Abstract: In a visual inspection method and apparatus, a picture processing unit converts an original picture, obtained by taking a photograph of a BGA illuminated by a ring illuminator from above, by using a camera, and labels a binary picture obtained by this binary conversion. Then, it forms a circumscribing rectangle circumscribing an outer circumference of a labeling picture obtained by the labeling, and inverts a labeling picture within the formed rectangle, and removes a portion of a region formed by the outer circumference and the circumscribing rectangle in a picture obtained by the inversion, and then generates an inspection picture by adding a picture obtained by the removal to the labeling picture, and accordingly judges a pass or rejection of the inspection target sample based on the generated inspection picture. Thus, the inspection can be carried out at a high accuracy irrespectively of a low cost.
Abstract: An image retrieval device and an image retrieval method are provided which are capable of improving performance of image retrieval, of retrieving images at higher speed with simplified configurations and of retrieving images by simplified calculating processes.
Abstract: Embodiments described herein provide for a system for creating a data collection of recognized images. The system includes an image analysis module that is configured to programmatically analyze individual images in a collection of images in order to determine information about each image in the collection. The system may also include a manual interface that is configured to (i) interface with one or more human editors, and (ii) displays a plurality of panels concurrently. Individual panels may be provided for one or more analyzed images, and individual panels may be configured to display information that is at least indicative of the one or more images of that panel and/or of the information determined from the one or more images. Additionally, the manual interface enables the one or more human editors to view the plurality of panels concurrently and to interact with each of the plurality of panels in order to correct or remove any information that is incorrectly determined from the image of that panel.
Type:
Grant
Filed:
November 7, 2007
Date of Patent:
February 9, 2010
Assignee:
Like.com
Inventors:
Salih Burak Gokturk, Baris Sumengen, Diem Vu, Navneet Dalal, Danny Yang, Xiaofan Lin, Azhar Khan, Munjal Shah, Dragomir Anguelov, Lorenzo Torresani, Vincent Vanhoucke
Abstract: A digital-signal processing apparatus in which the efficiency of processing data can be prevented uniformly. The apparatus includes: an extracting unit that extracts a part of real data input; a selecting unit that selects a data size from similar data sizes falling in a range, in accordance with the size of the extracted data, which is a reference value, and with data-decomposing conditions described in an algorithm that realizes fast Fourier transform; and an operation unit that decomposes the data extracted and having the selected size into data items, in accordance with the data-decomposing conditions, and that performs Fourier transform on each data item, thereby effecting convolution on the data item.
Abstract: A watermarked halftone image is generated from a continuous tone image by performing error diffusion halftoning on the continuous tone image while adding a watermark to the halftone image during the halftoning. A location of the watermark in the halftone image may be based on prediction criteria. Error caused by adding the watermark may be diffused into the halftone image.
Type:
Grant
Filed:
July 11, 2004
Date of Patent:
February 9, 2010
Assignee:
Hewlett-Packard Development Company, L.P.
Abstract: Systems and methods using an unstructured pattern for three-dimensional model construction, such as may be used for processing images of documents are disclosed. The method generally includes projecting an unstructured infrared pattern onto a target object, capturing a pair of stereoscopic images of the pattern projected onto the target object, such as with infrared cameras, performing signal processing on the pair of stereoscopic images to cross-correlate portions of the stereoscopic images in order to identify corresponding portions in the stereoscopic images, and constructing a three-dimensional model of the target object. The three-dimensional model can be utilized to rectify, e.g., de-warp, an image of the target object, e.g., a page or both facing pages of a bound document.
Abstract: In accordance with one aspect of the present exemplary embodiments, a calibration arrangement is configured to assist in calibration of a surface scanning system where the calibration arrangement includes a preconfigured physical object which may embody dimensional information wherein the dimensional information is used to calibrate a surface of the scanning system. In an alternative embodiment, the preconfigured physical object is configured to obtain data for use in calibration of the surface of a pan/tilt surface scanning system.
Type:
Grant
Filed:
December 20, 2004
Date of Patent:
February 2, 2010
Assignee:
Palo Alto Research Center Incorporated
Inventors:
Eric Saund, Bryan Pendleton, Kimon Roufas, Hadar Shemtov
Abstract: Embodiments enable searching of portions of objects in images, including programmatically analyzing each image in a collection in order to determine image data that, for individual images in the collection, represents one or more visual characteristics of a portion of an object shown in that image. A user is enabled to specify one or more search criteria that includes image data, and a search result may be determined based on one or more images in the collection that show a corresponding object that has a portion that satisfies a threshold. The threshold is defined at least in part by the one or more search criteria.
Type:
Grant
Filed:
November 7, 2007
Date of Patent:
February 2, 2010
Assignee:
Like.com
Inventors:
Salih Burak Gokturk, Baris Sumengen, Diem Vu, Navneet Dalal, Danny Yang, Xiaofan Lin, Azhar Khan, Munjal Shah, Dragomir Anguelov, Lorenzo Torresani, Vincent Vanhoucke
Abstract: Location information representing the location of a defective pixel in an image pickup device and pixel defect level information representing the pixel defect level of the defective pixel are stored beforehand in a memory. The defect level of the defective pixel is determined after shipment of the image pickup device, and the pixel defect information is updated based on the defect determination result. If a new defective pixel is detected at a location different from the location of the defective pixel with the pixel defect information stored in the memory, the pixel defect information of that defective pixel is added. If a defective pixel is detected at the same location as the defective pixel with the pixel defect information stored in the memory, and if the defect level of the detected defective pixel is worse than the original defect level, the defect level of the defective pixel is updated.
