Electronic Template Patents (Class 382/217)
-
Patent number: 6621929Abstract: A method for matching images is provided that includes the step of providing a template image and an input image. The images include pixels, each pixel having an intensity associated therewith. An energy function formed by weighting a sum of squared differences of data constraints corresponding to locations of both the template image and the input image is minimized to determine estimated geometric and spatially-varying illumination-parameters for the input image with respect to the template image. The estimated geometric and spatially-varying illumination parameters are outputted for further processing.Type: GrantFiled: June 22, 1999Date of Patent: September 16, 2003Assignee: Siemens Corporate Research, Inc.Inventors: Shang-Hong Lai, Ming Fang
-
Publication number: 20030165276Abstract: A document image capture (scanning) system and control method are described for scanning and processing document images received live from a camera. A motion detector detects image motion between two image frames. When the image is stationary, image processing (such as OCR) is carried out automatically and made available to the operator. In one form, when movement is detected, the image processing results are discarded until the image is newly stationary, whereupon new image processing is carried out on the new image. In another form, the degree of movement is evaluated; if the movement is small, then at least some of the previous image processing results are re-used by re-mapping on to the new image.Type: ApplicationFiled: March 4, 2002Publication date: September 4, 2003Applicant: XEROX CORPORATIONInventors: Mauritius Seeger, Stuart A. Taylor, Christopher R. Dance
-
Patent number: 6614926Abstract: The invention provides methods and apparatuses for generating from an image of an object a model of the object, when the object contains similar features. First, the invention finds the similar features. The similar features are found by searching portions of the image for features that are substantially similar to a feature prototype. More particularly, individual features in the image are located and designated candidate features, and optionally a spatial pattern representing a majority of the candidate features is generated. Next, feature profiles are generated therefrom, and the feature prototype is constructed using at least a subset of feature profiles. Thereafter, the model is generated using the similar features and, optionally, the spatial pattern.Type: GrantFiled: June 28, 2000Date of Patent: September 2, 2003Assignee: Cognex CorporationInventors: Raymond Fix, Ivan Bachelder
-
Publication number: 20030152276Abstract: An image of an inspection object is picked up by an image pickup unit, and a characteristics quantity of an image of a defect part extracted by a defect extracting unit is extracted and digitized by a characteristics extracting unit. A database preparing unit regroups defects having similar characteristics by a defect classification unit on the basis of the characteristics information digitized by the characteristics extracting unit with respect to defects belonging to a defect group selected and designated by an operator via a display/input unit, and prepares on a database memory a database in which the defects of the inspection object are hierarchically classified.Type: ApplicationFiled: August 16, 2002Publication date: August 14, 2003Inventors: Hiroshi Kondo, Masatsuga Okuyama, Yo Saito
-
Patent number: 6603882Abstract: A fast multi-resolution template search method uses a manually selected or an automatically selected template set, learned application specific variability, and optimized image pre-processing to provide robust, accurate and fast alignment without fiducial marking. Template search is directed from low resolution and large area into high resolution and smaller area with each level of the multi-resolution image representation having its own automatically selected template location and pre-processing method. Measures of discrimination power for template selection and image pre-processing selection increase signal to noise and consistency during template search. Signal enhancement means for directing discrimination of optimum template location are taught.Type: GrantFiled: April 12, 2001Date of Patent: August 5, 2003Inventors: Seho Oh, Shih-Jong J. Lee
-
Patent number: 6591010Abstract: A method and a system employing the method is disclosed for detecting and qualifying a portion of a digital image stored in a computer-readable memory with respect to a reference digital image. The method and system accesses video memory of a video controller device, retrieves a portion of the rasterized image, and compares the portion of the image on a pixel-by-pixel basis for approximate match to the reference image. Because graphics accelerators in video controller devices render inexact graphic images which are nonetheless visually acceptable, the method and system further employs a step of comparison for a match on a pixel region basis.Type: GrantFiled: July 29, 1999Date of Patent: July 8, 2003Assignee: International Business Machines CorporationInventor: Robert Michael Russin
-
