Directional Codes And Vectors (e.g., Freeman Chains, Compasslike Codes) Patents (Class 382/197)
  • Patent number: 10349832
    Abstract: A method of detecting the boundary between the iris and the sclera, particularly, a method of detecting the boundary between the iris and the sclera from part of a captured image of a user's eye area, is provided. According to this method, the boundary between the iris and the sclera can be quickly and precisely detected, compared to when using an existing circular boundary detector.
    Type: Grant
    Filed: June 26, 2017
    Date of Patent: July 16, 2019
    Assignee: 3E CO., LTD.
    Inventor: Min Ho Kim
  • Patent number: 10318735
    Abstract: In some embodiments, a processor can receive an input string associated with a potentially malicious artifact and convert each character in the input string into a vector of values to define a character matrix. The processor can apply a convolution matrix to a first window of the character matrix to define a first subscore, apply the convolution matrix to a second window of the character matrix to define a second sub score and combine the first subscore and the second subscore to define a score for the convolution matrix. The processor can provide the score for the convolution matrix as an input to a machine learning threat model, identify the potentially malicious artifact as malicious based on an output of the machine learning threat model, and perform a remedial action on the potentially malicious artifact based on identifying the potentially malicious artifact as malicious.
    Type: Grant
    Filed: June 22, 2017
    Date of Patent: June 11, 2019
    Assignee: Invincea, Inc.
    Inventor: Joshua Daniel Saxe
  • Patent number: 10223612
    Abstract: In a video frame processing system, a feature extractor generates, based on a plurality of data sets corresponding to a plurality of frames of a video, a plurality of feature sets, respective ones of the feature sets including features extracted from respective ones of the data sets. A first stage of the feature aggregator generates a kernel for a second stage of the feature aggregator. The kernel is adapted to content of the feature sets so as to emphasize desirable ones of the feature sets and deemphasize undesirable ones of the feature sets. In the second stage of the feature aggregator the kernel generated by the first stage is applied to the plurality of feature sets to generate a plurality of significances corresponding to the plurality of feature sets. The feature sets are weighted based on corresponding significances and weighted feature sets are aggregated to generate an aggregated feature set.
    Type: Grant
    Filed: September 1, 2016
    Date of Patent: March 5, 2019
    Assignee: MICROSOFT TECHNOLOGY LICENSING, LLC
    Inventors: Gang Hua, Peiran Ren, Jiaolong Yang
  • Patent number: 10217204
    Abstract: A method of evaluating an image quality for an imaging system and the imaging system are provided. The method in some examples includes: acquiring an image to be evaluated which is generated by the imaging system; extracting a number of sub-images from the image; obtaining a coefficient vector indicating a degree of sparsity by applying a sparse decomposition on the sub-images based on a pre-set redundant sparse representation dictionary; and performing a linear transformation on the coefficient vector so as to obtain an evaluation value for the image quality. The sparse dictionary is learned by only using a few high quality perspective images, and then the image quality is evaluated based on the sparse degree of the image obtained by using the sparse dictionary. A convenient and rapid no-reference image quality evaluation is achieved.
    Type: Grant
    Filed: January 19, 2017
    Date of Patent: February 26, 2019
    Assignee: Nuctech Company Limited
    Inventors: Zhiqiang Chen, Yuanjing Li, Li Zhang, Ziran Zhao, Yaohong Liu, Jianping Gu, Zhiming Wang
  • Patent number: 10187546
    Abstract: The present application relates to a method and a device for correcting a document image captured by an image pick-up device. The method includes: determining world coordinates of four vertices of the document image; calculating an original aspect ratio of the document image based on a correspondence between the world coordinates of the four vertices and projective coordinates of the four vertices in a projective space, and an intrinsic matrix and characteristics of an extrinsic matrix of the image pick-up device; determining a projective transformation matrix based on the world coordinates of the four vertices and the aspect ratio; and obtaining a corrected document image based on the determined projective transformation matrix and the document image. According to the application, perspective transformation can be corrected by using only one captured image and an original image can be recovered based on an original aspect ratio.
    Type: Grant
    Filed: January 19, 2017
    Date of Patent: January 22, 2019
    Assignee: FUJITSU LIMITED
    Inventors: Wei Liu, Wei Fan, Jun Sun
  • Patent number: 10140315
    Abstract: Methods for identifying visual portions of visual media files that are responsive to at least one visual media input file submitted as a search query are provided. In one aspect, a method includes generating an index of visual portions of a plurality of visual media files from a collection of media files, and receiving from a user a search query, for the collection, including at least one visual media input file. The method also includes analyzing the index to identify at least one responsive visual media file from the collection that includes a visual portion associated with a visual similarity score, to the at least one visual media input file, which exceeds a similarity threshold value, and providing, in response to the search query, an identifier of the at least one responsive visual media file for display as responsive to the search query. Systems and machine-readable media are also provided.
