Patents Examined by Wenpeng Chen
-
Patent number: 7991196Abstract: Methods and systems for image processing are provided. A method for processing images of a scene includes receiving image data of a reference and a current frame; generating N motion vectors that describe motion of the image data within the scene by computing a correlation function on the reference and current frames at each of N registration points; registering the current frame based on the N motion vectors to produce a registered current frame; and updating the image data of the scene based on the registered current frame. Optionally, registered frames may be oversampled. Techniques for generating the N motion vectors according to roll, zoom, shift and optical flow calculations, updating image data of the scene according to switched and intermediate integration approaches, re-introducing smoothed motion into image data of the scene, re-initializing the process, and processing images of a scene and moving target within the scene are provided.Type: GrantFiled: June 17, 2010Date of Patent: August 2, 2011Assignee: Lockheed Martin CorporationInventors: Gene D. Tener, Barry Mattox, Jennifer K. Park, John Cowan, William R. Herzog, Hai-Wen Chen, James P. Kane, III, David J. Huber, Timothy Alderson
-
Patent number: 7983499Abstract: A method and apparatus prioritizing video information during coding and decoding. Video information is received and an element of the video information, such as a visual object, video object layer, video object plane or keyregion, is identified. A priority is assigned to the identified element and the video information is encoded into a bitstream, such as a visual bitstream encoded using the MPEG-4 standard, including an indication of the priority of the element.Type: GrantFiled: November 23, 2009Date of Patent: July 19, 2011Assignee: AT&T Intellectual Property II, L.P.Inventors: Barin Geoffry Haskell, Atul Puri, Robert Lewis Schmidt
-
Patent number: 7983463Abstract: A method for suppressing tagged material in prepless colonoscopy includes identifying and segmenting multiple density classes, intensity classes, or both in a volumetric medical image of a patient including a volume of the patient's colon and classifying the volume of the patient's colon into a plurality of types based upon the identified and segmented density classes, intensity classes, or both. The method further includes subtracting, suppressing, or leaving the tagged material in the image in accordance with the type classifications to produce a resulting image and visualizing the resulting image.Type: GrantFiled: November 22, 2006Date of Patent: July 19, 2011Assignee: General Electric CompanyInventors: Saad Ahmed Sirohey, Renaud Capolunghi, Huy-Nam Doan
-
Patent number: 7978880Abstract: Methods and systems for image processing are provided. A method for processing images of a scene includes receiving image data of a reference and a current frame; generating N motion vectors that describe motion of the image data within the scene by computing a correlation function on the reference and current frames at each of N registration points; registering the current frame based on the N motion vectors to produce a registered current frame; and updating the image data of the scene based on the registered current frame. Optionally, registered frames may be oversampled. Techniques for generating the N motion vectors according to roll, zoom, shift and optical flow calculations, updating image data of the scene according to switched and intermediate integration approaches, re-introducing smoothed motion into image data of the scene, re-initializing the process, and processing images of a scene and moving target within the scene are provided.Type: GrantFiled: June 17, 2010Date of Patent: July 12, 2011Assignee: Lockheed Martin CorporationInventors: Gene D. Tener, Barry Mattox, Jennifer K. Park, John Cowan, William R. Herzog, Hai-Wen Chen, James P. Kane, III, David J. Huber, Timothy Alderson
-
Patent number: 7974435Abstract: A method and system for encoding a content signal having one or more content segments. The one or more content segments of the content signal are identifying or obtained. One or more encoded patterns are superimposed or inserted in a segment portion of at least one of the one or more content segments by use of an insertion method that renders the one or more encoded patterns substantially invisible within the content signal. A method for detecting an encoded message within an encoded content signal. A pattern format is identified in at least one of the content segments. One or more encoded patterns are extracted from the one or more pattern formatted identified content segments. The one or more encoded patterns are decoded. The encoded message decoded from the encoded content signal is utilized.Type: GrantFiled: September 18, 2006Date of Patent: July 5, 2011Assignee: Koplar Interactive Systems International LLCInventors: Alan G. Maltagliati, Yousri H. Barsoum, James G. Withers
-