Publication number: 20030118249Abstract: A method, system and software are disclosed for correcting defects formed in a physical medium of an original image. Multiple scans of the original image are recorded, where the multiple scans have different properties. For example, the angle of the light incident to the physical medium or the properties of the light may be changed between scans of original image. The multiple scans can be used to generate a reference image from which defect corrections are made. Alternatively, a reference image can be generated directly from the original image. The multiple scans can also be used to determine the degree of defectiveness and/or an estimate of the signal strength of each portion of the original image. A decision is made on whether or not an image portion having one or more defects should be corrected, where the decision can be based on an evaluation of the potential benefit compared to the potential damage caused by correction of an image portion.Type: ApplicationFiled: April 19, 2002Publication date: June 26, 2003Inventor: Albert D. Edgar
-
Patent number: 6584224Abstract: A template is analyzed to determine a step size for searching within a search area. First, the template is padded with pixels to increase its size. Cross-correlation between the padded template and the original template leads to identification of an effective step size. Step sizes for each of horizontal and vertical axes are derived. Windows of the search area, selected based on the step size, then are tested in a fast search stage by correlating the template to the window. Any tested window which has a correlation coefficient exceeding a specific value is a local match. A full search of the vicinity of the local match then is performed for all potential windows within an area bounded by one step to either side of the local match along either axis. The location(s) corresponding to the highest correlation(s) exceeding the threshold value are matches.Type: GrantFiled: August 23, 2001Date of Patent: June 24, 2003Assignee: University of WashingtonInventors: Shijun Sun, HyunWook Park, Yongmin Kim
-
Publication number: 20030099397Abstract: There is disclosed an image extraction method for extracting, from a subject image that records both the background and the object to be extracted, image data of the object using a mask. An initial mask used for extracting the subject region is generated on the basis of difference data between the subject image and the background image that records the background alone, the region of the initial mask is grown on the basis of the similarity between the features of a first region and its neighboring second region in the subject image corresponding to the initial mask, and the image data of the object is extracted from the first image on the basis of the grown mask region.Type: ApplicationFiled: May 1, 2000Publication date: May 29, 2003Inventors: Masakazu Matsugu, Hideo Takiguchi, Koji Hatanaka
-
Publication number: 20030099396Abstract: An image processing apparatus for extracting a desired target object from an image which includes the target object, includes a target object identification part identifying the target object from the image, a rectangle extracting part extracting a rectangle which includes the target object identified by said target object identification part, and an image cutting part cutting out a rectangular image including the target object from the rectangle extracted from the image by said rectangle extracting part.Type: ApplicationFiled: July 8, 1999Publication date: May 29, 2003Inventors: YASUFUMI NAKAMURA, YASUHIRO KAWAKATSU
-
Patent number: 6567551Abstract: Image data and their image feature amounts are managed in correspondence with each other using an image storage unit and an information management DB. A desired one of query conditions each of which is defined by image feature amounts set with ambiguity levels is designated by an input unit. A matching unit matches the image feature amounts that define the designated query condition, with corresponding managed image data. Image data as search results are output to a display on the basis of the matching results.Type: GrantFiled: April 26, 1999Date of Patent: May 20, 2003Assignee: Canon Kabushiki KaishaInventor: Hirotaka Shiiyama
-
Publication number: 20030081857Abstract: A data processing apparatus is operable to determine non-linear distortion applied to an image in which template signals have been introduced into at least first and second frequency bands of the image respectively, the template signals having a predetermined relationship with each other. The apparatus comprises an image dividing processor operable to generate data representative of selected parts of the image, a transform detection processor operable to generate at least one parameter representative of a linear transformation applied to each of the parts of the image, and a non-linear transform processor operable to estimate parameters of a non-linear transformation representative of the non-linear distortion applied to the image from the linear transform parameters of at least two parts of the image.Type: ApplicationFiled: August 19, 2002Publication date: May 1, 2003Inventor: Daniel Warren Tapson
-
Publication number: 20030072481Abstract: A method for determining whether a defect has previously occurred by searching a database of defect spatial signatures for a signature that matches that of a newly inspected semiconductor wafer. If a match occurs, an engineer is notified of the match. The defect spatial signature of the newly inspected wafer is added to the database of defect spatial signatures.Type: ApplicationFiled: October 11, 2001Publication date: April 17, 2003Applicant: ADVANCED MICRO DEVICES, INC.Inventors: Christopher L. Wooten, Arturo Morosoff
