Patents Examined by Daniel Zeilberger
-
Patent number: 7916971Abstract: An image processing technique includes acquiring a main image of a scene and determining one or more facial regions in the main image. The facial regions are analysed to determine if any of the facial regions includes a defect. A sequence of relatively low resolution images nominally of the same scene is also acquired. One or more sets of low resolution facial regions in the sequence of low resolution images are determined and analysed for defects. Defect free facial regions of a set are combined to provide a high quality defect free facial region. At least a portion of any defective facial regions of the main image are corrected with image information from a corresponding high quality defect free facial region.Type: GrantFiled: May 24, 2007Date of Patent: March 29, 2011Assignee: Tessera Technologies Ireland LimitedInventors: Petronel Bigioi, Peter Corcoran, Eran Steinberg
-
Patent number: 7916933Abstract: A method of detecting objects in water, comprises the steps of capturing a plurality of images of a region of interest, extracting voxel data from the images, and processing the voxel data to detect items of interest in the region of interest. An apparatus that performs the method is also included.Type: GrantFiled: January 18, 2007Date of Patent: March 29, 2011Assignee: Northrop Grumman Systems CorporationInventors: Steven Allen Schwartz, Kenneth J. Wallenstein, David S. Lee
-
Patent number: 7912284Abstract: A method of filtering video packets for video stream monitoring is provided. A video packet of a video stream is received. One or more features are extracted from a specified frame of the video packet via one or more histograms and frequency domain coefficients of the specified frame. One or more concept detectors are implemented on the one or more features creating one or more confidence values. The confidence values are transmitted to a display module for filtering of video packets.Type: GrantFiled: May 13, 2008Date of Patent: March 22, 2011Assignee: International Business Machines CorporationInventors: Lisa Dronette Amini, Ching-Yung Lin, Olivier Verscheure
-
Patent number: 7889892Abstract: An index representing the probability that a fraction image is a face image including a face in an input image is calculated for each of the positions of the face to be detected on the basis of a feature value. When the sum of the indexes of the fraction images is not smaller than the first threshold value, the image formed by the fraction images is determined to be a face image.Type: GrantFiled: October 13, 2006Date of Patent: February 15, 2011Assignee: Fujifilm CorporationInventor: Kensuke Terakawa
-
Patent number: 7873228Abstract: A method identifying a ligature within a scanned document, the ligature including two or more touching characters. The two or more touching characters of the ligature are then compared to a plurality of prototypes to identify two or more matched prototypes. A synthetic ligature is then created based on the two or more matched prototypes.Type: GrantFiled: February 21, 2006Date of Patent: January 18, 2011Assignee: Adobe Systems IncorporatedInventors: Dennis G. Nicholson, Robert Filippini
-
Patent number: 7865018Abstract: Handwriting recognition techniques employing a personalized handwriting recognition engine. The recognition techniques use examples of an individual's previous writing style to help recognize new pen input from that individual. The techniques also employ a shape trainer to select samples of an individual's handwriting that accurately represent the individual's writing style, for use as prototypes to recognize subsequent handwriting from the individual. The techniques also alternately or additionally employ an intelligent combiner to combine the recognition results from the personalized recognition engine and the conventional recognition engine (or engines). The combiner may use a comparative neural network to combine the recognition results from multiple recognition engines. The combiner alternately may use a rule-based system based on prior knowledge of different recognition engines.Type: GrantFiled: June 10, 2005Date of Patent: January 4, 2011Assignee: Microsoft CorporationInventors: Ahmad A. Abdulkader, Ioannis A. Drakopoulos, Qi Zhang
-
Patent number: 7860326Abstract: The present invention improves the encoding efficiency, image quality, resolution, and the like of an input image having a special property, for example, an image input from a special camera. The input image is sliced according to a recognized shape and divided into blocks having a size according to a pixel density or an image resolution. Then, an object in the image is defined using a parameter including the numbers of the sliced images and the amounts of shift of the sliced images from a reference position, and the respective blocks are encoded using the movement of the object. Further, when a partial image is cut out from a base image, encoded, and then transmitted, a processing load in the encoding processing is reduced by encoding the partial image making use of preprocessing data.Type: GrantFiled: September 16, 2004Date of Patent: December 28, 2010Assignee: KDDI CorporationInventors: Yasuhiro Takishima, Satoshi Miyaji
-
