 2a, 2b, 2c; 3a, 3b, 3c; 4a, 4b) of said objects, and multiple transfers of the same object between the first subregion (8) and the second subregion (9) are taken into account when determining the number (13) of transfer objects. The invention likewise proposes an appropriate system which can be used to carry out the method, said system comprising a sensor arrangement and a computation unit connected to the sensor arrangement.
    Type: Grant
    Filed: December 16, 2015
    Date of Patent: March 7, 2017
    Assignee: HELLA KGAA HUECK & CO.
    Inventor: Marko Stefanovic
  • Patent number: 9582719
    Abstract: A method and system for mapping the conditions of locations is provided. The method includes retrieving image data associated with a plurality of locations within a specified geographical area. The image data is compared to a plurality of stored image data that includes baseline measurement values associated with an expected condition level of baseline locations within a baseline geographical area. Condition score values associated with the plurality of locations are calculated. The condition score values indicate real time condition values associated with the plurality of locations. An overall condition score value associated with the specified geographical area is generated. A map indicating the overall condition score value is generated.
    Type: Grant
    Filed: November 6, 2015
    Date of Patent: February 28, 2017
    Assignee: International Business Machines Corporation
    Inventors: Richard M. Haas, Randy S. Johnson, Tedrick N. Northway, H. William Rinckel
  • Patent number: 9582483
    Abstract: Systems and methods inputting a plurality of different input documents, automatically identify repeated elements and unique elements within each of the different input documents, and automatically generate templates having the repeated elements and dynamic objects corresponding to the unique elements. The repeated elements are similar for all documents that are represented by a given template, and the unique elements have at least one difference between documents that are represented by the given template. The dynamic objects are placeholders for the unique elements within the templates. The systems and methods automatically store variable data for each of the dynamic objects from the unique elements, output the templates and the variable data, and receive user acceptance and refusal of the templates.
    Type: Grant
    Filed: July 13, 2012
    Date of Patent: February 28, 2017
    Assignee: Xerox Corporation
    Inventor: Matthew H. DeRoller
  • Patent number: 9576201
    Abstract: The present disclosure relates to a method and apparatus for classifying pixels in an input image and image processing system. The apparatus includes a first confidence determination unit configured for determining a first confidence of each of the pixels in the input image belonging to each of a plurality of predetermined classes; a detection unit configured for acquiring, by using a detector for detecting a particular class among the plurality of predetermined classes, detection results used to indicate pixels which belong to the particular class; an updating unit configured for updating the first confidences according to the detection results; and a class determination unit configured for determining the class to which each of the pixels in the input image belongs among the plurality of predetermined classes, according to the updated first confidences.
    Type: Grant
    Filed: July 22, 2014
    Date of Patent: February 21, 2017
    Assignee: Canon Kabushiki Kaisha
    Inventors: Bo Wu, Yong Jiang, You Lv, Xingwei Xu
  • Patent number: 9576210
    Abstract: A system to select video frames for optical character recognition (OCR) based on feature metrics associated with blur and sharpness. A device captures a video frame including text characters. An edge detection filter is applied to the frame to determine gradient features in perpendicular directions. An “edge map” is created from the gradient features, and points along edges in the edge map are identified. Edge transition widths are determined at each of the edge points based in local intensity minimum and maximum on opposite sides of the respective edge point in the frame. Sharper edges have smaller edge transition widths than blurry images. Statistics are determined from the edge transition widths, and the statistics are processed by a trained classifier to determine if the frame is or is not sufficiently sharp for text processing.
    Type: Grant
    Filed: September 29, 2014
    Date of Patent: February 21, 2017
    Assignee: AMAZON TECHNOLOGIES, INC.
    Inventors: Yue Liu, Qingfeng Yu, Xing Liu, Pradeep Natarajan
  • Patent number: 9576052
    Abstract: Methods and systems for dynamically training a web crawler. The web crawler maintains one or more categories each comprising a set of words. The method includes selecting at least one hyperlink in response to a query received from a user. The method further includes determining a hyperlink score for the at least one hyperlink based on a category score associated with each of one or more categories. The category score associated with each of the one or more categories is updated based at least in part on the hyperlink score. The updated category score is compared with the hyperlink score to select a category from the one or more categories. The set of words associated with the category is updated based on content of a web page pointed by the at least one hyperlink.
    Type: Grant
    Filed: July 16, 2013
    Date of Patent: February 21, 2017
    Assignee: XEROX CORPORATION
    Inventors: Nidhi Singh, Jean-Marc Coursimault, Herve Poirier, Nicolas Monet
  • Patent number: 9569086
    Abstract: A user interface module includes a display and a motion detector arranged to detect a motion. The module is arranged to display an emulated movement of a graphical representation of a first object on the display according to a motion detected by the motion detector. The first object has a physical characteristic and the user interface module further includes a controller module arranged to generate the emulated movement based on the physical characteristic.
    Type: Grant
    Filed: December 12, 2007
    Date of Patent: February 14, 2017
    Assignee: Nokia Technologies Oy
    Inventors: Kalle Pikkujamsa, Jukka Linjama, Juha Kela, Panu Korpipaa, Sanna Kallio, Tapani Rantakokko
  • Patent number: 9552524
    Abstract: A system and method for detecting a seat belt violation in a vehicle. The method may include receiving an image including a region of interest where a seat belt is visible if the seat belt is buckled. One or more first features may be extracted from the image. The one or more first features may be compared to one or more second features extracted from a plurality of sample images. It may be determined whether the seat belt is buckled in the image based upon the comparison.
    Type: Grant
    Filed: September 15, 2014
    Date of Patent: January 24, 2017
    Assignee: XEROX CORPORATION
    Inventors: Yusuf Artan, Orhan Bulan, Robert P. Loce, Peter Paul
  • Patent number: 9552513
    Abstract: Systems and methods of automatic detection of a facial feature are disclosed. Moreover, methods and systems of yaw estimation of a human head based on a geometrical model are also disclosed.
    Type: Grant
    Filed: June 2, 2015
    Date of Patent: January 24, 2017
    Inventors: Athinarayanan Sankaranarayanan, Kamal Bijlani