-
Patent number: 6546137Abstract: A fast localization with advanced search hierarchy system for fast and accurate object localization in a large search space is based on an assumption that surrounding regions of a pattern within a search range are always fixed. The FLASH system comprises a hierarchical nearest-neighbor search system and an optical-flow based energy minimization system. The hierarchical nearest-neighbor search system produces rough estimates of the transformation parameters for the optical-flow based energy minimization system which provides very accurate estimation results and associated confidence measures.Type: GrantFiled: January 25, 1999Date of Patent: April 8, 2003Assignee: Siemens Corporate Research, Inc.Inventors: Shang-Hong Lai, Ming Fang
-
Patent number: 6546151Abstract: An information processing apparatus comprises an indication means placed on a manuscript; a reading unit for reading information on the manuscript; an indication means detection unit for detecting the indication means placed on the manuscript, thereby detecting a region indicated by the indication means on the manuscript; and an information processing unit performing a predetermined process to the information obtained from the region indicated by the detected indication means, amongst the information on the manuscript read by the reading unit. Therefore, by only placing the indication means on a target portion to be processed on the manuscript, the portion is detected by the indicated means detection unit to be subjected to a process such as scale-up by the information processing unit.Type: GrantFiled: January 20, 1999Date of Patent: April 8, 2003Assignee: Matsushita Electric Industrial Co., LtdInventors: Shoichi Araki, Kouji Miura, Satoshi Matsuura, Takeshi Imanaka, Hiroshi Kutsumi
-
Patent number: 6539116Abstract: The structure of entered document image data is analyzed and a character string in a text block that has been analyzed is subjected to pattern recognition. Synonyms and equivalents of words obtained as results of language analysis are extracted and words obtained as results of language analysis are converted to words of another language. A character string in a text block that has been analyzed is translated to another language. At least results of analyzing the structure of document image data, results of character recognition and results of language analysis are stored, and at least one of the results of extraction, results of conversion and results of translation are stored in a RAM in association with the results of character recognition.Type: GrantFiled: October 2, 1998Date of Patent: March 25, 2003Assignee: Canon Kabushiki KaishaInventor: Makoto Takaoka
-
Patent number: 6538759Abstract: An non-reproducible image formation preventing apparatus has a recognition section, a storage section and a comparison/determination section. The recognition section performs recognition processing for received image data, determines whether or not the image data is an non-reproducible image, and outputs the determination result to an image formation section. The storage section stores non-reproducible image data recognized as an non-reproducible image by the recognition section. The comparison/determination section compares the non-reproducible image data stored in the storage section with an image based on input image data given through an image process section and finding a match degree therebetween. If either of the comparison/determination section and the recognition section determines that the image is an non-reproducible image, a detection signal is sent to the image formation section for stopping normal print.Type: GrantFiled: June 4, 1999Date of Patent: March 25, 2003Assignee: Omron CorporationInventors: Junji Hiraishi, Kiyoshi Imai, Takashi Kakiuchi, Keitaro Taniguchi
-
Publication number: 20030039404Abstract: A method of aligning a first image with a second image, the method comprising the steps of; approximately locating the second image; scaling the first image to the approximate size of the second image; positioning the scaled first image over the approximate location of the second image; generating a measure of fit between the first image and the second image; re-positioning the second image and repeating the above steps to obtain an increased measure of fit value.Type: ApplicationFiled: September 17, 2002Publication date: February 27, 2003Inventor: Alexandre J Bourret
-
Publication number: 20030035583Abstract: A method of determining segments in a series of images based on previous segmentation results and on motion estimation. A second segment (108) of a second image (106) is determined based on a first segment (102) of a first image (100), with the first segment (102) and the second segment (108) corresponding to one object (104). The method comprises a calculation step to compare values of pixels of the first image (100) and the second image (106) in order to calculate a motion model which defines the transformation of the first segment (102) into the second segment (108). An example of such a motion model allows only translation. Translation can be described with one motion vector (110).Type: ApplicationFiled: May 14, 2002Publication date: February 20, 2003Inventors: Anna Pelagotti, Gerard Anton Lunter, Massimo Ciacci
-