Patent number: 7848572Abstract: An image processing method according to the present invention includes extracting from a document image an area to be determined, calculating the number of closed loops within the extracted area, and making a determination based on the calculated number of closed loops, whether the area is a character area. This invention makes it possible to determine with a high accuracy whether an area to be determined is a character area.Type: GrantFiled: April 19, 2006Date of Patent: December 7, 2010Assignee: Canon Kabushiki KaishaInventor: Reiji Misawa
-
Patent number: 7844103Abstract: An automated optical inspection system includes a pulsed light source illuminating an article to be inspected thereby to generate at least one image thereof, at least one camera having a field of view, and a relative motion provider operative to provide relative motion between the camera and at least one image of at least a portion of the article. The relative motion provider may include a first continuous motion provider and a second, velocity-during-imaging-lessening motion provider. The relative motion is a superposition of a first continuous component of motion provided by the first motion provider and a second, smaller component of motion provided by the second motion provider which lessens the velocity of the at least one image relative to the camera, during imaging.Type: GrantFiled: October 12, 2006Date of Patent: November 30, 2010Assignee: Applied Materials Israel, Ltd.Inventor: Ehud Tirosh
-
Patent number: 7835541Abstract: An apparatus, method, and medium for detecting a face in an image using classifiers generated by a boost algorithm are provided. The apparatus may include an image dividing unit dividing a black and white image into sub-windows; a sub-window detection unit detecting a sub-window corresponding to a face image in the divided sub-windows using the front part of the cascade of the classifiers and storing the sub-window detected as the face image; and a post-processing unit verifying whether each of sub-windows included in a group among a plurality of groups into which stored sub-windows are grouped is the face image using the number of child detections in each group and the latter part of the cascade of classifiers, integrating the sub-windows in the group according to the result of the verification, and outputting the result of the integration as the face image included in the black and white image.Type: GrantFiled: July 1, 2005Date of Patent: November 16, 2010Assignee: Samsung Electronics Co., Ltd.Inventors: Sujin Lee, Jungbae Kim
-
Patent number: 7835546Abstract: The present invention allows a device to easily generate pseudorandom numbers using information about a living body. An imaging element outputs an image signal equivalent to imaging light reflecting a target of identification inside a certain part of a living body. The device extracts data from the image signal output from the imaging element. This data reflects a change arising from the living body and a change of the imaging element. The device then generates a pseudorandom number using the data as a seed. Specifically, to produce the seed, the device sequentially selects two image signals output from the imaging element as a pair, sequentially detects difference between the pair of image signals, recognizes one of the detected differences as criterion, and then calculates a degree of difference between the criterion and the rest of differences.Type: GrantFiled: April 21, 2006Date of Patent: November 16, 2010Assignee: Sony CorporationInventor: Hiroshi Abe
-
Patent number: 7822265Abstract: An image processing system for performing a transformation of an input image associated with an input viewpoint to an output image associated with an output viewpoint. The input image is a pre-filtered 2D representation of 3D objects as seen from the input viewpoint, and comprises for each input pixel an associated input pixel value and an associated input pixel depth. Additional to the input image a hidden image is received, being another 2D representation of the 3D objects and comprising information, which information is occluded from the input viewpoint. The system comprises a video processor being operative to create the output image by transforming each input pixel to a transformed input pixel. The transformation is a function of the input pixel depth. The output image is created, based on the transformed input pixels, using hidden image pixels for filling de-occluded areas and for at least one pixel position adjacent to the de-occluded areas.Type: GrantFiled: April 8, 2005Date of Patent: October 26, 2010Assignee: Koninklijke Philips Electronics N.V.Inventor: Robert-Paul Mario Berretty
-
Patent number: 7817847Abstract: A robot system having a vision sensor. The robot system includes a robot; a robot controlling section for controlling an operation of the robot; an imaging section provided on the robot and obtaining image data of a working environment of the robot; an image processing section for processing the image data obtained in the imaging section; a vision controlling section for controlling the imaging section and the image processing section to cause execution of obtaining the image data, transmitting the image data thus obtained, and processing the image data; and a communication network to which the robot controlling section, the image processing section and the vision controlling section are connected.Type: GrantFiled: October 5, 2005Date of Patent: October 19, 2010Assignee: Fanuc LtdInventors: Yoshiki Hashimoto, Minoru Enomoto
-