    Type: Grant
    Filed: June 20, 2016
    Date of Patent: November 27, 2018
    Assignee: Shutterstock, Inc.
    Inventors: Heath Hohwald, Kevin Scott Lester
  • Patent number: 10095912
    Abstract: A biological image processing device includes: a memory; and a processor coupled to the memory and the processor configured to execute a process, the process comprising: extracting at least one line from contours of a palm and principal lines of the palm shown in an image of the palm; extracting pixels of the line extracted by the extracting of at least one line; setting a local region for each of the pixels, each local region including each corresponding one of the pixels; calculating, for each local region, local orientations of the line extracted by the extracting of at least one line, in accordance with gradient intensities related to respective orientations of each corresponding local region; and calculating an orientation of the palm in accordance with a statistical amount of the respective local orientations.
    Type: Grant
    Filed: February 6, 2017
    Date of Patent: October 9, 2018
    Assignee: FUJITSU LIMITED
    Inventor: Tomoaki Matsunami
  • Patent number: 10096108
    Abstract: A medical image segmentation method includes: step 1, initialize of the number of clusters and determine the initial values of the cluster centroids; step 2, calculate a Hausdorff distance between each cluster centroid and each pixel in the image; step 3, calculate a membership function of each pixel based on the Hausdorff distance and a Euclid distance between the cluster centroids and each pixel; step 4, calculate an objective function, cluster the pixels of the image based on the updated membership function, and update the centroid value; and repeating step 2-4, until a difference between two objective function values is less than a threshold value. Then the membership function from the last iteration is a final segmentation.
    Type: Grant
    Filed: December 22, 2016
    Date of Patent: October 9, 2018
    Inventors: Yuan Feng, Long Huang
  • Patent number: 10049097
    Abstract: The present disclosure discloses methods and systems for creating a multi-layered Optical Character Recognition (OCR) document, the multi-layered OCR document facilitates selection of the desired text from the multi-layered OCR document. The method includes receiving a scanned image corresponding to a document, the document includes text information. A binary image is generated from the scanned image. Then, a morphological dilation operation is performed to create one or more text groups, using a horizontal structuring element and a vertical structuring element. Thereafter, OCR operation is applied on each text group to generate a corresponding OCR layer. The one or more OCR layers are then combined while creating a multi-layered OCR document. Finally, the combined OCR layers are superimposed as invisible text layers over the scanned image to create the multi-layered OCR document.
    Type: Grant
    Filed: January 27, 2017
    Date of Patent: August 14, 2018
    Assignee: XEROX CORPORATION
    Inventors: Sainarayanan Gopalakrishnan, Rajasekar Kanagasabai, Sudhagar Subbaian
  • Patent number: 10038873
    Abstract: A method for managing temporary images. The method comprises a computer system. The computer system identifies a context for an image for a user generating the image. The computer system determines whether the image is a temporary image based on the context of the image for the user. The computer system preforms a group of alternate actions on the image when the image is the temporary image.
    Type: Grant
    Filed: January 4, 2017
    Date of Patent: July 31, 2018
    Assignee: International Business Machines Corporation
    Inventors: Trudy L. Hewitt, Francesco C. Schembari, Merrilee F. Tomlinson, Christina L. Wetli
  • Patent number: 9905021
    Abstract: A method is provided for processing image data, wherein the image data are given by an image bitmap including pixels, wherein each pixel has exactly one pixel color value. Regions including pixels having a prespecifiable first color value and a prespecifiable maximum pixel spacing are overwritten by a rectangle having a prespecifiable second color value. This method advantageously allows for anonymization of selected image data on the basis of pixel color values to be carried out. The method may be used universally, independently of the form of the information to be anonymized (e.g., embedded texts, tables, images).
    Type: Grant
    Filed: April 13, 2015
    Date of Patent: February 27, 2018
    Assignee: Siemens Aktiengesellschaft
    Inventor: Thomas Hertlein
  • Patent number: 9830528
    Abstract: A method may include determining a value indicative of an average intensity of blocks in an image. The blocks include a primary and outer blocks. Each of the outer blocks may have three, five, or more than five pixels. The image may describe an external pixel lying between the primary and at least one of the outer blocks. The external pixel may not contribute to the value indicative of the average intensity of any of the blocks. The image may also describe a common internal pixel lying within two of the blocks. The common pixel may contribute to the value indicative of the average intensity of the two of the blocks. The method may include comparing the value indicative of the average intensity of the primary block to the values of the outer blocks, and quantifying a feature represented by the image by generating a characteristic number.