Publication number: 20030002740Abstract: In one aspect the invention provides processes for locating a pattern within an image that can comprise the acts of providing a template representative of the pattern to be located within an image. The image can be subdivided into a plurality of sub-images, each being representative of a portion of the image. The process can then compare each of the sub-images to the template to generate a plurality of score signals representative of a location of the pattern, and can then process the score signals to determine a location for the pattern. In one practice, the step of processing the score signals can include a step of identifying at least one sub-image that includes a degraded image. As described above, a degraded image can include an image that has had a portion obscured by shadows or debris. Additionally, an image can be degraded at spots of high-reflection that create glare and wash-out portions of the image.Type: ApplicationFiled: May 23, 2002Publication date: January 2, 2003Inventors: Simon H. Melikian, Rohit G. Israni
-
Publication number: 20020181783Abstract: Image data and their image feature amounts are managed in correspondence with each other using an image storage unit and an information management DB. A desired one of query conditions each of which is defined by image feature amounts set with ambiguity levels is designated by an input unit. A matching unit matches the image feature amounts that define the designated query condition, with corresponding managed image data. Image data as search results are output to a display on the basis of the matching results.Type: ApplicationFiled: April 26, 1999Publication date: December 5, 2002Inventor: HIROTAKA SHIIYAMA
-
Patent number: 6490369Abstract: A robot manipulator having a vision system for viewing and identifying a known part. The vision system includes a computer and a camera with known viewing geometry and optics to view the part. The method of viewing and identifying the part comprises the following steps. First, a template of the known part is stored in the computer. Secondly, the relative position of the camera in relation to the part is determined and stored. Thirdly, the part is viewed through the camera and a distorted image of the part is captured. Fourthly, the stored relative position information of the camera in relation to the part and the known viewing geometry and optics of the camera are used to convert the distorted image into a true image of the part which removes any distortions from the distorted image. Fifthly, the true edges of the true image of the part are determined.Type: GrantFiled: July 6, 1999Date of Patent: December 3, 2002Assignee: Fanuc Robotics North AmericaInventor: Leonard H. Beiman
-
Patent number: 6477275Abstract: In one aspect the invention provides processes for locating a pattern within an image that can comprise the acts of providing a template representative of the pattern to be located within an image. The image can be subdivided into a plurality of sub-images, each being representative of a portion of the image. The process can then compare each of the sub-images to the template to generate a plurality of score signals representative of a location of the pattern, and can then process the score signals to determine a location for the pattern. In one practice, the step of processing the score signals can include a step of identifying at least one sub-image that includes a degraded image. As described above, a degraded image can include an image that has had a portion obscured by shadows or debris. Additionally, an image can be degraded at spots of high-reflection that create glare and wash-out portions of the image.Type: GrantFiled: June 16, 1999Date of Patent: November 5, 2002Assignee: Coreco Imaging, Inc.Inventors: Simon H. Melikian, Rohit G. Israni
-
Patent number: 6463176Abstract: An image recognition method comprising an extraction step of extracting local feature elements in an input image; a conversion step of arranging the local feature elements extracted at said extraction step on each of cells of predetermined lattice pattern composed of a plurality of cells having predetermined sizes and figures while substantially maintaining relative position relationship between a plurality of extracted local feature elements which are near each other; and a matching step of estimating a resemblance between a plurality of input images by matching the plurality of local feature elements arranged on the lattice pattern at said conversion step.Type: GrantFiled: July 15, 1997Date of Patent: October 8, 2002Assignee: Canon Kabushiki KaishaInventors: Masakazu Matsugu, Katsumi Iijima
-
Patent number: 6411731Abstract: Position, size and attribute regarding one or a plurality of areas in an image are held as template information. Block areas such as text areas and figure areas are extracted from a document image that has entered from a scanner and an attribute is added to each block area. A block area that at least partially overlaps an area indicated by the template information and whose attribute agrees with the attribute included in the template information is acquired as extracted information from the block areas that have been extracted. This makes it possible to reliably extract a desired area from an entered document image while employing a template.Type: GrantFiled: December 29, 1998Date of Patent: June 25, 2002Assignee: Canon Kabushiki KaishaInventor: Kazuyuki Saito
