Patents Examined by Brian Q Le
-
Patent number: 7486825Abstract: An image processing apparatus for detecting, at high precision, the contours of the eyes and the lips from an image including a shot of the face is provided. The apparatus includes an image input portion used to input an image, a feature point detecting portion that detects plural feature points of the object from the input image, a facial pose estimating portion that estimates a rotation angle indicating a facial pose from the detected, plural feature points, an initial parameter correcting portion that estimates initial parameters to detect the contours of the eye and the lips from the detected, plural feature points, and corrects the estimated initial parameters using the estimated rotation angle, an initial contour generating portion that generates an initial contour of the object using the corrected initial parameters, and a contour generating portion that generates a final contour from the initial contour.Type: GrantFiled: May 26, 2005Date of Patent: February 3, 2009Assignee: Kabushiki Kaisha ToshibaInventor: Mayumi Yuasa
-
Patent number: 7487438Abstract: Methods and apparatus for comparing blank forms represented in a digital format to digitized filled-in forms are described. Different errors are attributed different weights when attempting to correlate regions of blank and filled-in forms. Foreground pixels in the blank form which are not found in a corresponding portion of a filled-in form are attributed greater error significance than foreground pixels, e.g., pixels which may correspond to added text, found in the filled-in form which correspond to a background pixel value in the blank form. A virtual filled-in form including content, e.g., pixel values, from the filled-in form is generated from the content of the filled-in form and pixel value location mapping information determined from comparing the blank and filled-in forms. Various analysis is performed on a block basis, but in some embodiments the final pixel mapping to the virtual form is performed on a pixel by pixel rather than a block basis.Type: GrantFiled: March 8, 2005Date of Patent: February 3, 2009Assignee: Pegasus Imaging CorporationInventor: William Douglas Withers
-
Patent number: 7483572Abstract: A method determines a warping function between a first image and a second image. For each pixel in each scan-line of the first image, a distance matrix is constructed which represents distances between the pixel and a plurality of corresponding pixels of a second image. The distance matrices are ordered according a scan-line order of the pixels.Type: GrantFiled: August 25, 2004Date of Patent: January 27, 2009Assignee: Mitsubishi Electric Research Laboratories, Inc.Inventor: Fatih M. Porikli
-
Patent number: 7483576Abstract: A method for decoding a compressed video file comprises receiving the encoded video file, wherein the encoded video file includes a plurality of encoded video data tables and a plurality of reference pixel value sets. The plurality of encoded video data tables are decoded using the plurality of reference pixel value sets. Decoded video data is returned.Type: GrantFiled: February 2, 2004Date of Patent: January 27, 2009Assignee: Adams Platform Pty Ltd.Inventor: Adam Leslie Clark
-
Patent number: 7480412Abstract: A method and apparatus for characterizing an image. The method selects one or more toboggan potentials from the image, or a portion thereof, to be tobogganed. It toboggans the selected toboggan potentials to generate one or more toboggan parameters, forming at least one toboggan cluster using one or more of the toboggan parameters. It also selects one or more of the toboggan clusters to compute at least one feature parameter to characterize the image or a portion thereof.Type: GrantFiled: December 7, 2004Date of Patent: January 20, 2009Assignee: Siemens Medical Solutions USA, Inc.Inventors: Jianming Liang, Luca Bogoni
-
Patent number: 7474800Abstract: A method and apparatus from removing image compression artifacts includes comparing a center pixel value with a perimeter pixel value to generate a compare pixel value. The compare pixel value is compared with a threshold value such that when the compare pixel value is below a threshold value, a count value is incremented and an accumulation value is accumulated. The method and apparatus includes repeating the comparison of perimeter pixel values with the center pixel values. The method and apparatus further includes if the count value has been incremented, generating an output center pixel value based on the quotient of the accumulation value divided by the count value as the center pixel value.Type: GrantFiled: January 29, 2004Date of Patent: January 6, 2009Assignee: ATI Technologies ULCInventor: Konstantin E. Moskvitin
-
Patent number: 7471843Abstract: A system for improving the an image displayed on a display.Type: GrantFiled: June 15, 2004Date of Patent: December 30, 2008Assignee: Sharp Laboratories of America, Inc.Inventors: Dean Messing, Louis J. Kerofsky
-