    Type: Grant
    Filed: December 9, 2015
    Date of Patent: November 28, 2017
    Assignee: Axis AB
    Inventors: Jiandan Chen, Anders Lloyd, Niclas Danielsson
  • Patent number: 9817471
    Abstract: An image transmitted, through a network, from any of at least one terminal having a function of capturing an image or obtaining an image from another device is obtained. A probability that the obtained image includes a certain imaging target is calculated. If the probability is higher than a first threshold, information indicating the certain imaging target is added to the image. If the probability is lower than a second threshold, the information indicating the certain imaging target is not to the image. If the probability is equal to or higher than the second threshold and if the probability is equal to or lower than the first threshold, the image and request reception information for requesting addition of the information is transmitted to the image to any of the at least one terminal through the network.
    Type: Grant
    Filed: August 4, 2016
    Date of Patent: November 14, 2017
    Assignee: PANASONIC INTELLECTUAL PROPERTY CORPORATION OF AMERICA
    Inventors: Reiko Hagawa, Yasunori Ishii, Sotaro Tsukizawa, Masaki Takahashi
  • Patent number: 9659231
    Abstract: Provided is an image analysis device for specifying an image region of an object being the basis for design similarity determination. An image analysis device 100 includes units 112, 113. The unit 112 moves original images Im1 and Im2 including a designated image region H with a rotationally symmetry on a first point P relative to a reference image Im2(j), calculates a correlation value cim(j) at each relative position, and detects the position of the point P in the images Im1, Im2, according to a geometric model, based on the amount of rotation ?(j) at which the cim(j) is the maximum and a vector v(j). The unit 113 specifies the range of the region H in the images Im1, Im2 based on the position of the point P and the distribution of the brightness values of pixels in the images Im1, Im2.
    Type: Grant
    Filed: October 8, 2015
    Date of Patent: May 23, 2017
    Assignee: HONDA MOTOR CO., LTD.
    Inventor: Saburo Sakai
  • Patent number: 9626579
    Abstract: A method includes receiving image data and performing a non-maximum suppression (NMS) operation on the image data. The method also includes initiating an edge tracking by hysteresis (ETH) operation on a portion of the image data prior to completion of the NMS operation.
    Type: Grant
    Filed: May 5, 2014
    Date of Patent: April 18, 2017
    Assignee: QUALCOMM Incorporated
    Inventors: Kim-Chyan Gan, Mao Zeng, Lucian Codrescu
  • Patent number: 9519838
    Abstract: The present invention relates to a character recognition method. Same comprises: step 1: reading in a character image; step 2: image preprocessing; step 3: character splitting; step 4: edge extraction; utilizing an edge detection operator to detect edge points of characters; step 5: feature extraction, where the features of each edge point of each character are all expressed with distances running from other edge points of the character to the edge point and are represented by eigenvectors (Pk, Pk, . . . P); step 6: feature processing, mapping the eigenvectors (Pk, Pk, . . . P) into a matrix, T, thus allowing all characters to be identical in eigenvector dimension; and, step 7: template matching recognition.
    Type: Grant
    Filed: June 26, 2013
    Date of Patent: December 13, 2016
    Assignee: GRG Banking Equipment Co., Ltd.
    Inventors: Tiancai Liang, Kun Wang, Weifeng Wang, Siwei Liu
  • Patent number: 9508018
    Abstract: An object detection system and a method of detecting an object in an image are disclosed. In an embodiment, a method for detecting the object includes computing one or more feature planes of one or more types for each image pixel of the image. A plurality of cells is defined in the image, where each cell includes first through nth number of pixels, and starting locations of each cell in the image in horizontal and vertical directions are integral multiples of predefined horizontal and vertical step sizes, respectively. One or more feature plane summations of one or more types are computed for each cell. A feature vector is determined for an image portion of the image based on a set of feature plane summations, and the feature vector is compared with a corresponding object classifier to detect a presence of the corresponding object in the image portion of the image.
    Type: Grant
    Filed: November 24, 2014
    Date of Patent: November 29, 2016
    Assignee: Texas Instruments Incorporated
    Inventors: Kumar Arrakutti Desappan, Manu Mathew, Pramod Kumar Swami
  • Patent number: 9417224
    Abstract: A system and method using a native mobile application for gardening is disclosed herein. An image of a plant or plant disease is uploaded and sent to a site for identification and advise from an expert or expert database. A response is provided to the gardener, along with helpful information.
    Type: Grant
    Filed: July 20, 2014
    Date of Patent: August 16, 2016
    Inventor: Alex Shah
  • Patent number: 9344745
    Abstract: Systems and methods for generating group portraits using video chat systems are provided. A communications system can generate a group portrait that includes an image of two or more users. For example, a communications system can generate a group portrait that includes an image of each user participating in a video conference, and the group portrait may serve as a memento from the video conference.