-
Publication number: 20020076108Abstract: A region extraction apparatus for extracting a desired region from a moving image, includes a region specifying unit for specifying contour points of a region which is to be extracted, a template creating unit for creating a template of the region which is to be extracted based on the contour points of the region specified by the region specifying unit, a template matching unit for matching the template created by the template creating unit and the moving image, and a template correcting unit for correcting the template depending on a matching result obtained by the template matching unit.Type: ApplicationFiled: August 21, 1998Publication date: June 20, 2002Inventors: MAKIKO KONOSHIMA, MORITO SHIOHARA, ATSUKO TADA, TAKASHI ITOH
-
Patent number: 6408107Abstract: An apparatus and method for image registration of a template image with a target image with large deformation. The apparatus and method involve computing a large deformation transform based on landmark manifolds, image data or both. The apparatus and method are capable of registering images with a small number of landmark points. Registering the images is accomplished by applying the large deformation transform.Type: GrantFiled: November 14, 2000Date of Patent: June 18, 2002Inventors: Michael I. Miller, Sarang C. Joshi, Gary E. Christensen
-
Publication number: 20020067404Abstract: A xerographic marking device includes a source of pixels representative of digitized image information. A rendering image processor analyzes the pixels for an image characteristic, and based on the analysis assigns one of at least three values to each pixel. The three values are selected from a set of values representing positive, negative or neutral relative to a predetermined exposure bias. The device also includes an imaging output device which converts the values assigned to the pixels into areas of selective exposure on a charge retentive surface. A charge on the charge retentive surface such as a photoreceptor accordingly, includes an area of positive charge determined to electrostatically react with a charged monochrome marking agent or toner in a first manner, for example by attraction. The charge also includes a negative charge area which is determined to electrostatically react with the toner in an opposite manner, for example to repel toner.Type: ApplicationFiled: December 6, 2000Publication date: June 6, 2002Applicant: XEROX CORPORATIONInventor: Robert P. Loce
-
Publication number: 20020067509Abstract: The present invention is methods, apparatuses and systems for dynamic switching of image processing techniques. Image processing techniques include resolution enhancement for improving the quality of text and line art, halftone image enhancement for improving the quality of gray value or halftone images and toner conservation for reducing toner consumption. Toner conservation may be used simultaneously with other image enhancement processing techniques according to the present invention.Type: ApplicationFiled: December 6, 2000Publication date: June 6, 2002Inventor: Eugene A. Roylance
-
Patent number: 6389163Abstract: The present invention is a method and apparatus for automatic image segmentation using template matching filters. The invention generally segments differing binary textures or structures within an input image by passing one or more structures while removing other structures. More particularly, the method and apparatus segment a stored binary image using a template matching filter that is designed to pass therethrough, for example, text regions while removing halftone regions.Type: GrantFiled: November 18, 1994Date of Patent: May 14, 2002Assignee: Xerox CorporationInventors: Ronald E. Jodoin, Robert P. Loce, Reiner Eschbach
-
Patent number: 6373971Abstract: The method of the present invention discovers patterns in a protein sequences in two phases. In a sampling phase, preferably proper templates corresponding to a group of protein sequences are generated. Patterns corresponding to the templates are then generated and stored in memory. In a convolution phase, the patterns stored in memory are combined to identify a set of maximal patterns.Type: GrantFiled: February 13, 1998Date of Patent: April 16, 2002Assignee: International Business Machines CorporationInventors: Aristidis Floratos, Isidore Rigoutsos
-
Patent number: 6370271Abstract: Pattern detection methods, which may be embodied in copiers or scanners/printers, examine documents to be copied or printed to determine whether the documents contain one or more targeted patterns. The pattern detection methods include algorithms that look for certain features and geometric relationships between the features in determining whether one or more of the targeted patterns are present.Type: GrantFiled: April 30, 1999Date of Patent: April 9, 2002Assignee: Seiko Epson CorporationInventors: Bin Fu, Anoop Bhattacharjya
-
Patent number: 6370568Abstract: An apparatus is provided comprised of a digital camera having a button, a processor connected to the digital camera, and a telecommunications line connected to the processor. When the button of the digital camera is pressed the digital camera creates a digital image and sends the digital image to the processor. The processor sends the digital image out to the telecommunications line. The processor may automatically send the digital image out or may wait for inputs from a user before it sends the digital image out. The processor can be connected to an interactive device such as a computer keyboard or computer mouse. The processor is adaptable to receive a user input in the form of an electronic mailing address and the processor can send the digital image out on the telecommunications line to the electronic mailing address. The processor produces a first screen on a computer monitor.Type: GrantFiled: September 18, 2000Date of Patent: April 9, 2002Inventor: Jeffrey Garfinkle
