Patents Examined by Kate R Duffy
-
Patent number: 9704235Abstract: When the lengths of FEM wafers are automatically measured, not only the sizes of targets, the lengths of which are to be measured, are often varied from those in registration, but also the patterns of the targets are often deformed. Therefore, it is difficult to automatically determine whether the length measurement is possible or not. Therefore, the following are executed with a semiconductor inspection system: (1) a process of identifying the position of the contour line of an inspected image using a distance image calculated from a reference image, (2) a process of calculating a defect size image based on the position of the contour line with respect to the identified distance image, and detecting a defect candidate from the defect size image, and (3-1) a process of, upon detection of the defect candidate, calculating the size of the detected defect candidate, or (3-2) a process of detecting a portion different between the first and second contour lines as the defect candidate.Type: GrantFiled: July 20, 2012Date of Patent: July 11, 2017Assignee: HITACHI HIGH-TECHNOLOGIES CORPORATIONInventors: Akiyuki Sugiyama, Yuichi Abe
-
Patent number: 9697435Abstract: An object of the present invention is to reduce a size of a feature descriptor while maintaining accuracy of object identification.Type: GrantFiled: November 15, 2012Date of Patent: July 4, 2017Assignee: NEC CorporationInventors: Kota Iwamoto, Ryota Mase
-
Patent number: 9690986Abstract: A method for processing a physical document image. The method includes receiving a temporary quality control thumbnail file, a temporary metadata thumbnail file, and a high resolution archive file. Each of the temporary quality control thumbnail file, the temporary metadata thumbnail file, and the high resolution archive file is associated with a raw image captured by an imaging station. The method also includes performing a quality control process on the temporary quality control thumbnail file. The method further includes performing a metadata process on the temporary metadata thumbnail file. The method additionally includes performing an optical character recognition optimization process on the high resolution archive file. The method includes performing an image optimization process on the high resolution archive file.Type: GrantFiled: March 17, 2014Date of Patent: June 27, 2017Assignee: ArcaSearch CorporationInventors: James R. Walker, Calvin G. Sixta
-
Patent number: 9672590Abstract: An image interpolation method and device based on an autoregressive model. The method first, interpolating a low-resolution image up to a target scale to obtain an interpolated image M; determining a local area W in the image M to be interpolated, establishing two autoregressive models for each pixel point in the local area W except for the edge pixel points, and determining an initial objective function F0 according to the autoregressive models; down sampling the local area W except for the edge pixel points to the same size as the low-resolution image to obtain a local area W?, subtracting a corresponding area in the low-resolution image from W? one pixel value by one pixel value, and adding the result to the initial objective function F0 to obtain an objective function F; performing iteration on the objective function F to obtain a pixel point value of a center block of W.Type: GrantFiled: November 6, 2013Date of Patent: June 6, 2017Assignees: PEKING UNIVERSITY FOUNDER GROUP CO. LTD., BEIJING FOUNDER ELECTRONICS CO., LTD.Inventors: Mading Li, Jiaying Liu, Jie Ren, Zongming Guo
-
Patent number: 9672636Abstract: A particular implementation decomposes an image into a structure component and a texture component. An edge strength map is calculated for the structure component, and a texture strength map is calculated for the texture component. Using the edge strength and the texture strength, texture masking weights are calculated. The stronger the texture strength is, or the weaker the edge strength is, the more distortion can be tolerated by human eyes, and thus, the smaller the texture masking weight is. The local distortions are then weighted by the texture masking weights to generate an overall distortion level or an overall quality metric.Type: GrantFiled: April 23, 2012Date of Patent: June 6, 2017Assignee: Thomson LicensingInventors: Fan Zhang, Kai Xie, Wenfei Jiang, Zhibo Chen
-