    Type: Grant
    Filed: April 1, 2010
    Date of Patent: May 17, 2016
    Assignee: Shindig, Inc.
    Inventor: Steven M. Gottlieb
  • Patent number: 9292931
    Abstract: The invention provides an apparatus and method for extracting a boundary of an object in an image and an electronic device. The apparatus includes: a position determining unit, configured to determine a start point and an end point of a boundary of an object in an image and to determine a position of a reference point relevant to the start point and the end point; a first direction determining unit, configured to determine a first direction of the boundary; a gradient map obtaining unit, configured to obtain a gradient map of a first region; a gradient attenuating unit, configured to attenuate in the gradient map the gradients of a second region; and an extracting unit, configured to extract a boundary of an object. The technology of the invention can improve the accuracy of boundary extracting, and can be applied in the field of image processing.
    Type: Grant
    Filed: June 7, 2013
    Date of Patent: March 22, 2016
    Assignee: FUJITSU LIMITED
    Inventors: Pan Pan, Yuan He, Jun Sun, Satoshi Naoi
  • Patent number: 9183041
    Abstract: According to one aspect of the present disclosure a system and technique for input/output traffic backpressure prediction is disclosed. The system includes a processor unit and logic executable by the processor unit to: determine, for each of a plurality of memory transactions, a traffic value corresponding to a time for performing the respective memory transactions; responsive to determining the traffic value for a respective memory transaction, determine a median value based on the determined traffic values; determine whether successive median values are incrementing; and responsive to a quantity of successively incrementing median values exceeding a threshold, indicate a prediction of a backpressure condition.
    Type: Grant
    Filed: September 21, 2012
    Date of Patent: November 10, 2015
    Assignee: INTERNATIONAL BUSINESS MACHINES CORPORATION
    Inventors: Norbert Hagspiel, Matthias Klein
  • Patent number: 9183042
    Abstract: According to one aspect of the present disclosure, a method and technique for input/output traffic backpressure prediction is disclosed. The method includes: performing a plurality of memory transactions; determining, for each memory transaction, a traffic value corresponding to a time for performing the respective memory transactions; responsive to determining the traffic value for a respective memory transaction, determining a median value based on the determined traffic values; determining whether successive median values are incrementing; and responsive to a quantity of successively incrementing median values exceeding a threshold, indicating a prediction of a backpressure condition.
    Type: Grant
    Filed: November 11, 2013
    Date of Patent: November 10, 2015
    Assignee: INTERNATIONAL BUSINESS MACHINES CORPORATION
    Inventors: Norbert Hagspiel, Matthias Klein
  • Patent number: 9175966
    Abstract: A monitoring device receives an aerial image, data relating to a vehicle and a request for an image of the vehicle. The vehicle is located in the image. A portion of the aerial image that includes the vehicle and/or information based thereon is provided.
    Type: Grant
    Filed: October 15, 2013
    Date of Patent: November 3, 2015
    Assignee: FORD GLOBAL TECHNOLOGIES, LLC
    Inventors: Douglas R. Martin, Kenneth J. Miller
  • Patent number: 9142037
    Abstract: When encoding a set of texture data elements 30 for use in a graphics processing system, the direction along which the data values of the set of texture data elements in question exhibit the greatest variance in the color space is estimated by using one or more infinite planes 41 to divide the texture data elements in the color space. For each such plane, texture data element values on each side of the plane are added up to give respective sum points 48, 49, and the vector 50 between these two sum points determined. The direction in the data space of one of the determined vectors 50 is then used to derive endpoint color values to use when encoding the set of texture data elements.
    Type: Grant
    Filed: July 2, 2013
    Date of Patent: September 22, 2015
    Assignee: ARM LIMITED
    Inventor: Jorn Nystad
  • Patent number: 9074868
    Abstract: Measurement accuracy of a remote visual inspection (RVI) system is tested using a test object including a test feature having a known geometric characteristic. Using a controller, attachment of a detachable measurement optical tip to an RVI probe is detected. A user is then prompted to perform testing of the measurement accuracy. When the user indicates the test feature is visible, the system captures one or more images of the test feature, determines coordinates of the test feature from the images, and measures a geometric characteristic of the test feature using the coordinates. An accuracy result is determined using the measured geometric characteristic and the known geometric characteristic, and an indication is provided, e.g., to the user, of the result of the comparison. An RVI system with a user-prompt device is also described.