-
Publication number: 20020033956Abstract: The image processing equipment produces an image of enhanced quality, having dots with higher density than in an original image, by collating an original image with templates. The present invention reduces required number of templates. Original image data in a collation window is either mirror-converted, flip-converted or flip-and-mirror-converted in an array converter (33) and selected on a time-division basis to collate with a single template pattern (34). The effect is equivalent to the collation with maximum four templates. Thus the number of template patterns to be prepared is greatly reduced. Also, by preparing additional patterns to be stored into an area that became vacant attributed to the reduction of patterns, more sophisticated image enhancement may be attained.Type: ApplicationFiled: February 9, 2001Publication date: March 21, 2002Applicant: Fujitsu Limited, Kawasaki, JapanInventor: Mikio Koga
-
Patent number: 6360013Abstract: A method is described that greatly reduces the computational cost of determining the globally optimal match between a template and one or more samples. For every sample in a search range, intermediate distance measures between the template and each sample are first computed in one designated dimension. The computed distance measures are then sorted according to their magnitude. The sample with the minimal distance measure is selected to accumulate a new distance measure in the next higher dimension. This new distance measure is recorded and the samples are sorted again according to the updated distance measures. The above process is repeated until a minimal distance measure has been computed in all dimensions. For motion estimate, this method can reduce the number of computational operations by about 90% to 99%. This method can also provide a list of globally best matches.Type: GrantFiled: March 16, 1999Date of Patent: March 19, 2002Assignee: Academia SinicaInventors: Yi-Ping Hung, Yong-Sheng Chen
-
Publication number: 20020025075Abstract: A system for processing stereo matching of a video image sequence in a real-time mode is provided. The system includes a signal converting means for converting an image input from a first camera and a second camera into a digital signal; and an image matching means for calculating a determined matching cost based on a pair of pixels in one scan line of the first and second digital image signals, tracing the decision value which determines the minimum matching cost, and outputting the decided value as an estimated disparity according to determined activation information; and a display means for displaying the output from the image matching means. According to the system, real-time stereo matching is enabled, by parallel processing or video image sequences using an algorithm which is based on a new dynamic trellis based method and is optimal in the Bayesian sense.Type: ApplicationFiled: May 29, 2001Publication date: February 28, 2002Inventors: Hong Jeong, Yun-Soo Oh
-
Publication number: 20020009231Abstract: A plurality of images and image features of each of the plurality of images are stored in a form correlated with the images in an external storage device of an image retrieval apparatus. In retrieval of a desired image from the plurality of images that have been stored in the external storage device, a plurality of image features regarding one image are acquired by varying, over a predetermined range, each of the image features that have been stored in the external storage device. A fuzzy image search is conducted by calculating the degree of similarity between each of the plurality of images and a retrieval source image based upon the image features that have been acquired and the image features of the retrieval-source image.Type: ApplicationFiled: March 27, 2001Publication date: January 24, 2002Inventors: Kunihiro Yamamoto, Kentaro Matsumoto, Kiyoshi Kusama
-
Patent number: 6339651Abstract: A method and system for recognizing the characters on surfaces where alphanumeric identification code (“ID” for short) may be present such as a license plate. The present system is particularly adapted for situations where visual distortions can occur, and utilizes a highly robust method for recognizing the characters of an ID. Multiple character recovery schemes are applied to account for a variety of conditions to ensure high accuracy in identifying the ID. Accuracy is greatly enhanced by taking a comprehensive approach where multiple criteria are taken into consideration before any conclusions are drawn. Special considerations are given to recognizing the ID as a whole and not just the individual characters.Type: GrantFiled: February 25, 1998Date of Patent: January 15, 2002Assignee: Kent Ridge Digital LabsInventors: Qi Tian, Kong Wah Wan, Karianto Leman, Chade Meng Tan, Chun Biao Guo
-
