Local Neighborhood Operations (e.g., 3x3 Kernel, Window, Or Matrix Operator) Patents (Class 382/205)
-
Patent number: 6885763Abstract: For the purpose of providing an image processing method for efficiently performing filtering, a local region containing a pixel of interest is defined in a plurality of modes in an original image (502); sequentially for the region in the plurality of modes, the variance of pixel values is obtained and a decision is made whether the value of the variance falls within a predetermined range (504, 506, 508 512, 514); and an image is produced using as a new pixel value for the pixel of interest an average value of pixel values of the region for which the value of the variance first falls within the range (508, 510, 524).Type: GrantFiled: June 18, 2001Date of Patent: April 26, 2005Assignee: GE Medical Systems Global Technology Company, LLCInventor: Tetsuo Ogino
-
Patent number: 6879721Abstract: A pixel processing system receives a target pixel value and neighbor pixel values. The neighbor pixel values may correspond to eight neighbor pixels that surround a target pixel corresponding to the target pixel value. The pixel processing system determines a minimum value and a maximum value among the neighbor pixel values. If the target pixel value is less than the minimum value, then the pixel processing system increases the target pixel value. If the target pixel value is greater than the maximum value, then the pixel processing system reduces the target pixel value. In response to increasing the target pixel value, the pixel processing system may reduce the neighbor pixel values to maintain an average value. In response to reducing the target pixel value, the pixel processing system may increase the neighbor pixel values to maintain an average value.Type: GrantFiled: January 2, 2001Date of Patent: April 12, 2005Assignee: Hewlett-Packard Development Company, L.P.Inventor: Brent M. Bradburn
-
Efficient block transform including pre-processing and post processing for autostereoscopic displays
Patent number: 6868177Abstract: Image data processing techniques utilizing multiple processes operating on one or more processors of one or more computer systems can be used to efficiently rearrange or reparameterizing image data to form hologram element (hogel) images which can then be used to produce holographic stereograms. These techniques utilize matrix manipulation of portions of image data both within a single process, and across multiple processes typically executing on different processors, and each utilizing a subset of the overall set of image data.Type: GrantFiled: October 19, 2001Date of Patent: March 15, 2005Assignee: Zebra Imaging, Inc.Inventors: Emilio Camahort, Mark E. Holzbach, Robert L. Sitton -
Patent number: 6867773Abstract: A method of determining neighboring voxels that contribute to a surface of an object includes evaluating a primary voxel that corresponds to a side of a reference voxel. The side of the reference voxel is adjacent to an outside voxel, which is outside the object. The primary voxel is adjacent to the outside voxel and is adjacent to a corresponding secondary voxel. The secondary voxel is age adjacent to a corresponding tertiary voxel. The method also includes determining whether the primary voxel, the secondary voxel or the tertiary voxel is on the same surface as the reference voxel.Type: GrantFiled: May 13, 2002Date of Patent: March 15, 2005Assignee: Intel CorporationInventor: William A. Hux
-
Patent number: 6865296Abstract: With respect to two pattern sets obtained on different conditions, a feature-extraction matrix, which maximizes between-class scatter and minimizes within-class scatter, is found respectively. A first feature amount is calculated using one of the feature-extraction matrices. The first feature amount and the two matrices are retained in a referential database. A pattern is determined to a second feature amount—extracted by applying another feature-extraction matrix to a pattern input—by extracting a most similar element out of the first feature amount retained in the referential database.Type: GrantFiled: June 5, 2001Date of Patent: March 8, 2005Assignee: Matsushita Electric Industrial Co., Ltd.Inventor: Kenji Nagao
-
Patent number: 6862367Abstract: An artificial vision system includes means (12, 14) for generating an image percept vector, means (16) for transforming this image percept vector into a feature vector, and means (16) for generating a response vector by multiplying the feature vector by a corresponding trained linkage matrix modeling a percept-response system.Type: GrantFiled: June 10, 1999Date of Patent: March 1, 2005Inventor: Gosta Granlund
-
Publication number: 20040264781Abstract: Methods and systems used to automatically identify the marking process used for an image on a substrate based on spatial characteristics and/or color of the image. Image types which are classified and identified include continuous tone images and halftone images. Among halftone images separately identified are inkjet images, xerographic images and lithographic images. Locally adaptive image threshold techniques may be used to determine the spatial characteristics of the image.Type: ApplicationFiled: June 30, 2003Publication date: December 30, 2004Applicant: XEROX CORPORATIONInventors: Reiner Eschbach, Gaurav Sharma, Shen-ge Wang