Patent number: 7970191Abstract: A method for downsampling fluoroscopic images and enhancing guidewire visibility during coronary angioplasty includes providing a first digitized image, filtering the image with one or more steerable filters of different angular orientations, assigning a weight W and orientation O for each pixel based on the filter response for each pixel, wherein each pixel weight is assigned to a function of a maximum filter response magnitude and the pixel orientation is calculated from the angle producing the maximum filter response if the magnitude is greater than zero, wherein guidewire pixels have a higher weight than non-guidewire pixels, and downsampling the orientation and weights to calculate a second image of half the resolution of the first image, wherein the downsampling accounts for the orientation and higher weight assigned to the guidewire pixels.Type: GrantFiled: September 25, 2007Date of Patent: June 28, 2011Assignee: Siemens Medical Solutions USA, Inc.Inventors: Bogdan Georgescu, Peter Durlak, Vassilis Athitsos, Adrian Barbu, Dorin Comaniciu
-
Patent number: 7970170Abstract: Methods and systems for image processing are provided. A method for processing images of a scene includes receiving image data of a reference and a current frame; generating N motion vectors that describe motion of the image data within the scene by computing a correlation function on the reference and current frames at each of N registration points; registering the current frame based on the N motion vectors to produce a registered current frame; and updating the image data of the scene based on the registered current frame. Optionally, registered frames may be oversampled. Techniques for generating the N motion vectors according to roll, zoom, shift and optical flow calculations, updating image data of the scene according to switched and intermediate integration approaches, re-introducing smoothed motion into image data of the scene, re-initializing the process, and processing images of a scene and moving target within the scene are provided.Type: GrantFiled: May 9, 2006Date of Patent: June 28, 2011Assignee: Lockheed Martin CorporationInventors: Gene D. Tener, Barry Mattox, Jennifer K. Park, John Cowan, William R. Herzog, Hai-Wen Chen, James P. Kane, III, David J. Huber, Timothy Alderson
-
Patent number: 7970199Abstract: A surface inspection apparatus is provided based on an optical interference scheme using a wide-band laser light source, such as diode laser, for an interferometer. In the apparatus, a diode laser with a large spectrum width having a short coherence length is used as an emitted light source; modulation optical elements for performing modulation with slightly different frequencies, and optical path length varying optical elements for adjusting the optical path length are located in each of two optical paths between a branching optical element and a combining optical element; and the above-mentioned optical path length varying optical elements are adjusted, while measuring an interference intensity, so as to maximize the interference intensity.Type: GrantFiled: June 4, 2007Date of Patent: June 28, 2011Assignee: Hitachi High-Technologies CorporationInventors: Minoru Yoshida, Yoshimasa Oshima
-
Patent number: 7961965Abstract: An apparatus for transmitting multimedia, the apparatus including an image compression unit for compressing an image portion of a multimedia data frame, an audio compression unit for compressing an audio portion of the multimedia frame, a bit rate manager in communication with said image compression unit, and said audio compression unit for controlling frame flow, and a merger unit in communication with each of said image compression unit and said audio compression unit for merging a compressed image portion and a compressed audio portion into a single compressed multimedia frame, wherein the image compression unit utilizes multidimensional hierarchical categorization trees to compress said image portion, and wherein the multidimensional hierarchical categorization trees examine pixel blocks that are structurally similar to a reference block, and organize pixel blocks in a hierarchical manner.Type: GrantFiled: November 20, 2008Date of Patent: June 14, 2011Assignee: Indinell Sociedad AnonimaInventor: Eduardo Moreira
-
Patent number: 7961916Abstract: The present invention provides an identification method. A video capture device captures an identification video at a random time. Then a suitable identification image is obtained from the identification video. Subsequently, a current user characteristic value obtained from the identification image is compared with a stored user characteristic value of a user data stored in a recognition database. When the current user characteristic value corresponds to the stored user characteristic value, then returns to capture the identification video step. When the current user characteristic value does not corresponds to the stored user characteristic value, then an error counter is incremented, and returns to the acquiring the identification video step.Type: GrantFiled: June 4, 2007Date of Patent: June 14, 2011Assignee: Compal Electronics, Inc.Inventors: Jen-Li Wang, Chien-Ming Lee