Patent number: 9659224Abstract: Disclosed are techniques for merging optical character recognized (OCR'd) text from frames of image data. In some implementations, a device sends frames of image data to a server, where each frame includes at least a portion of a captured textual item. The server performs optical character recognition (OCR) on the image data of each frame. When OCR'd text from respective frames is returned to the device from the server, the device can perform matching operations on the text, for instance, using bounding boxes and/or edit distance processing. The device can merge any identified matches of OCR'd text from different frames. The device can then display the merged text with any corrections.Type: GrantFiled: March 31, 2014Date of Patent: May 23, 2017Assignee: Amazon Technologies, Inc.Inventors: Matthew Joseph Cole, Sonjeev Jahagirdar, Matthew Daniel Hart, David Paul Ramos, Ankur Datta, Utkarsh Prateek, Emilie Noelle McConville, Prashant Hegde, Avnish Sikka
-
Patent number: 9652857Abstract: An object detection apparatus acquires the value of an overlapping area which indicates a high probability that an object is located within a three dimensional space using data including a back projection of an object area on a three-dimensional space, acquires the integral quantity of the value of the overlapping area in a shape model including a three-dimensional target, and acquires the distribution degree of the value of the overlapping area in the shape model of the detection target. The apparatus further determines whether the inside of the shape model is an object or non-object using the integral quantity and the distribution degree.Type: GrantFiled: June 29, 2012Date of Patent: May 16, 2017Assignee: NEC CorporationInventor: Hiroo Ikeda
-
Patent number: 9633430Abstract: A method for analyzing fMRI brain data, comprising: collecting the fMRI data including spatial information and temporal information from subjects; preprocessing the fMRI data; computing independent components (ICs) and their corresponding time course for each individual subjects; constructing an initial functional connectivity pattern; constructing a classifier based on the functional connectivity pattern; and applying the classifier to functional connectivity patterns of individual subjects for statistical analysis or diagnosis. The method may be used in fMRI based studies of a brain function and brain disorder diagnosis.Type: GrantFiled: December 28, 2011Date of Patent: April 25, 2017Assignee: INSTITUTE OF AUTOMATION, CHINESE ACADEMY OF SCIENCESInventor: Yong Fan
-
Patent number: 9569866Abstract: Techniques involving flexible video object boundary tracking are described. One or more curves, such as Bezier curves, are received as drawn by a user on an initial frame of video to define a boundary of an object in the frame. The curves are then mapped to a subsequent or previous frame of the video where the object is included but has a new or changed boundary. A segmentation boundary is determined for the object in the subsequent frame and endpoints of segments of the curves are snapped to the segmentation boundary. Additionally, confidence values are determined for subregions of the frame that include portions of the curves. These confidence values are used to update control points on the curve segments to fit the curve segments to the new or changed boundary of the object in the frame.Type: GrantFiled: March 10, 2015Date of Patent: February 14, 2017Assignee: Adobe Systems IncorporatedInventors: Xue Bai, Jue Wang, Yao Lu
-
Patent number: 9547807Abstract: A method for classifying objects from one or more images comprising generating a trained classification process and using the trained classification process to classify objects in the images. Generating the trained classification process can include extracting features from one or more training images and clustering the features into one or more groups of features termed visual words; storing data for each of the visual words, including color and texture information, as descriptor vectors; and generating a vocabulary tree to store clusters of visual words with common characteristics. Using the trained classification process to classify objects can include extracting features from the images and clustering the features into groups of features termed visual words; searching the vocabulary tree to determine the closest matching clusters of visual words; and classifying objects based on the closest matching clusters of visual words in the vocabulary tree.Type: GrantFiled: October 19, 2012Date of Patent: January 17, 2017Assignee: The Univeristy of SydneyInventors: Teresa Vidal Calleja, Rishi Ramakrishnan
-