-
Publication number: 20040264771Abstract: Methods and systems used to associate color calibration profiles with scanned images based on identifying the marking process used for an image on a substrate using spatial characteristics and/or color of the image. Image types which are classified and identified include continuous tone images and halftone images. Among halftone images separately identified are inkjet images, xerographic images and lithographic images. Locally adaptive image threshold techniques may be used to determine the spatial characteristics of the image.Type: ApplicationFiled: June 30, 2003Publication date: December 30, 2004Inventors: Gaurav Sharma, Reiner ESCHBACH, Shen-ge WANG
-
Publication number: 20040264770Abstract: Methods and systems used to associate color calibration profiles with scanned images based on identifying the marking process used for an image on a substrate using spatial characteristics and/or color of the image. Image types which are classified and identified include continuous tone images and halftone images. Among halftone images separately identified are inkjet images, xerographic images and lithographic images. Locally adaptive image threshold techniques may be used to determine the spatial characteristics of the image.Type: ApplicationFiled: June 30, 2003Publication date: December 30, 2004Applicant: XEROX CORPORATIONInventors: Gaurav Sharma, Reiner ESCHBACH, Shen-ge WANG
-
Publication number: 20040264768Abstract: Methods and systems used to associate color calibration profiles with scanned images based on identifying the marking process used for an image on a substrate using spatial characteristics and/or color of the image. Image types which are classified and identified include continuous tone images and halftone images. Among halftone images separately identified are inkjet images, xerographic images and lithographic images. Locally adaptive image threshold techniques may be used to determine the spatial characteristics of the image.Type: ApplicationFiled: June 30, 2003Publication date: December 30, 2004Applicant: XEROX CORPORATIONInventors: Gaurav Sharma, Reiner ESCHBACH, Shen-ge WANG
-
Publication number: 20040264769Abstract: Methods and systems used to associate color calibration profiles with scanned images based on identifying the marking process used for an image on a substrate using spatial characteristics and/or color of the image. Image types which are classified and identified include continuous tone images and halftone images. Among halftone images separately identified are inkjet images, xerographic images and lithographic images. Locally adaptive image threshold techniques may be used to determine the spatial characteristics of the image.Type: ApplicationFiled: June 30, 2003Publication date: December 30, 2004Applicant: XEROX CORPORATIONInventors: Gaurav Sharma, Reiner ESCHBACH, Shen-ge WANG
-
Publication number: 20040258304Abstract: Automatic selection of best shots from among a plurality of photographic images is enabled, while retaining photographic images of scenes, for which a possibility exists that a user feels are necessary, regardless of image quality. A readout portion reads out a plurality of photographic images, which are recorded on a memory card. A classifying portion classifies these photographic images into similar photographic image groups, which include similar photographic images. A selection condition setting portion of an extracting portion sets selection conditions so that at least one qualified photographic image is selected from each similar photographic image group. The extracting portion extracts qualified photographic image groups form each similar photographic image group, based on the set selection conditions. A printing portion prints only the qualified photographic images, and a recording portion records the qualified photographic images and other photographic images in different folders of a CD-R.Type: ApplicationFiled: January 29, 2004Publication date: December 23, 2004Inventors: Kazuo Shiota, Kiyotaka Kaneko
-
Patent number: 6807286Abstract: A system and process for recognizing an object in an input image involving first generating training images depicting the object. A set of prototype edge features is created that collectively represent the edge pixel patterns encountered within a sub-window centered on each pixel depicting an edge of the object in the training images. Next, a Hough kernel is defined for each prototype edge feature in the form of a set of offset vectors representing the distance and direction, from each edge pixel having an associated sub-window exhibiting an edge pixel pattern best represented by the prototype edge feature, to a prescribed reference point on a surface of the object. The offset vectors are represented as originating at a central point of the kernel. For each edge pixel in the input image, the prototype edge feature which best represents the edge pixel pattern exhibited within the sub-window centered on the edge pixel is identified.Type: GrantFiled: April 13, 2000Date of Patent: October 19, 2004Assignee: Microsoft CorporationInventors: John Krumm, Richard J. Campbell
-