Patent number: 6333998Abstract: Disclosed are an image processing method and apparatus capable of improving character quality by applying smoothing processing even in a case where a faint character image is binarized. A multilevel digital image signal from an image input unit is binarized by a binarizing processor and the resulting bilevel image signal is delayed by a delay circuit for as long as required for pattern matching. Using the bilevel image signal from the delay circuit and the latest bilevel image signal, and in accordance with the predetermined size of a matrix for pattern matching, a pattern matching unit performs pattern matching between the bilevel image signal and all reference patterns stored in a ROM.Type: GrantFiled: September 4, 1998Date of Patent: December 25, 2001Assignee: Canon Kabushiki KaishaInventor: Atsushi Matsumoto
-
Patent number: 6304667Abstract: A method for incorporating dyslexia detection into existing handwriting pattern recognition software includes a database of distorted characters along with the database of legitimate characters with which the handwriting samples are compared. The distorted characters are those recognized as typical of those produced by dyslexic persons. The system used may be a personal computer, a personal digital assistant, or an Internet system. Whatever system is used, a means of freehand drawing must be included, such as a stylus with pressure sensitive drawing area, a drawing tablet, or pen-pad-type device. A character is first presented to the student for reproduction. The student then draws a character on the pad, attempting to reproduce the character. The system software detects a correspondence between the character input and the legitimate characters, when occurring.Type: GrantFiled: June 21, 2000Date of Patent: October 16, 2001Inventor: Carmen T. Reitano
-
Publication number: 20010028741Abstract: In a method for matching two images, each consisting of a plurality of pixels and having partially overlapping contents, the degree of correspondence between the contents of the images is determined for different displacement positions representing different overlappings of the images. More specifically, a plurality of numbers are determined for each one of a plurality of said displacement positions. Each number is formed with the aid of pixel values from both images and is used to simultaneously retrieve predefined overlap assessment values for at least two of said displacement positions. The overlap assessment values retrieved are subsequently used to determine the degree of correspondence between the images for the different displacement positions. The method is carried out with the aid of a computer and can be implemented as a computer program.Type: ApplicationFiled: February 17, 1998Publication date: October 11, 2001Inventors: CHRISTER FAHRAEUS, OLA HUGOSSON, PETTER ERICSON
-
Patent number: 6301387Abstract: A template is analyzed to determine a step size for searching within a search area. First, the template is padded with pixels to increase its size. Cross-correlation between the padded template and the original template leads to identification of an effective step size. Step sizes for each of horizontal and vertical axes are derived. Windows of the search area, selected based on the step size, then are tested in a fast search stage by correlating the template to the window. Any tested window which has a correlation coefficient exceeding a specific value is a local match. A full search of the vicinity of the local match then is performed for all potential windows within an area bounded by one step to either side of the local match along either axis. The location(s) corresponding to the highest correlation(s) exceeding the threshold value are matches.Type: GrantFiled: December 18, 1998Date of Patent: October 9, 2001Assignee: University of WashingtonInventors: Shijun Sun, Hyun Wook Park, Yongmin Kim
-
Patent number: 6285788Abstract: A method for expediting a search of a digital image database comprises, for each digital image, constructing a set of at least two abstracted images varying in level of detail and searching the abstracted images. Selected ones of the abstracted images are down-loaded to a user thereby reducing download time through a finite bandwidth communicative medium relative to the time needed to download complete digital images.Type: GrantFiled: May 27, 1998Date of Patent: September 4, 2001Assignee: Sharp Laboratories of America, Inc.Inventors: Muhammed Ibrahim Sezan, Richard Junqiang Qian
-
Patent number: 6272246Abstract: In a positioning device, a maximum-value/minimum-value calculating section calculates the maximum value and minimum value of densities with respect to each of a target image and a reference image that has been preliminarily stored. Further, based upon the maximum values and minimum values of densities of the two images, a barrel-shifter gradation conversion section carries out bit shifts of the original images of the two images respectively so as to allow them to contain effective bits related to contrast, and gradation-converts the respective images so as to have numbers of density levels that are fewer than the original numbers of density levels thereof, thereby outputting the resulting images that have been gradation-converted.Type: GrantFiled: January 6, 1998Date of Patent: August 7, 2001Assignee: Sharp Kabushiki KaishaInventor: Kensuke Takai
-