Patent number: 9542625Abstract: The present disclosure discloses device and method for detecting objects placed at multiple ranges from vehicle. Images of the objects may be captured by an image capturing unit housed in the vehicle. The image may be splitted into plurality of sub-images. Further, one or more features may be extracted from the plurality of sub-images. Further, each of the plurality of sub-images may be simultaneously processed for computing gradients associated with the plurality of sub-images. Further, a cell histogram may be created by casting weighted vote for an orientation based histogram channel based on values associated with the gradient. The gradients computed may be normalized by grouping the cells in spatial blocks. Further, a Support vector Machine (SVM) linear classifier may be applied on the plurality of sub-images in order to classify the near object and the far object in a category of a pedestrian or a vehicle.Type: GrantFiled: March 9, 2015Date of Patent: January 10, 2017Assignee: Tata Consultancy Services LimitedInventors: Manoj C. R., Thiyagarajan Paramasivam
-
Patent number: 9530202Abstract: An inspection apparatus includes an optical image acquisition unit to acquire an optical image of a photomask on which a plurality of figure patterns are formed, a first measurement unit to measure a first positional deviation amount in the horizontal direction at each position on the photomask accompanying deflection of the surface of the photomask generated by supporting the photomask using a support method which is used for acquiring the optical image, a second measurement unit to measure a second positional deviation amount of each of the plurality of figure patterns, by using the optical image, and a difference map generation unit to generate a difference map in which a difference value obtained by subtracting the first positional deviation amount from the second positional deviation amount is used as a map value, with respect to a region on the surface of the photomask.Type: GrantFiled: January 28, 2015Date of Patent: December 27, 2016Assignee: NuFlare Technology, Inc.Inventor: Nobutaka Kikuiri
-
Patent number: 9495740Abstract: A mask inspection apparatus including, a driving unit configured to drive a stage holding an inspection target mask, in which a pattern is formed, or a calibration mask, a light irradiation device configured to irradiate light on the inspection target mask or the calibration mask, an image sensor configured to detect a light quantity signal of transmitted light or reflected light of the inspection target mask or the calibration mask at a plurality of pixels. A sensor amplifier configured to amplify an output of the image sensor with respect to each pixel, generates an optical image, and normalizes a gain and an offset of signal amplitude, wherein at a first setting the sensor amplifier sets the gain and the offset using the calibration mask, and at a second setting the sensor amplifier sets the gain and offset of the inspection target mask based on the first setting.Type: GrantFiled: January 27, 2015Date of Patent: November 15, 2016Assignee: NuFlare Technology, Inc.Inventor: Toshiyuki Watanabe
-
Patent number: 9432675Abstract: An image decoding method, according to the present invention, includes the steps of: deriving an MPM candidate mode from neighboring blocks adjacent to a target block to be decoded; generating an MPM list using the MPM candidate mode derived from the neighboring blocks; and deriving an intra prediction mode for the target block to be decoded using the generated MPM list. According to the present invention, image compression efficiency can be improved.Type: GrantFiled: September 10, 2012Date of Patent: August 30, 2016Assignee: Electronics and Telecommunications Research InstituteInventors: Jin Ho Lee, Hui Yong Kim, Sung Chang Lim, Jin Soo Choi, Jin Woong Kim
-
Patent number: 9418438Abstract: Systems, devices and methods are described including receiving a source image having a foreground portion and a background portion, where the background portion includes image content of a three-dimensional (3D) environment. A camera pose of the source image may be determined by comparing features of the source image to image features of target images of the 3D environment and using the camera pose to segment the foreground portion from the background portion may generate a segmented source image. The resulting segmented source image and the associated camera pose may be stored in a networked database. The camera pose and segmented source image may be used to provide a simulation of the foreground portion in a virtual 3D environment.Type: GrantFiled: September 12, 2011Date of Patent: August 16, 2016Assignee: INTEL CORPORATIONInventors: Joshua Ratcliff, Yi Wu, Maha El Choubassi, Yoram Gat, Wei Sun, Kalpana Seshadrinathan, Igor V. Kozintsev
-