Patent number: 7471831Abstract: The present invention relates to a system and method for performing rapid and programmable analysis of data. The present invention relates to a reconfigurable detector comprising at least one array of a plurality of pixels, where each of the plurality of pixels can be selected to receive and read-out an input. The pixel array is divided into at least one pixel group for conducting a common predefined analysis. Each of the pixels has a programmable circuitry programmed with a dynamically configurable user-defined function to modify the input. The present detector also comprises a summing circuit designed to sum the modified input.Type: GrantFiled: January 16, 2004Date of Patent: December 30, 2008Assignee: California Institute of TechnologyInventors: Greg Bearman, Michael J. Pelletier, Suresh Seshadri, Bedabrata Pain
-
Patent number: 7469056Abstract: A system for performing automated cell screening in drug discovery, including an automated microscope, a fast autofocus device and a digital imaging system. Processes are implemented in software through which relevant cellular material is segmented and quantified with minimal user interaction. Improvements in the following areas: known methods for image processing are implemented in such a way that automated segmentation is achieved; sets of known measurements (pixel counting, etc.) are implemented as methods which demonstrate aspects of biology in a reliable fashion; components for automated positioning, focusing, imaging and processing of a multiplicity of samples are integrated as systems within which the segmentation and measurement methods may be mounted; and components and methods are adapted into systems which yield more highly automated and more rapid cell screening.Type: GrantFiled: May 9, 2003Date of Patent: December 23, 2008Assignee: GE Healthcare Niagara Inc.Inventors: Peter Ramm, Yuriy Alexandrov, Jurich Cybuch, Paul Donders, Carlos Zarate, Bohdan J. Soltys
-
Patent number: 7469068Abstract: The invention is directed to a method and apparatus for serially receiving and dimensionally transforming an image in an ordered sequence of interleaved blocks. Each block comprises an ordered sequence of samples of the image, and the samples of a single block all corresponding to a single color component. The method includes steps of counting the samples in the order received to determine a sequence number for each sample, comparing the sequence number with a set of predetermined values; and dimensionally transforming the image by either discarding the sample or storing the sample in a memory depending on the result of the comparison.Type: GrantFiled: May 27, 2004Date of Patent: December 23, 2008Assignee: Seiko Epson CorporationInventors: Eric Jeffrey, Barinder Singh Rai
-
Patent number: 7463777Abstract: A method for the scaling down of data is provided. At least two blocks of transformed data samples representing at least two blocks of original data samples are received. One of at least two tables of constants is selected wherein each table of constants is capable of reducing the number of transformed data samples by a different factor. The constants taken from the selected table are applied to the at least two blocks of transformed data samples to produce one block of transformed data samples representing one block of final data samples. The data is processed one dimension at a time by multiplying the data in one dimension with selected constants taken from previously developed tables corresponding to the desired scale down factor. Scaling down by different factors in each dimension as well as scaling down in one dimension and scaling up in the other dimension may be achieved.Type: GrantFiled: November 4, 2004Date of Patent: December 9, 2008Assignee: International Business Machines CorporationInventors: Joan La Verne Mitchell, Timothy James Trenary, Nenad Rijavec, Ian Richard Finlay
-
Patent number: 7460729Abstract: A data transform processing apparatus comprising a first lossless transform circuit to perform two step ladder operation processings of receiving unweighted normalized data then outputting weighted nonnormalized rotation-transformed data, and a second lossless transform circuit to perform two step ladder operation processings of receiving the weighted nonnormalized rotation-transformed data from the first lossless transform circuit then performing inverse weighting and outputting unweighted normalized rotation-transformed data, wherein the outputs from the first lossless transform circuit are interchanged and supplied to the second lossless transform circuit.Type: GrantFiled: June 21, 2004Date of Patent: December 2, 2008Assignee: Canon Kabushiki KaishaInventor: Tadayoshi Nakayama
-
Patent number: 7460688Abstract: A digital image sensor comprising an array of pixels and a processor is provided. The array of pixels comprises a current pixel in a first color plane that is configured to produce a current sensor value, a first plurality of pixels in the first color plane that is configured to produce a first plurality of sensor values, and a second plurality of pixels in the second color plane that is configured to produce a second plurality of sensor values. The processor is configured to generate a plurality of estimate values using the first plurality of sensor values and a plurality of intensity ratios associated with the second plurality of sensor values, and the processor is configured to determine whether the current pixel is defective using the plurality of estimate values and the current sensor value.Type: GrantFiled: December 9, 2004Date of Patent: December 2, 2008Assignee: Aptina Imaging CorporationInventors: John H. Stanback, Xuemei Zhang, Ramakrishna Kakarala, Bond Yu Pong Ying