    Type: Grant
    Filed: May 13, 2013
    Date of Patent: July 7, 2015
    Assignee: General Electric Company
    Inventors: Clark Alexander Bendall, Joshua Lynn Scott
  • Publication number: 20150146988
    Abstract: An image angle variation detection means 81 detects an image angle variation of a imaging device from videos shot by the imaging device based on fixed point information including the position of a fixed point specified by an image angle of the imaging device and the features indicating the characteristics of the fixed point, and a situation determination means 82 determines that the image angle of the imaging device is varied when a change in the image angle variation is stable for a certain period of time.
    Type: Application
    Filed: June 17, 2013
    Publication date: May 28, 2015
    Inventor: Ryoma Oami
  • Publication number: 20150146989
    Abstract: An image processing method includes acquiring the number of feature vectors similar to at least one specific feature vector, including the at least one specific feature vector, among feature vectors of a plurality of keypoints in a comparison source image, and the number of feature vectors similar to the at least one specific feature vector in at least one comparison destination image, and comparing the comparison source image with the at least one comparison destination image by using, as a value for evaluating a similarity between the comparison source image and the at least one comparison destination image, a smaller value instead of a larger value of the numbers of feature vectors acquired with respect to the at least one specific feature vector.
    Type: Application
    Filed: November 24, 2014
    Publication date: May 28, 2015
    Inventors: Hirotaka Shiiyama, Masahiro Matsushita
  • Patent number: 9036903
    Abstract: A learning device includes a gradient feature extraction unit which extracts a gradient feature amount including a gradient direction at each coordinate and a gradient intensity value thereof based on an amount of variation between luminance at each coordinate of an inputted learning target pattern and luminance at a periphery thereof, a sum difference feature extraction unit which calculates a predetermined sum difference feature amount by adding the gradient intensity values according to the gradient directions included in a predetermined gradient range indicating a range of the predetermined gradient direction based on the extracted gradient feature amount and subtracting the gradient intensity values according to the gradient directions included in the other gradient range adjacent to the predetermined gradient range from the calculated sum, and a learning unit which acquires a learning parameter at each coordinate.
    Type: Grant
    Filed: December 24, 2010
    Date of Patent: May 19, 2015
    Assignee: NEC CORPORATION
    Inventor: Toshinori Hosoi
  • Patent number: 9036918
    Abstract: Whether a processing target block belongs to a quiescence area is determined based on a motion vector of an adjacent block whose encoding mode is determined to not be the intra-encoding mode and a reference index indicating a reference image of the adjacent block, whereby a reference index indicating a reference image of the processing target block is determined. If all the encoding modes of the adjacent blocks are the intra-encoding mode, an index number assigned to an image temporally closest and in a field different from a field to which the processing target block belongs is selected. If the processing target block belongs to a quiescence area, an index number assigned to an image temporally closest and in a same field as the processing target block is selected. If the processing target block belongs to a moving area, an index number of a reference image temporally closest is selected.
    Type: Grant
    Filed: September 12, 2012
    Date of Patent: May 19, 2015
    Assignee: FUJITSU LIMITED
    Inventor: Hidenobu Miyoshi
  • Patent number: 9025885
    Abstract: Disclosed is a global motion detecting method which includes receiving a video sequence of input images, calculating local motion vectors, one for each image block of a current input image, grouping image blocks of the current input image into image block groups, calculating a group motion parameter of each of the image block groups based on local motion vectors of the image blocks in each respective image block group, and determining a global motion parameter of the currently input image according to the group motion parameters.
    Type: Grant
    Filed: March 14, 2013
    Date of Patent: May 5, 2015
    Assignee: Samsung Electronics Co., Ltd.
    Inventors: Sharon Alpert, Yowon Jeong, Omri Govrin
  • Patent number: 9025884
    Abstract: In a method for processing an image file using a computing device, an image file from a storage system is read. If an image in the image file is slanted, an incision coordinates according to a configuration file in the storage system and a preset formula is calculated. The method incises the image using the calculated incision coordinates and storing the incised image in a new image file. If an image in the new image file is not slanted, the method further determines whether the image has been incised. If the image has been incised, the method records the calculated incision coordinates that make the image not be slanted as optimal incision coordinates, and stores the optimal incision coordinates into a database.
    Type: Grant
    Filed: August 14, 2012
    Date of Patent: May 5, 2015
    Assignee: Zhongshan Innocloud Intellectual Property Services Co., Ltd.
    Inventors: Chung-I Lee, Hai-Hong Lin, De-Yi Xie, Shuai-Jun Tao, Zhi-Qiang Yi, Jun-Chao Yao
  • Patent number: 9020271
    Abstract: Systems and methods for clustering a plurality of feature vectors. A hierarchical clustering algorithm is performed on the plurality of feature vectors to provide a plurality of clusters and a cluster similarity measure for each cluster representing the quality of the cluster. Each cluster of the plurality of clusters with a cluster similarity measure meeting a threshold value is accepted. A clustering algorithm is performed on each cluster that fails to meet the threshold value to provide a set of subclusters each having an associated cluster similarity measure. Each subcluster having a cluster similarity measure meeting the threshold value is accepted.