Patent number: 9412163Abstract: A method for detecting a cerebral infarct includes receiving an image of a brain of a subject from a magnetic resonance imaging scanner, wherein the image has a plurality of voxels, and each of the voxels has a voxel intensity. Then, the voxel intensities are normalized, wherein the normalized voxel intensities have a distribution peak, and the normalized voxel intensity of the distribution peak is Ipeak. A threshold is determined, which is the Ipeak+ a value. Voxel having the normalized voxel intensity larger than the threshold is selected, wherein the selected voxel is the cerebral infarct. A method for quantifying the cerebral infarct is also provided.Type: GrantFiled: March 10, 2015Date of Patent: August 9, 2016Assignees: National Central UniversityInventors: Syu-Jyun Peng, Jang-Zern Tsai, Yu-Wei Chen, Kuo-Wei Wang
-
Patent number: 9411042Abstract: Multi-sensor compressive imaging systems can include an imaging component (such an RF, microwave, or mmW metamaterial surface antenna) and an auxiliary sensing component (such as an EO/IR sensor). In some approaches, the auxiliary sensing component includes a structured light sensor configured to identify the location or posture of an imaging target within a field of view of the imaging component. In some approaches, a reconstructed RF, microwave, or mmW image may be combined with a visual image of a region of interest to provide a multi-spectral representation of the region of interest.Type: GrantFiled: October 10, 2014Date of Patent: August 9, 2016Assignee: Duke UniversityInventors: David Brady, Tom Driscoll, John Hunt, Daniel Marks, Alexander Mrozack, Matthew Reynolds, David R. Smith
-
Patent number: 9404866Abstract: A fluorescence observation device includes a light source supplying excitation light, an interlaced imaging device, a field memory storing an image output from the imaging device, and a difference computing unit generating a difference image. The light source supplies the excitation light such that one of the first and second field image acquiring periods in the imaging device is a fluorescence image acquiring period, and the other is a background image acquiring period. The difference computing unit switches between a first mode of subtracting a background image of the memory from a fluorescence image output from the imaging device, and a second mode of subtracting a background image output from the imaging device from a fluorescence image of the memory, to apply the mode. Accordingly, it is possible to generate images in which fluorescence is extracted in real time.Type: GrantFiled: July 13, 2012Date of Patent: August 2, 2016Assignee: HAMAMATSU PHOTONICS K.K.Inventor: Takayuki Sato
-
Patent number: 9390523Abstract: Z-effective (e.g., atomic number) values are generated for one or more sets of voxels in a CT density image using sparse (measured) multi-energy projection data. Voxels in the CT density image are assigned a starting z-effective value, causing a CT z-effective image to be generated from the CT density image. The accuracy of the assigned z-effective values is tested by forward projecting the CT z-effective image to generate synthetic multi-energy projection data and comparing the synthetic multi-energy projection data to the sparse multi-energy projection data. When the measure of similarity between the synthetic data and the sparse data is low, the z-effective value assigned to one or more voxels is modified until the measure of similarity is above a specified threshold (e.g., with an associated confidence score), at which point the z-effective values substantially reflect the z-effective values that would be obtained using a (more expensive) dual-energy CT imaging modality.Type: GrantFiled: May 1, 2012Date of Patent: July 12, 2016Assignee: ANALOGIC CORPORATIONInventors: Andrew Litvin, Ram Naidu, David Lieblich, Sergey Simanovsky
-
Patent number: 9390509Abstract: Provided is a medical image processing device capable of notifying the diagnosis personnel that a segmentation error has occurred or may have occurred during tissue segmentation processing. This medical image processing device specifies a gray matter image of a subject, smoothes the gray matter image, and, in accordance with an elevation function for calculating an absolute Z score, calculates an elevation value. Next, the medical image processing device compares the evaluation value with a pre-defined threshold value and determines the segmentation result, and, if the separation result is determined to be abnormal, warns that segmentation result is abnormal and displays a segmentation result display screen showing the segmentation result.Type: GrantFiled: September 19, 2012Date of Patent: July 12, 2016Assignee: DAI NIPPON PRINTING CO., LTD.Inventors: Tetsutaro Ono, Tomoaki Goto