Patent number: 6804375Abstract: Electronic watermarks in digital documents can be detected, even if the content of the digital documents has been subjected to distortion or other attempts to hide or destroy the watermarks. A method involves inputting the distorted image and comparison information, the comparison information including at least one of the original image or information used for embedding the electronic watermark; dividing a domain of the original image into a plurality of patches, based on the comparison information; inputting affine parameters of a predetermined patch from among the patches in the original image; extracting a patch candidate from the distorted image; using a predetermined electronic watermark detection method, judging whether the patch candidate in the distorted image adequately correlates with a neighboring patch in the original image; when the judging indicates an adequate correlation, outputting a part of the electronic watermark.Type: GrantFiled: June 27, 2001Date of Patent: October 12, 2004Assignee: Kabushiki Kaisha ToshibaInventor: Hirofumi Muratani
-
Patent number: 6795214Abstract: The method generates trapping contours on color edges in a page to be printed that is defined in digital data. The trapping contours reduce the defects that are caused by register errors in the printing machine. The method operates similarly to a filter, using a window that is moved over the entire page, picture element by picture element. Certain neighboring picture elements of the central element are selected and used for calculating a trapping color. The color separation proportions of the central picture element are compared with the corresponding color separation proportions of the trapping color. The result of the comparison determines which color separation proportions of the central picture element are to be changed.Type: GrantFiled: September 19, 2001Date of Patent: September 21, 2004Assignee: Heidelberger Druckmaschinen AGInventors: Peter Weinholz, Volkmar Funke
-
Publication number: 20040179747Abstract: There are provided an image reading apparatus, in which a plurality of image reading sensors are joined one another, and which is capable of simply and precisely compensating for defective data in joint portions of the plurality of image reading sensors through interpolation so as to obtain a high-quality image. The image reading apparatus includes an image reading portion for reading an image on an original for each pixel with the use of a plurality of image reading sensors arranged in a row in the direction of the array of pixels, and a pixel interpolation portion for computing, with the use of a filter portion, interpolation pixel data for compensating, through interpolation, for a target pixel that is an unreadable pixel situated between adjacent image reading sensors.Type: ApplicationFiled: March 8, 2004Publication date: September 16, 2004Applicant: CANON KABUSHIKI KAISHAInventor: Atsushi Matsumoto
-
Publication number: 20040174546Abstract: A method for segmenting a compound document initiates with receiving image data, followed by processing the image data at both a first scale and a second scale. Next, region types of the compound document are determined according to a combination of processing output at the first scale and processing output at the second scale. Methods for labeling digital image data and labeling and enhancing documents defined through digital data, as well as associated computer readable medium embodiments are provided. An image replication device and a microprocessor configured to segment and enhance image data associated with a compound document are also provided.Type: ApplicationFiled: July 23, 2003Publication date: September 9, 2004Inventor: Onur G. Guleryuz
-
Publication number: 20040175054Abstract: The invention provides an image processing apparatus and an image processing method. An area to which image data belongs is discriminated, and a correction coefficient to be used for correction of a pixel value of the image data is produced based on a result of the discrimination. Then, the pixel value of the image data is corrected with the correction coefficient. The relationship in magnitude among pixel values in the same area is maintained because the same coefficient is used, but pixel values which belong to different areas can be varied or even reversed. This allows the gradation of an entire image to be corrected while preventing partial deterioration of the contrast.Type: ApplicationFiled: March 16, 2004Publication date: September 9, 2004Inventors: Masami Ogata, Takashi Tsuchiya, Kazuhiko Ueda
-
Publication number: 20040170325Abstract: Systems and methods for image pattern recognition comprise digital image capture and encoding using vector quantization (“VQ”) of the image. A vocabulary of vectors is built by segmenting images into kernels and creating vectors corresponding to each kernel. Images are encoded by creating a vector index file having indices that point to the vectors stored in the vocabulary. The vector index file can be used to reconstruct an image by looking up vectors stored in the vocabulary. Pattern recognition of candidate regions of images can be accomplished by correlating image vectors to a pre-trained vocabulary of vector sets comprising vectors that correlate with particular image characteristics.Type: ApplicationFiled: February 9, 2004Publication date: September 2, 2004Applicant: Aperio Technologies, Inc.Inventors: Ole Eichhorn, Dirk G. Soenksen