    Type: Grant
    Filed: July 31, 2012
    Date of Patent: April 28, 2015
    Assignee: Hewlett-Packard Development Company, L.P.
    Inventors: Vinay Deolalikar, Hernan Laffitte
  • Patent number: 9020272
    Abstract: In one aspect, the present disclosure can be embodied in a method that includes approximating an outline of a vector image using a set of circular arcs. A signed distance value is computed for a selected group of points in a two-dimensional grid associated with the vector image based on a location of each point relative to the approximated outline of the vector image. The nearest are from the respective location of each point in the selected group is identified and the corresponding signed distance value is assigned to each point. The vector image is reproduced based on the signed distance value assigned to each point in the selected group.
    Type: Grant
    Filed: March 15, 2013
    Date of Patent: April 28, 2015
    Assignee: Google Inc.
    Inventors: Seyed Behdad Esfahbod Mirhosseinzadeh Sarabi, Maysum Panju
  • Patent number: 9008439
    Abstract: A method of comparing two object poses, wherein each object pose is expressed in terms of position, orientation and scale with respect to a common coordinate system, the method comprising: calculating a distance between the two object poses, the distance being calculated using the distance function: d sRt ? ( X , Y ) = d s 2 ? ( X , Y ) ? s 2 + d r 2 ? ( X , Y ) ? r 2 + d t 2 ? ( X , Y ) ? t 2 .
    Type: Grant
    Filed: August 14, 2014
    Date of Patent: April 14, 2015
    Assignee: Kabushiki Kaisha Toshiba
    Inventors: Minh-Tri Pham, Oliver Woodford, Frank Perbet, Atsuto Maki, Bjorn Stenger, Roberto Cipolla
  • Patent number: 8989503
    Abstract: A method for identifying a set of key video frames from a video sequence comprising extracting feature vectors for each video frame and applying a group sparsity algorithm to represent the feature vector for a particular video frame as a group sparse combination of the feature vectors for the other video frames. Weighting coefficients associated with the group sparse combination are analyzed to determine video frame clusters of temporally-contiguous, similar video frames. The video sequence is segmented into scenes by identifying scene boundaries based on the determined video frame clusters.
    Type: Grant
    Filed: August 3, 2012
    Date of Patent: March 24, 2015
    Assignee: Kodak Alaris Inc.
    Inventors: Mrityunjay Kumar, Alexander C. Loui, Bruce Harold Pillman
  • Patent number: 8977059
    Abstract: Integrating features is disclosed, including: determining a value associated with a temporal feature for a point; determining a value associated with a spatial feature associated with the temporal feature; including the value associated with a spatial feature and the value associated with the temporal feature into a feature vector; and using the feature vector to decode for a character. Determining a transform is also disclosed, including: determining, for a point associated with a sequence of points, a set of points including: the point, a first subset of points of the sequence preceding a sequence position associated with the point, and a second subset of points following the sequence position associated with the point; and determining the transform associated with the point based at least in part on the set of points.
    Type: Grant
    Filed: June 4, 2012
    Date of Patent: March 10, 2015
    Assignee: Apple Inc.
    Inventors: Jerome R. Bellegarda, Jannes G. A. Dolfing, Devang K. Naik
  • Patent number: 8977060
    Abstract: The techniques and systems described herein are directed to isolating part-centric motion in a visual scene and stabilizing (e.g., removing) motion in the visual scene that is associated with camera-centric motion and/or object-centric motion. By removing the motion that is associated with the camera-centric motion and/or the object-centric motion, the techniques are able to focus motion feature extraction mechanisms (e.g., temporal differencing) on the isolated part-centric motion. The extracted motion features may then be used to recognize and/or detect the particular type of object and/or estimate a pose or position of a particular type of object.
    Type: Grant
    Filed: May 16, 2013
    Date of Patent: March 10, 2015
    Assignee: Microsoft Technology Licensing, LLC
    Inventors: Piotr Dollar, Charles Lawrence Zitnick, III, Dennis I. Park
  • Publication number: 20150063706
    Abstract: Disclosed is a computer-implemented method of vectorizing a curve. The method determines a reference point distanced from the curve and a first point on the curve associated with the reference point, and a line running through the reference point from a tangent to the curve at the first point, where the line and tangent are parallel. A shift point is then determined on the determined line, and the curve is vectorized by a path formed from a plurality of linear segments, the plurality of segments comprising at least a segment joining the reference point and the shift point. From the vectorization of the curve the curve can be rendered, particularly using dashed strokes.