-
Patent number: 7460698Abstract: An ultrasonic picture processing method comprising the steps of extracting contour information of a target from respective frame pictures of an ultrasonic moving image, dividing the extracted contour information into a plurality of regions at a preset interval, comparing the divided contour information parts with one another and making a predetermined determination of the ultrasonic moving pictures based on the comparison results.Type: GrantFiled: January 3, 2005Date of Patent: December 2, 2008Assignee: Kabushiki Kaisha ToshibaInventors: Hideki Yoshioka, Mutsumi Watanabe, Mayumi Yuasa, Masahide Nishiura
-
Patent number: 7457478Abstract: An image processing apparatus, an image processing system and an image processing method can correct the distortion of an image at a low cost and can generate a high quality image in real time. An image processing apparatus is provided which is an image processing apparatus for correcting an original image having distortion.Type: GrantFiled: August 18, 2003Date of Patent: November 25, 2008Assignee: Sony CorporationInventors: Koji Aoyama, Tohru Kurata, Atsushi Nose, Kimitaka Wada
-
Patent number: 7457438Abstract: Methods and apparatus, including computer program products, implementing and using techniques for camera pan vector estimation, are disclosed. A camera model is provided for representing motion vectors based on a plurality of parameters. A distribution of the motion vectors is generated. A cluster in the distribution of motion vectors is identified. A center of mass of the identified cluster is determined. The camera pan vector is provided based upon the determined center of mass.Type: GrantFiled: October 7, 2004Date of Patent: November 25, 2008Assignee: Genesis Microchip Inc.Inventor: Hari N. Nair
-
Patent number: 7454090Abstract: A method is provided for augmenting a set of image recordings in which, in addition to making image recordings (95), location data (175) is recorded for locations for which the user desires an, or a further, image recording. This desired-image-recording location data is subsequently used to retrieve one or more corresponding image recordings from a service system (40).Type: GrantFiled: February 21, 2001Date of Patent: November 18, 2008Assignee: Hewlett-Packard Development Company, L.P.Inventors: Lawrence Wilcock, Michael P. Spratt, Andrew Thomas
-
Patent number: 7450738Abstract: The invention relates to a method for detecting the contour of an obstacle in the surroundings of a moving vehicle by means of a sensor device that is preferably integrated in the lateral area of the vehicle. In previously known methods of this type, the sensor device usually emits a sensor signal to the obstacle already when said obstacle lies ahead in the direction of travel while also emitting a sensor signal once the vehicle has passed the obstacle. In order to improve evaluation of the sensor signals with regard to the reflection thereof on the obstacle, the reflection signals thus generated are first mathematically averaged so as to be able to then derive in a more precise manner the actual position and the actual shape of the contour of the obstacle from the obtained averaged reflection signal.Type: GrantFiled: July 8, 2004Date of Patent: November 11, 2008Assignee: Valeo Schalter und Sensoren GmbHInventors: Nicolas Jecker, Heinrich Gotzig
-
Patent number: 7450778Abstract: A method performed by a processing system is provided. The method comprises performing a first set of correlations between a first plurality of pixel values in a first frame and a target pixel value in a target frame of a digital video across all of a plurality of color channels of the target pixel value, performing a second set of correlations between the first plurality of pixel values in the first frame and the target pixel value in the target frame across less than all of the plurality of color channels of the target pixel value, and determining whether the target pixel value is an artifact using the first set of correlations and the second set of correlations.Type: GrantFiled: March 17, 2005Date of Patent: November 11, 2008Assignee: Hewlett-Packard Development Company, L.P.Inventor: D. Amnon Silverstein
-
Patent number: 7447344Abstract: A system and method for visualizing a tree structure in medical image data is provided. The method comprises: segmenting the tree structure in the image data (230-265); coloring an exterior of the tree structure using data associated with interior components of the tree structure (270-290); and outputting an image of the tree structure colored by the interior components of the tree structure (295).Type: GrantFiled: April 11, 2005Date of Patent: November 4, 2008Assignee: Siemens Medical Solutions USA, Inc.Inventors: Eric Pichon, Carol L. Novak, Atilla Peter Kiraly