-
Patent number: 6781718Abstract: An image correcting method and apparatus is disclosed in which processing other than a smoothing process can be performed in accordance with conditions of a line such as correction of a line width with a small circuit size and at low cost. The disclosed dot image correcting method stores correction data corresponding to a correction dot pattern(s) in a window(s), determines whether or not a dot pattern coincides with a correction dot pattern, and when coincidence is detected, performs correction in accordance with corresponding correction data, characterized in that there are a plurality of windows, and correction data are stored corresponding to a combination of correction dot patterns in various windows, and determination is made as to whether or not dot patterns in a plurality of windows coincide with a combination of correction dot patterns in various windows, and when coincidence is detected, correction is performed in accordance with corresponding correction data.Type: GrantFiled: August 30, 2002Date of Patent: August 24, 2004Assignee: Fuji Xerox Co., Ltd.Inventor: Kazuhiko Sato
-
Publication number: 20040161153Abstract: Structured defects in a digital image are detected by examining at least one context-dependent property of candidate image regions of the digital image.Type: ApplicationFiled: February 18, 2003Publication date: August 19, 2004Inventor: Michael Lindenbaum
-
Publication number: 20040156079Abstract: A laser print apparatus includes a memory for storing a multi-bit image including a plurality of pixels. Each pixel is represented by an N-bit value, wherein N is greater than one. A modulation code generator analyzes three adjacent pixels. The three adjacent pixels include a left pixel, a center pixel, and a right pixel. The modulation code generator is configured to generate a pulse width value based on the value of the center pixel, and a justification value based on the values of the left pixel and the right pixel. A laser print engine forms an output pixel on media based on the pulse width value and the justification value.Type: ApplicationFiled: February 12, 2003Publication date: August 12, 2004Inventors: John D. Marshall, Douglas G. Keithley
-
Patent number: 6775409Abstract: The method comprises acquiring a bidimensional image formed by a plurality of dots, characterized by a brightness value; dividing the bidimensional image into a plurality of image regions, each formed by a respective first number of dots; for each of the image regions, determining a first quantity correlated to the brightness variation intensity of each dot with respect to adjacent dots; for each image region, determining at least one second quantity correlated to the brightness variation direction of each dot with respect to adjacent dots; and determining any presence and type of a code in each of the image regions, on the basis of the determined quantities. These quantities typically comprise the contrast level of the image region, obtained by adding the gradient variation intensity of all contrast dots of the image region and directionality of the image region, obtained by calculating the gradient variation direction for all dots and searching for the prevalent direction or directions.Type: GrantFiled: October 22, 1999Date of Patent: August 10, 2004Assignee: Data Logic S.p.A.Inventors: Elena Brunelli, Francesco Caselli, Marco Bottazzi
-
Patent number: 6775031Abstract: An image forming apparatus which is composed of an image reading unit for reading a document image and converting the document image into image data, an image processing unit for applying image processes containing edge enhancing process and smoothing process to the image data, and an image forming unit for forming the processed image data on a printing medium. The image processing unit includes a halftone dot detector for detecting a halftone dot region using the image data, a first edge detector for executing edge detection over the detected halftone dot region using a derivative filter, and a second edge detector for executing edge detection over a non-halftone dot region other than the detected halftone dot region using a derivative filter with a size different from a size of the derivative filter.Type: GrantFiled: March 23, 2000Date of Patent: August 10, 2004Assignee: Minolta Co., Ltd.Inventor: Yoko Fujiwara
-
Patent number: 6763129Abstract: An input image is inputted into a pre-process circuit and a spatial filter. The pre-process circuit emphasizes a linear shadow or an edge to some extent, or reduces noises overall. Following pre-processing, the image signal is supplied to a direction detection circuit, a direction and its intensity of an area of interest are detected. These two results of the detections are correlated with peripheral areas of interest to some extent by a vector filter. An output of the vector filter is supplied to a spatial filter coefficient generation circuit, and a filter coefficient of the spatial filter is decided. The spatial filter processes the input image based on the direction of the shadow and its intensity, whereby selectively suppressing noise components of the background in real time without damaging the contrasts of a linear shadow or an edge-like shadow.Type: GrantFiled: October 4, 2000Date of Patent: July 13, 2004Assignee: Kabushiki Kaisha ToshibaInventors: Michitaka Honda, Hiroko Umazaki