Patent number: 7813574Abstract: A method and apparatus is disclosed herein for decoding data (e.g., video data) using transforms. In one embodiment, the decoding process comprises scaling a block of coefficients using a scaling factor determined for each coefficient by computing an index for said each coefficient and indexing a look-up table (LUT) using the index. The index is based on a quantization parameter, a size of the block of coefficients, and a position of said each coefficient within the block. The method also comprises applying a transform to the block of scaled coefficients.Type: GrantFiled: February 15, 2008Date of Patent: October 12, 2010Assignee: NTT DoCoMo, Inc.Inventor: Frank Jan Bossen
-
Patent number: 7760950Abstract: A method and apparatus is disclosed herein for decoding data (e.g., video data) using transforms. In one embodiment, the decoding process comprises scaling a block of coefficients using a scaling factor determined for each coefficient by computing an index for said each coefficient and indexing a look-up table (LUT) using the index. The index is based on a quantization parameter, a size of the block of coefficients, and a position of said each coefficient within the block. The method also comprises applying a transform to the block of scaled coefficients.Type: GrantFiled: September 24, 2003Date of Patent: July 20, 2010Assignee: NTT DoCoMo, Inc.Inventor: Frank Jan Bossen
-
Patent number: 7760930Abstract: The present disclosure provides a system and method for recognizing a defect image associated with a semiconductor substrate. In one example, the method includes collecting defect data of the defect image by testing and measuring the semiconductor substrate, extracting a pattern from the defect data, normalizing a location, orientation, and size of the pattern, and identifying the pattern after the pattern is normalized.Type: GrantFiled: February 21, 2006Date of Patent: July 20, 2010Assignee: Taiwan Semiconductor Manufacturing Company, Ltd.Inventors: Chen-Ting Lin, Chih-Cheng Chou, Chih-Hung Wu, Chia-Hua Chang
-
Patent number: 7734074Abstract: A finger sensor may include a plurality of finger image sensing arrays for generating a respective plurality of finger image data sets based upon sliding finger movement over the finger image sensing arrays, and a processor cooperating with the finger image sensing arrays. The processor may determine finger movement based upon the finger image data sets, and generate a resampled finger image data set by resampling the finger image data sets based upon the determined finger movement. The processor may further deskew the finger image data sets when generating the resampled finger image data set.Type: GrantFiled: February 17, 2006Date of Patent: June 8, 2010Assignee: Authentec, Inc.Inventors: Dale R. Setlak, Barbara E. Cornett, Kuntal Sengupta
-
Patent number: 7724971Abstract: Image compression, decompression and motion detection methods are described. Two temporally adjacent frame images, a previous time frame and a current time frame, are compressed in round-off and averaging techniques. Next, according to the compressed data of two corresponding pixels of the two frame images, whether or not the pixel of the current time frame image is of a motion picture is detected. If the pixel is of a motion picture, the compressed pixel data of the previous time frame image is decompressed, and an overdrive process is performed on the decompressed pixel data and the original pixel data of the current time frame image to produce an overdrive output. If the pixel is not of a motion picture, an overdrive process is not performed.Type: GrantFiled: July 29, 2008Date of Patent: May 25, 2010Assignee: Chunghwa Picture Tubes, Ltd.Inventors: Juin-Ying Huang, Wen-Tse Tseng, Chien-Hsun Cheng, Shih-Sung Wen
-
Patent number: 7715616Abstract: A PC board inspecting method capable of detecting deviation of an IC component at high speed with small storage capacity utilizes inspection logic particular to a component to be inspected. The inspection logic includes a color condition for specifying a color appearing in the body of the component to be inspected. The component is radiated with a plurality of color rays at different angles of incidence. An image formed by reflection light of the color rays is captured, and an inspection image including at least part of an edge of the body of the component to be inspected is captured. By using the color condition, a component body area satisfying the color condition is extracted from the inspection image by an image process.Type: GrantFiled: February 21, 2006Date of Patent: May 11, 2010Assignee: OMRON CorporationInventors: Toshihiro Moriya, Hirotaka Wada, Takako Onishi, Atsushi Shimizu, Akira Nakajima
-
Patent number: 7715594Abstract: A method of iris recognition using a cumulative-sum-based change point analysis and an apparatus using the same are disclosed. The method includes: transforming an iris image to a polar coordinated image having n×m pixel size and dividing the polar coordinated image into at least one cell; grouping the divided iris images into at least one of first groups having a predetermined number of cells, and at least one of second groups each having more cells than the first group has; performing a cumulative-sum-based change point analysis using a predetermined characteristic pattern value of each cell as a representative value; and generating a pattern vector by assigning a predetermined value to a cell having the change point and assigning a different value to other cells.Type: GrantFiled: April 18, 2006Date of Patent: May 11, 2010Assignee: Electronics and Telecommunications Research IntituteInventors: Jong Gook Ko, Youn Hee Gil, Jang Hee Yoo, Kyo Il Chung