Patent number: 6272247Abstract: A system for digital image recognition which combines sparse correlation with image pyramiding to reduce the number of pixels used in correlation provides effective recognition of a reference image template without exhaustive correlation of all pixels in the reference image template. An optimal sparse pixel set is selected from the pixels of the reference image template by correlating the reference image template against a search image scene which is to be searched. Such a sparse pixel set includes those pixels which are optimal in defining the correlation sensitive features of the reference image template. By terminating the accumulation of sparse pixels at an optimal point, performance is maximized without compromising accuracy of recognition. The resultant optimal sparse pixel set is then correlated against the pixels in the search image scene through a series of transformations to find a match of the reference image template within the search image scene.Type: GrantFiled: May 18, 1999Date of Patent: August 7, 2001Assignee: Datacube, Inc.Inventors: Swaminathan Manickam, Scott D. Roth, Thomas W. Bushman
-
Patent number: 6266445Abstract: A sample image (142) is recognized by normalizing (404) the size of a sample image (142) to the size of a referent images (146); and determining (406) a set of candidate images (147) from a set of referent images (146), wherein each of the candidate images (147) is within an acceptable distance from a different binarization (145) of the sample image (142). A system (120) for image recognition includes a scanning device (126), a normalization unit (134), a distance calculation unit (136), a classification unit (138), a disambiguation unit (140), and a display device (128).Type: GrantFiled: March 13, 1998Date of Patent: July 24, 2001Assignee: Canon Kabushiki KaishaInventors: Radovan V. Krtolica, Roger D. Melen
-
Patent number: 6263097Abstract: Method of determining (a) symmetry line(s) in a radiation image. A method of determining (a) symmetry line(s) in horizontal (vertical) direction in an image that is represented by a matrix of digital pixel values by (i) calculating a symmetry point for each column (row) of the image, and (ii) interconnecting calculated symmetry points.Type: GrantFiled: June 24, 1998Date of Patent: July 17, 2001Assignee: Agfa-GevaertInventor: Piet Dewaele
-
Patent number: 6243494Abstract: A template is analyzed to determine step sizes for searching within a search area. The template is analyzed by first padding the template with data points to increase its size. Cross-correlation between the padded template and the original template leads to identification of an effective step size along multiple axes. Step sizes for each of a horizontal, vertical and a third axis are derived. Third axis step sizes may correspond to rotation, scaling factor, subsampling factor, linear distance, time or frequency. Windows of the search area, selected based on the step sizes, then are tested in a fast search by correlating the template to selected windows to derive correlation coefficients. Any tested window which has a correlation coefficient exceeding a given value is a potential match for the template and is subject to a refined stage of comparison.Type: GrantFiled: December 18, 1998Date of Patent: June 5, 2001Assignee: University of WashingtonInventors: Shijun Sun, Yongmin Kim
-
Patent number: 6229914Abstract: To reliably and simply prevent the output of specified image data, a unit to detect a pattern is installed in one of the devices in a system consisting of an image scanner 1, control device 2 (a personal computer) and printer 3 which are connected consecutively. This pattern detection unit contains a unit which detects a specified pattern from the mass of image data representing a document. Once this pattern has been found, either the subsequent image data are altered before being output or output of the image data is interrupted. When the system starts up, the control unit for each device checks whether that device, or one of the other devices in the system, has a pattern detection unit as described above. If it determines that no device has such a unit, it halts operation.Type: GrantFiled: March 6, 2000Date of Patent: May 8, 2001Assignee: Omron CorporationInventors: Junji Hiraishi, Tetsuya Miwa, Hitoshi Nakamura, Kiyoshi Imai, Kouichi Ohmae, Shinya Sonoda, Souichi Matsushita, Minoru Inada
-
Patent number: 6222940Abstract: A system and method for performing pattern matching to locate zero or more instances of a template image in a target image. The method first comprises sampling the template image using a Low Discrepancy sequence, also referred to as a quasi-random sequence, to determine a plurality of sample pixels in the template image which accurately characterize the template image. The Low Discrepancy sequence is designed to produce sample points which maximally avoid each other. After the template image is sampled or characterized, the method then performs pattern matching using the sample pixels and the target image to determine zero or more locations of the template image in the target image. The method may also perform a local stability analysis around at least a subset of the sample pixels to determine a lesser third number of sample pixels which have a desired degree of stability, and then perform pattern matching using the third plurality of sample pixels.Type: GrantFiled: January 6, 1999Date of Patent: April 24, 2001Assignee: National Instruments CorporationInventors: Lothar Wenzel, Dinesh Nair, Nicolas Vazquez, Samson Dekey