-
Publication number: 20040125409Abstract: In the image processing apparatus, image data is divided into large blocks of a prescribed size and the large blocks are subdivided into small blocks by the dividing unit. The number of isolated points in each large block is then calculated by the large block isolated point calculation unit, and the number of isolated points in each small block is then calculated by the small block isolated point calculation units. It is then determined by the halftone-dot region determination unit whether or not the large block is a halftone-dot region. This determination considers both the number of isolated points in the large block and the number of isolated points in each small block.Type: ApplicationFiled: September 16, 2003Publication date: July 1, 2004Applicant: MINOLTA CO., LTD.Inventors: Tomohiro Yamaguchi, Yoshihiko Hirota
-
Publication number: 20040114186Abstract: A method for selecting a halftoning mode to be applied to regions within an image. The method operates on received image data to identify low-level regions within the received image data. The low-level regions are defined and separated by edges. Pixels that are not separated by edges are considered to belong to the same region. After the regions have been identified, low-level image attributes for the image data within each region are compiled. A halftoning method suited for the image data within the region can be selected based upon low-level image attributes.Type: ApplicationFiled: December 17, 2002Publication date: June 17, 2004Applicant: Xerox CorporationInventor: Zhigang Fan
-
Publication number: 20040114185Abstract: A method to detect frequency and angle of a binary halftone pattern. The method employs an exclusive-or operation which is applied locally to a region of a binary bit map and its spatially shifted version. The resulting bits from the exclusive-or operation are summed over the region. The exclusive-or operation is repeated for a range of shift values. In a halftone region, the shift at which the minimum sum occurs reflects the angle and the frequency of the halftone.Type: ApplicationFiled: December 12, 2002Publication date: June 17, 2004Applicant: Xerox CorporationInventors: Jeng-nan Shiau, Ying-wei Lin
-
Patent number: 6751363Abstract: Methods of imaging objects based on wavelet retrieval of scenes utilize wavelet transformation of plural defined regions of a query image. By increasing the granularity of the query image to greater than one region, accurate feature vectors are obtained that allow for robust extraction of corresponding regions from a database of target images. The methods further include the use of sliding windows to decompose the query and target images into regions, and the clustering of the regions utilizing a novel similarity metric that ensures robust image matching in low response times.Type: GrantFiled: August 10, 1999Date of Patent: June 15, 2004Assignee: Lucent Technologies Inc.Inventors: Apostol Ivanov Natsev, Rajeev Rastogi, Kyuseok Shim
-
Patent number: 6750984Abstract: The present invention is directed to the provision of an image processing apparatus which enables the characteristic of an image region to be discriminated accurately, even in the presence of pixel displacements, noise, etc. Within masks for three colors, each mask consisting of m×n pixel including an object pixel, sub-region masks of u×v are set, and the sum of pixel densities is calculated by a sum calculation circuit. Based on the thus calculated sum, a parameter calculation circuit calculates a parameter based on which a multi-valued signal generation circuit generates a multi-valued signal. A characteristic signal represented by multi-valued data is generated by a characteristic signal generation circuit from the distribution of data within a region discriminating mask represented by the multi-valued signal, and a segmentation signal representing the result of region characteristic discrimination is output from a segmentation signal generation circuit.Type: GrantFiled: February 11, 2000Date of Patent: June 15, 2004Assignee: Sharp Kabushiki KaishaInventors: Toshihiro Kanata, Yasushi Adachi
-
Publication number: 20040105108Abstract: A parameter representing whether an effective density pixel exists in an objective image is calculated. As to each interest pixel in the objective image having one pixel neighboring a downstream side in a first direction, the interest pixel is defined as an effective density pixel when a first condition and a second condition is an effective density pixel are satisfied, and in other case, the interest pixel is defined as an ineffective density pixel. The objective image is converted to an update image on one-line reduced lines in the first direction. A parameter representing whether an effective density pixel exists in the update image is calculated. The conversion and the calculation are repeated about the update image as the objective image. When the objective image converts one line in the first direction, the number of continuous pixels of the effective density pixels spatially continuing is calculated based on each parameter.Type: ApplicationFiled: September 25, 2003Publication date: June 3, 2004Inventor: Yasushi Adachi