Abstract: A digital imaging device such as a digital camera is used to obtain multiple digital images of all of a text document segment. The text document segment could be the entire text document so that each digital image would be of the entire text document. The multiple images are obtained while lateral jittering is imparted between the digital imaging device and the text document. The lateral jittering may be in a pair of transverse (e.g., perpendicular) directions. An enhanced resolution representation of the text document is formed from the multiple laterally displaced images and de-blurred as a multi-frame reconstruction.
Abstract: Generation of ringing components during extraction of striped blue components from an image is suppressed, in an image processing apparatus for removing striped blur components from the image. A preprocess is executed on an image data set, which is obtained by detecting light with line sensors that perform scanning in a sub scanning direction. The preprocess reduces drastic changes in pixel values in a first direction, which is perpendicular to the sub scanning direction. A filter process is executed on the preprocessed image data set to extract image data that represents striped blur components that extend in the sub scanning direction. The extracted image data is subtracted from the image data set to remove striped blur components.
Abstract: Embodiments programmatically analyze each of a plurality of images in order to determine one or more visual characteristics about an item shown in each of the plurality of images. Data is stored corresponding to the one or more visual characteristics. An interface in is provided for which a user is able to specify one or more search criteria. In response to receiving the one or more search criteria, a search operation is performed to identify one or more items that have a visual characteristic that satisfies at least some of the one or more search criteria.
Type:
Grant
Filed:
November 7, 2007
Date of Patent:
February 2, 2010
Assignee:
Like.com
Inventors:
Salih Burak Gokturk, Baris Sumengen, Diem Vu, Navneet Dalal, Danny Yang, Xiaofan Lin, Azhar Khan, Munjal Shah, Dragomir Anguelov, Lorenzo Torresani, Vincent Vanhoucke
Abstract: In a digital multifunction peripheral that has a fingerprint reader capability, a display panel displays the duration of time from when the user places his finger immovably at the fingerprint read position until a carriage of the scanner actually starts reading the fingerprint. Consequently, the user becomes careful not to move his finger during the fingerprint scan or before or after the scan starts, and a rate of erroneous reading in fingerprint reading can be lowered.
Abstract: Visual tracking over a sequence of images is formulated by defining an object class and one or more background classes. The most discriminant features available in the images are then used to select a portion of each image as belonging to the object class. Fisher's linear discriminant method is used to project high-dimensional image data onto a lower-dimensional space, e.g., a line, and perform classification in the lower-dimensional space. The projection function is incrementally updated.
Abstract: The present invention can extract a change not only from an orthorectified image but also from a plurality of images captured at different camera positions, under different lighting conditions, at different angles, and at different scaling factors. Foot-print information is extracted from each of images to be compared, and a change is detected by comparing the foot-print information.
Abstract: A digital watermark detection apparatus includes an extraction unit extracting a specific frequency component signal from an input image signal, a first transformation unit orthogonal-transforming the image signal, a second transformation unit orthogonal-transforming the specific frequency component signal, a combination unit combining first and second transformed image signals from the first and second transformation units, a third transformation unit orthogonal-transforming a combined image signal, an estimation unit estimating embedded watermark information by searching for a peak appearing in a third transformed image signal of the third transformation unit, and a compressor compressing a target signal in amplitude based on a compression characteristic varying according to an amplitude and frequency of the target signal to obtain a compression signal, the target signal formed of at least one of the first transformed image signal, the second transformed image signal and the combined image signal.
Abstract: Method and apparatus for compressed sensing yields acceptable quality reconstructions of an object from reduced numbers of measurements. A component x of a signal or image is represented as a vector having m entries. Measurements y, comprising a vector with n entries, where n is less than m, are made. An approximate reconstruction of the m-vector x is made from y. Special measurement matrices allow measurements y=Ax+z, where y is the measured m-vector, x the desired n-vector and z an m-vector representing noise. “A” is an n by m matrix, i.e. an array with fewer rows than columns. “A” enables delivery of an approximate reconstruction, x#, of x. An embodiment discloses approximate reconstruction of x from the reduced-dimensionality measurement y. Given y, and the matrix A, x# of x is possible. This embodiment is driven by the goal of promoting the approximate sparsity of x#.
Abstract: A device for user enrollment is adapted to capture an image of a pattern projected on a target and distorted on the three-dimensional (3D) surface of the target, such as a human face, and an image of a two-dimensional (2D) surface of the target. The device can include an illumination unit adapted to project a patterned light onto the surface of the target in an invisible light wavelength, a 3D image capturing device to capture an image of a pattern projected on the target and distorted on the 3D surface of the target. The device can also include a 2D image capturing device adapted to capture an image of a 2D frontal view of the target in visible light. The device can also include an orientation unit that displays the face of the target in order to facilitate the positioning of the target within the proper fields of view.
Abstract: A method of detecting an image of a reflective object includes the steps of capturing an exposure image by an image sensor while an LED is ON; down-sampling an exposure image signal from the image sensor to a first digital signal by comparing the level of the first image signal with a first reference level; capturing a dark image by the image sensor while the LED is OFF; down-sampling a second image signal from the image sensor to a second digital signal by comparing the level of the second image signal with a second reference level; computing an image of difference between the exposure image and the dark image by subtracting the second digital signal from the first digital signal; and specifying the image of the reflective object by locating a bright point remaining in the image of difference.
Abstract: Processing of a digital image includes applying a function to a plurality of pixels in the image, the function relating a noise statistic to pixel intensity values. By applying the noise statistic function, a noise statistic value is produced for each of the pixels. Each of the pixels is processed with respect to its noise statistic value.
Type:
Grant
Filed:
July 5, 2005
Date of Patent:
January 5, 2010
Assignee:
Hewlett-Packard Development Company, L.P.