-
Patent number: 7949175Abstract: A method is disclosed for detection of miniature security mark configurations within documents and images, wherein the miniature security marks are in the form of dispersed miniature security marks and may include data marks or a combination of data marks and anchor marks. The method includes sub-sampling a received image, which is a digital representation possible recipient(s) of the miniature security marks, to generate a reduced-resolution image of the received image. Maximum/minimum points detection is performed and the maximum/minimum points are grouped into one or more clusters according to location distances between the maximum/minimum points. Group configuration is checked to match the clusters with a pre-defined template configuration. Dot parameter verification is then performed to verify mark location and configuration between the received image and a pre-defined template dot specification.Type: GrantFiled: January 23, 2007Date of Patent: May 24, 2011Assignee: Xerox CorporationInventor: Zhigang Fan
-
Patent number: 7949195Abstract: A method, and apparatus, and logic encoded in one or more computer-readable tangible medium to carry out a method. The method is to code an ordered sequence of quantized transform coefficients of a block of image data using a hybrid coding method that includes determining a breakpoint location in the sequence between a low frequency region and a high frequency region; coding the low-frequency region using a selected low-frequency variable length coding method; coding the high-frequency region using a selected high-frequency variable length coding method; and coding the location of the breakpoint. The breakpoint location is determined as a function of the sequence such that different sequences can have different breakpoints.Type: GrantFiled: March 15, 2007Date of Patent: May 24, 2011Assignee: Cisco Technology, Inc.Inventors: Dihong Tian, Pi Sheng Chang, Wen-hsiung Chen
-
Patent number: 7945102Abstract: An apparatus for encoding data includes a receiving section, a signal-deteriorating factor generation section, and a data-encoding section. The signal-deteriorating generating section generates a signal-deteriorating factor and includes a phase-shifting section to shift a phase of data received at the receiving section. The data-encoding section obtains encoded data by performing encoding processing on the data in which the signal-deteriorating factor is generated and includes an encoding section to perform encoding by use of sub-sampling on the data whose phase is shifted by the phase-shifting section.Type: GrantFiled: March 23, 2004Date of Patent: May 17, 2011Assignee: Sony CorporationInventor: Tetsujiro Kondo
-
Patent number: 7925079Abstract: The present invention provides a method and system for pattern recognition and processing. Information representative of physical characteristics or representations of physical characteristics is transformed into a Fourier series in Fourier space within an input context of the physical characteristics that is encoded in time as delays corresponding to modulation of the Fourier series at corresponding frequencies. Associations are formed between Fourier series by filtering the Fourier series and by using a spectral similarity between the filtered Fourier series to determine the association based on Poissonian probability. The associated Fourier series are added to form strings of Fourier series. Each string is ordered by filtering it with multiple selected filters to form multiple time order formatted subset Fourier series, and by establishing the order through associations with one or more initially ordered strings to form an ordered string.Type: GrantFiled: December 23, 1998Date of Patent: April 12, 2011Inventor: Randell L. Mills
-
Patent number: 7920759Abstract: A Mixed Media Reality (MMR) system and associated techniques are disclosed. The MMR system provides mechanisms for forming a mixed media document that includes media of at least two types. In one embodiment of the present invention, the MMR system includes an action processor and method, and MMR documents with an associated action. The MMR document structure specifies different actions for different MMR documents combined with the ability to create any number of MMR documents for a particular location on any media, which allows the MMR architecture to serve as a universal trigger or initiator for additional processing. In another embodiment, the MMR system, in particular MMR recognition and the input/output capabilities of the capture device, are used as a control mechanism for a third party system.Type: GrantFiled: July 31, 2006Date of Patent: April 5, 2011Assignee: Ricoh Co. Ltd.Inventors: Jonathan J. Hull, Berna Erol, Jamey Graham, Peter E. Hart, Dar-Shyang Lee, Kurt Piersol