-
Patent number: 6734990Abstract: Discloses an image formation apparatus capable of achieving a picture quality improvement of a multi-value image without increasing the size of a look-up table to a size larger than the size of a look-up table that is used for the picture quality improvement at the time of forming two-value image data. The image formation apparatus can form an image having at least one density in addition to two values of white and black in pixel units, and includes a density separation section that distributes a multi-value image data to a plurality of density planes according to a predetermined rule, a correction signal generator that outputs correction signals for correcting the layout of pixels in each of the plurality of density planes, and a correction signal combiner that combines the correction signals output from the density planes according to a predetermined combining rule.Type: GrantFiled: June 7, 2000Date of Patent: May 11, 2004Assignee: Fuji Xerox Co., Ltd.Inventor: Masaru Iida
-
Patent number: 6735336Abstract: A method of comparing an input pattern with a memory pattern includes the steps of loading a representation of said input pattern into cells in an input layer; loading a representation of said memory pattern into cells in a memory layer; loading an initial value into cells in an intermediate layers between said input layer and said memory layer; comparing values of cells in said intermediate layers with values stored in cells of adjacent layers; updating values stored in cells in said intermediate layers based on said step of comparing; and mapping cells in said memory layer to cells in said input layer.Type: GrantFiled: May 15, 2002Date of Patent: May 11, 2004Assignee: Applied Neural Computing Ltd.Inventors: Yossi Avni, Eytan Suchard
-
Patent number: 6735745Abstract: Method for detecting defects, the method comprising the procedures of identifying theoretically-symmetrical windows in an object-image; analyzing the theoretically-symmetrical windows according to expected symmetry of the theoretically-symmetrical windows; and determining the presence of defects according to a deviation from the expected symmetry.Type: GrantFiled: February 7, 2002Date of Patent: May 11, 2004Assignee: Applied Materials, Inc.Inventor: Nimrod Sarig
-
Publication number: 20040088723Abstract: Systems and methods to generate a video summary of a video data sequence are described. In one aspect, key-frames of the video data sequence are identified independent of shot boundary detection. A static summary of shots in the video data sequence is then generated based on key-frame importance. For each shot in the static summary of shots, dynamic video skims are calculated. The video summary consists of the calculated dynamic video skims.Type: ApplicationFiled: November 1, 2002Publication date: May 6, 2004Inventors: Yu-Fei Ma, Lie Lu, Hong-Jiang Zhang, Mingjing Li
-
Patent number: 6731806Abstract: A method for removing noise from a pixel of a digital image channel includes the steps of: storing an original value of the pixel of interest and using the value of the pixel of interest and neighboring contiguous pixels to determine a variable shape neighborhood region of cleaning pixels. The method further includes using the neighborhood region of cleaning pixels and the value of the pixel of interest to replace the original value of the pixel of interest with a noise cleaned pixel value.Type: GrantFiled: October 8, 1999Date of Patent: May 4, 2004Assignee: Eastman Kodak CompanyInventor: Edward B. Gindele
-
Patent number: 6728399Abstract: A method and apparatus is disclosed for converting raster images into a vector format by identifying and converting the borders of image features into a mathematical format referred to as string sequences. To enable the string sequencing of color images, the present invention identifies and resolves contrast conflicts in the image features in order to avoid misperceiving the vectorized image when converted back into a raster format. A contrast conflict occurs when there is a “contrast tie” between overlapping features of the image. The feature that would normally be perceived as the dominant feature breaks the contrast tie so that when the vector image is reconstructed, the dominate feature appears in the foreground of the image while the recessive feature appears in the background.Type: GrantFiled: March 31, 1999Date of Patent: April 27, 2004Assignee: Colorcom, Ltd.Inventor: Joseph Doll
-
Patent number: 6711303Abstract: An improved method and computer program product for detecting an amount of rotation or magnification in a modified image, of the type including the steps of: a) embedding a marker image having a pair of identical features separated by a distance d and oriented at an angle &agr; in an original image to produce a marked image, the marked image having been rotated and/or magnified to produce the modified image; b) performing an autocorrelation on the modified image to produce a pair of autocorrelation peaks corresponding to the location of the features of the marker image in the modified image; and c) comparing the separation d and orientation &agr; of the autocorrelation peaks with the separation d and orientation &agr; of the features in the marker image to determine the amount of rotation and magnification in the modified image, wherein the improvement comprises the step of: d) preprocessing the modified image to have a constant standard deviation over all local regions in the image.Type: GrantFiled: December 1, 1999Date of Patent: March 23, 2004Assignee: Eastman Kodak CompanyInventor: Chris W. Honsinger