    Type: Application
    Filed: August 26, 2014
    Publication date: March 5, 2015
    Inventors: Ari Pekka LAMPINEN, Hervé Soulard
  • Publication number: 20150055874
    Abstract: An image analyzing apparatus includes computer. The computer is programmed to obtain a motion vector from a first image toward a second image and calculate an evaluation value depending on a magnitude of frequency components. The frequency components have a higher frequency than a first frequency determined based on the motion vector. The computer is also programmed to detect a particular area from the first image based on the evaluation value.
    Type: Application
    Filed: April 3, 2014
    Publication date: February 26, 2015
    Applicant: KABUSHIKI KAISHA TOSHIBA
    Inventors: Takuma YAMAMOTO, Yasunori Taguchi, Toshimitsu Kaneko
  • Publication number: 20150055873
    Abstract: Provided are an image alignment apparatus and an image alignment method using the same. The image alignment apparatus includes a first conversion function estimation unit for estimating a first conversion function based on feature point information that is extracted from a first image, which is captured by using a first image sensor, and a second image, which is captured by using a second image; and a second conversion function estimation unit for estimating a second conversion function based on motion information that is extracted from the first image and the second image.
    Type: Application
    Filed: February 11, 2014
    Publication date: February 26, 2015
    Applicant: Samsung Techwin Co., Ltd.
    Inventors: Joon Sung LEE, Jae Yoon OH, Gon Soo KIM
  • Patent number: 8965117
    Abstract: Embodiments of the subject technology provide methods and systems of image pre-processing for improving the accuracy of optical character recognition (OCR) and reducing the power consumption on a given computing device (e.g., mobile computing device). The subject technology, in some examples, classifies an image received from a camera of a mobile computing device into one or more classes: 1) normal background, 2) textured background, 3) image with text, 4) image with barcode, 5) image with QR code, and/or 6) image with clutter or “garbage.” Based on the classes associated with the image, the subject technology may forgo certain image processing operations, when the image is not associated with a particular class, in order to save resources (e.g., CPU cycles, battery power, memory usage, etc.) on the mobile computing device.
    Type: Grant
    Filed: December 17, 2013
    Date of Patent: February 24, 2015
    Assignee: Amazon Technologies, Inc.
    Inventors: Oleg Rybakov, Christopher John Lish, Chang Yuan, Junxiong Jia, Rakesh Madhavan Nambiar, Matias Omar Gregorio Benitez
  • Patent number: 8965054
    Abstract: A method executed by a computer system for detecting edges comprises receiving an image comprising a plurality of pixels, determining a phase congruency value for a pixel, where the phase congruency value comprises a plurality of phase congruency components, and determining if the phase congruency value satisfies a phase congruency criteria. If the phase congruency value satisfies the phase congruency criteria, the computer system categorizes the pixel as an edge pixel. If the phase congruency value does not satisfy the phase congruency criteria, the computer system compares a first phase congruency component of the plurality of phase congruency components to a phase congruency component criteria. If the first phase congruency component satisfies the phase congruency component criteria, the computer system categorizes the pixel as an edge pixel, and if the first phase congruency component does not satisfy the phase congruency component criteria, categorizes the pixel as a non-edge pixel.
    Type: Grant
    Filed: December 31, 2013
    Date of Patent: February 24, 2015
    Assignee: Verint Systems Ltd.
    Inventors: Oren Golan, Shmuel Kiro, Itshak Horovitz
  • Publication number: 20150049953
    Abstract: A computer-implemented method of mapping. The method includes analyzing images of faces in a plurality of pictures to generate content vectors, obtaining information regarding one or more vector dimensions of interest, at least some of the one or more dimensions of interest corresponding to facial expressions of emotion, and generating a representation of the location. Appearance of regions in the map varies in accordance with values of the content vectors for the one or more vector dimensions of interest. The method also includes using the representation, the step of using comprising at least one of storing, transmitting, and displaying.
    Type: Application
    Filed: August 15, 2014
    Publication date: February 19, 2015
    Inventors: Javier MOVELLAN, Josua SUSSKIND
  • Publication number: 20150043827
    Abstract: A device includes a characteristic point extraction part configured to extract, as a characteristic point, a pattern projected on an object from an imaging image obtained by imaging the pattern projected on the object, and a parameter calculation part configured to calculate a parameter configured to minimize a distance between the characteristic point on an imaging image plane and a line obtained by projecting a projection light ray corresponding to the characteristic point onto the imaging image plane.