-
Patent number: 7920725Abstract: A characteristic amount calculating means calculates first characteristic amounts, which do not require normalization, and normalized second characteristic amounts. A first discriminating portion discriminates whether a candidate for a face is included in the target image, by referring to first reference data with the first characteristic amounts, calculated from the target image. The first reference data is obtained by learning the first characteristic amounts of a plurality of images, which are known either to be of faces or to not be of faces. In the case that the candidate is included, a second discriminating portion discriminates whether the candidate is a face, by referring to second reference data, obtained by learning the second characteristic amounts of a plurality of images, which known either to be of faces or to not to be of faces.Type: GrantFiled: September 8, 2004Date of Patent: April 5, 2011Assignee: FUJIFILM CorporationInventor: Yuanzhong Li
-
Patent number: 7912306Abstract: Coding quantized transform coefficients as occur in image compression combines a position identifying method to identify the relative position of clusters of consecutive non-zero-valued coefficients, an amplitude event identifying method to identify amplitude events in the clusters of non-zero-valued coefficients, and a coding method to code the position events, amplitude events, and signs of the amplitudes in the clusters. The method is particularly applicable to a series of quantized transform coefficients where clusters are likely to occur.Type: GrantFiled: November 25, 2009Date of Patent: March 22, 2011Assignee: Cisco Technology, Inc.Inventors: Wen-hsiung Chen, Fang Wu, John A. Toebes, Dov Rosenfeld, Gregory D. Pelton
-
Patent number: 7907794Abstract: A method for aligning a modified document and an original document is provided according to an aspect of the present invention. The method includes a step of receiving a first bitmap representative of the modified document, including a first anchor. Additionally, a second bitmap representative of the original document including a second anchor is received. The method also includes the step of deriving a set of first vertex coordinates of the first anchor, and a set of second vertex coordinates of the second anchor. The method further includes the step of transforming the first bitmap to a common reference based upon the first set of vertex coordinates, and the step of transforming the second bitmap to the common reference based upon the second set of vertex coordinates.Type: GrantFiled: January 24, 2007Date of Patent: March 15, 2011Assignee: Bluebeam Software, Inc.Inventors: Brian Hartmann, Benjamin Gunderson
-
Patent number: 7889938Abstract: In an image processing method, contour information is extracted on the basis of terminal points and intersections included in a thinned line drawing, the contour information being extracted for each of closed curves and line elements connecting the terminal points and the intersections included in the line drawing, and skeletonized vector data is generated on the basis of the contour information. Start and end terminal points in the skeletonized vector data are determined. Artificial vectors to be inserted between the determined start and end terminal points are generated, and artificial-vector-inserted vector data including the generated artificial vectors is generated. A smoothing process is performed for the artificial-vector-inserted vector data, and then smoothed, non-circumference vector data is generated on the basis of the start and end terminal points. The thinned line drawing is raster-scanned in units of a pixel matrix.Type: GrantFiled: March 21, 2007Date of Patent: February 15, 2011Assignee: Canon Kabushiki KaishaInventor: Yoshihiro Ishida
-
Patent number: 7885430Abstract: The present invention provides an automotive environment monitoring device to detect the existence of an object, a pedestrian or the like, with high accuracy even if one part of the object is overlapped with the background of an image obtained from an imaging unit mounted on the vehicle. An area where the object exists in the image obtained from an infrared camera (102) is set as a first area (A1i) and a lower portion of the first area (A1i) is set as a second area (A2i). A feature of the object is extracted with respect to the second area (A2i) and the object is identified based on the extracted feature. Thereby, even in the case where the first area (A1i) has been set but the feature of an upper part of the object can not be extracted due to the upper part of the object being overlapped with the background of the image, the object can still be identified with high accuracy by extracting the feature of a lower part of the object in the second area (A2i).Type: GrantFiled: April 7, 2008Date of Patent: February 8, 2011Assignee: Honda Motor Co., Ltd.Inventors: Nobuharu Nagaoka, Makoto Aimura, Masakazu Saka, Takayuki Tsuji