-
Patent number: 6707941Abstract: A method for reducing the quantity of digital information in a data set for the purpose of pattern recognition, comprising: a) determining digital a priori information values associated with at least one known pattern; b) determining digital information values of first elements associated with a pattern for recognition; c) pairing two or more first elements into second elements, wherein the number of digital information values for each second element is at least doubled; d) for each second element, on the basis of the digital a priori information values, merging a minimum of two digital information values into a reduced second element with a reduced number of digital information values.Type: GrantFiled: March 6, 2000Date of Patent: March 16, 2004Inventor: Dirk Ernest Maria Van Dyck
-
Patent number: 6701009Abstract: A method for separating foreground and background image pixels from other pixels. The method determines if a current pixel is an edge pixel, and then selects a window of pixels surrounding the current pixel if it is on an edge. This window is then analyzed to produce a classification of a gradient field within the window, and defines foreground and background image pixel values depending upon the classification. Once foreground and background pixel positions are defined, the process interpolates the value for the current pixel. The accurate designation of foreground and background pixels eliminates or mitigates any color misregistration error in the pixel value.Type: GrantFiled: June 6, 2000Date of Patent: March 2, 2004Assignee: Sharp Laboratories of America, Inc.Inventors: Otsu Makoto, William Ho Chang
-
Publication number: 20040027614Abstract: In a system, such as in a digital copier, for classifying image data derived from an original image, the image data is enhanced before submission to a classification analysis. In the enhancement, portions of the image having a measured roughness (degree of gray level discontinuity) within a certain range are artificially increased in roughness. This increased roughness in the enhanced image data serves to increase confidence of the classification analysis for detecting low-frequency halftone screens, so that such low-frequency halftone areas are not misclassified as contone or high-frequency halftone areas.Type: ApplicationFiled: August 9, 2002Publication date: February 12, 2004Applicant: Xerox CorporationInventor: Xing Li
-
Patent number: 6674915Abstract: A method of adjusting image descriptors when using steerable pyramid to extract features for content-based search is disclosed in this invention. Using steerable pyramid, the original image is filtered to produce Gaussian images and Laplacian images. The image descriptors are formed by filtering the Laplacian images. The filtering of the Laplacian images provides orientation data and energy data. From the orientation data, a first set of global information and local information is generated. From the energy data, a second set of global information and local information is generated. The descriptors are extracted from the respective sets of global information and local information associated with the orientation data and the energy data. The descriptors associated with the orientation data and the descriptors associated with the energy data are combined to form an image descriptor. The Laplacian image may be adjusted prior to the application of the filter.Type: GrantFiled: November 15, 1999Date of Patent: January 6, 2004Assignees: Sony Corporation, Sony Electronics Inc.Inventor: Lizhi Wang
-
Patent number: 6674904Abstract: An edge detection algorithm may use contour tracing followed by boundary detection to improve the accuracy of the detected boundaries and to enable the detection of boundaries in cases where the image is either distorted or subject to the effects of noise. The contour tracing and boundary detection may result in an object identification, which if unsatisfactory, may be retried with different threshold values. A high pass gaussian filter may be applied before tracing contours, resulting in the selection of a band limited set of pixel values which are centered around a value of one half the bit depth.Type: GrantFiled: December 14, 1999Date of Patent: January 6, 2004Assignee: Intel CorporationInventor: Ward A. McQueen
-
Patent number: 6670991Abstract: In an image sensing system where a camera apparatus and a client device are connected for controlling the camera apparatus by the client device, the client device displays and arbitrarily shifts a predetermined size of detection area in an object image sensed by the camera apparatus, enabling to quickly perform camera parameter control, such as focus adjustment or the like, based on image signals of the detection area, with a small amount of transferring data. Moreover, detection areas of a sensed image are synthesized by controlling camera parameters for each of the detection areas so as to optimize camera parameters for all of the areas in the object image.Type: GrantFiled: September 24, 1998Date of Patent: December 30, 2003Assignee: Canon Kabushiki KaishaInventors: Tsuneyoshi Takagi, Yoshihiro Ishida, Takashi Oya