    Type: Application
    Filed: August 11, 2014
    Publication date: February 12, 2015
    Inventor: Takayuki Hara
  • Publication number: 20150043814
    Abstract: An image code conversion device for image information and its method are provided. By extracting a plurality of features of an image based on raw image information taken by an image reading device and converting the features into a unique code, the image code conversion device for image information and its method can stably convert the same image into the same image code. Raw image information acquired by an image reading device is converted into a plurality of pieces of developed image information on the basis of geometrical or physical factors. From the plurality of developed image information, feature information is quantified by a self-organization processing based on a probability scale. The quantified feature information is converted into a digital code. Accordingly, a unique image code corresponding to the raw image information is generated.
    Type: Application
    Filed: January 30, 2014
    Publication date: February 12, 2015
    Applicant: APOLLO JAPAN CO., LTD.
    Inventors: Zecang GU, Ikuko KISHIGAMI
  • Publication number: 20150036933
    Abstract: A method and apparatus for matching key pixels of images and a method and apparatus for matching images are disclosed. The method for matching key pixels of images includes: determining, for a key pixel to be matched of a first image, a first eigenvector, in a set of eigenvectors of a second image, matching an eigenvector of the key pixel to be matched; determining a second eigenvector, in a set of eigenvectors of the first image, matching the first eigenvector; and determining the key pixel to be matched and a key pixel corresponding to the first eigenvector as a pair of matching pixels when a coordinate vector of the key pixel to be matched is the same as a coordinate vector of a key pixel corresponding to the second eigenvector.
    Type: Application
    Filed: November 27, 2013
    Publication date: February 5, 2015
    Applicants: PEKING UNIVERSITY FOUNDER GROUP CO., LTD., PEKING UNIVERSITY, BEIJING FOUNDER APABI TECHNOLOGY LTD.
    Inventors: Yan LIU, Xiaoqing LV, Jianbo XU, Zhi TANG
  • Publication number: 20150036934
    Abstract: A computer-implemented (including method implemented using laptop, desktop, mobile, and wearable devices) method for image filtering. The method includes analyzing each image to generate a content vector for the image; applying an interest operator to the content vector, the interest operator being based on a plurality of pictures with desirable characteristics, thereby obtaining an interest index for the image; comparing the interest index for the image to an interest threshold; and taking one or more actions or abstaining from one or more actions based on a result of the step of comparing. Also, related systems and articles of manufacture.
    Type: Application
    Filed: August 4, 2014
    Publication date: February 5, 2015
    Inventors: Javier MOVELLAN, Ken DENMAN, Joshua SUSSKIND
  • Publication number: 20150030252
    Abstract: The present invention is a method for carrying out high-level activity recognition on a wide variety of videos. In one embodiment, the invention leverages the fact that a large number of smaller action detectors, when pooled appropriately, can provide high-level semantically rich features that are superior to low-level features in discriminating videos. Another embodiment recognizes activity using a bank of template objects corresponding to actions and having template sub-vectors. The video is processed to obtain a featurized video and a corresponding vector is calculated. The vector is correlated with each template object sub-vector to obtain a correlation vector. The correlation vectors are computed into a volume, and maximum values are determined corresponding to one or more actions.
    Type: Application
    Filed: December 17, 2012
    Publication date: January 29, 2015
    Applicant: The Research Foundation for The State University of New York
    Inventors: Jason J. Corso, Sreemanananth Sadanand
  • Publication number: 20150030242
    Abstract: A method and system is provided for combining information from a plurality of source images to form a fused image. The fused image is generated by the combination of the source images based on both local features and global features computed from the source images. Local features are computed for local regions in each source image. For each source image, the computed local features are further processed to form a local weight matrix. Global features are computed for the source images. For each source image, the computed global features are further processed to form a global weight vector. For each source image, its corresponding local weight matrix and its corresponding global weight vector are combined to form a final weight matrix. The source images are then weighted by the final weight matrices to generate the fused image.
    Type: Application
    Filed: July 26, 2013
    Publication date: January 29, 2015
    Inventor: Rui Shen
  • Publication number: 20150030238
    Abstract: A system may be configured as an image recognition machine that utilizes an image feature representation called local feature embedding (LFE). LFE enables generation of a feature vector that captures salient visual properties of an image to address both the fine-grained aspects and the coarse-grained aspects of recognizing a visual pattern depicted in the image. Configured to utilize image feature vectors with LFE, the system may implement a nearest class mean (NCM) classifier, as well as a scalable recognition algorithm with metric learning and max margin template selection. Accordingly, the system may be updated to accommodate new classes with very little added computational cost. This may have the effect of enabling the system to readily handle open-ended image classification problems.
    Type: Application
    Filed: July 29, 2013
    Publication date: January 29, 2015
    Applicant: Adobe Systems Incorporated
    Inventors: Jianchao Yang, Guang Chen, Jonathan Brandt, Hailin Jin, Elya Shechtman, Aseem Omprakash Agarwala