-
Patent number: 6643400Abstract: The present invention provides an image processing apparatus capable of retrieving a specific pattern from an image at a high speed in an image recognizing processing. The apparatus creates a map image having pixels correspond to block regions of the image by one to one when retrieving a specific pattern from the image. Then, the apparatus labels each pixel of the map image based on feature of each block region of the image, detects a predetermined pattern specifying a region on the map image, specifies a block region on the original image corresponding to a position of the detected predetermined pattern on the map image, and retrieves the specific pattern in the specified block region.Type: GrantFiled: March 30, 2000Date of Patent: November 4, 2003Assignee: Minolta Co., Ltd.Inventors: Akira Murakawa, Keisuke Hashimoto
-
Patent number: 6636618Abstract: The invention concerns a method for detecting geological discontinuity in an environment using an optical flow. The method includes the steps of selecting in the seismic block a section containing at least one discontinuity. The section constitutes a seismic image. The method includes computing a rough optical flow on the seismic image to obtain a first representation of an optical flow wherein the discontinuity constitutes a moving front. The method also includes smoothing, by a non-supervised dynamic cluster process, the optical flow representation so as to obtain a second optical flow representation wherein the discontinuity shows accentuated contrasts.Type: GrantFiled: July 15, 1999Date of Patent: October 21, 2003Assignee: Elf Exploration ProductionInventors: Naamen Keskes, Fabien Pauget
-
Publication number: 20030189731Abstract: A print (e.g., on paper) user interface (PUI) device for decoding an embedded message to activate a function is described. The PUI can be used for transforming an embedded message into cognizable human sensory inputs. The device includes a reader and a processor. The reader can sense light from a pattern of pixels on a printed surface and determines the pixel values of the pixels. The pattern of pixels constitutes a foreground visual image that conveys cognizable information to an observer. The primary function of the embedded message need not be to launch a Web-site but to display more information related to the cognizable information.Type: ApplicationFiled: July 1, 2002Publication date: October 9, 2003Inventor: Kenneth H.P. Chang
-
Publication number: 20030185449Abstract: A system and associated method automatically and very rapidly determine the dominant periodicity in sequential binary data. The method incorporates a fast binary autocorrelation function used to discover the dominant periodicity in a sequential binary signal, and exploits this information in a periodicity exploiting application, such as using any one or more of a pixel periodicity or an edge periodicity based compression method in order to compress periodic halftoned bitonal images. The method implements four processes: The first process describes setting up the data in a transition-based representation to facilitate an algorithm to quickly find an autocorrelation distribution derived from a sequential binary signal. The second process describes a fast dominant periodicity search algorithm. The third process describes an algorithm that estimates the dominant window for estimating the dominant periodicity within one line of two-dimensional sequential binary signal.Type: ApplicationFiled: March 28, 2002Publication date: October 2, 2003Applicant: International Business Machines CorporationInventors: Ronald Barthold Arps, Cormeliu Mihail Constantinescu
-
Patent number: 6621937Abstract: A method for removing noise on a pixel by pixel basis from pixels of a digital image is disclosed. The pixels are used to produce a map of features, which then are used to produce a variable shape neighborhood region of cleaning pixels with respect to the original value of the pixel of interest. The neighborhood region of cleaning pixels and the value of the pixel of interest are then used to change the original value of the pixel of interest in the digital image so that it has been noise cleaned.Type: GrantFiled: October 8, 1999Date of Patent: September 16, 2003Assignee: Eastman Kodak CompanyInventors: James E. Adams, Jr., John F. Hamilton, Jr.
-
Patent number: 6603877Abstract: A novel method all of and apparatus for categorizing different material regions of an object-to-be inspected, by optical scanning of the object to produce a pixel image thereof, discriminating different pixel regions of the image corresponding to possibly different materials on the basis of color, and texture brightness measurements of the pixel regions and assigning preliminary likelihoods to such discrimination through with ambiguities; and comparing the measurements of the pixel regions with their local neighborhood surroundings in the image to assist in resolving said ambiguities and determining the material categorizations of the pixel regions with a high likelihood.Type: GrantFiled: June 1, 1999Date of Patent: August 5, 2003Assignee: Beltronics, Inc.Inventor: Robert Bishop