Median Filter Patents (Class 382/262)
  • Patent number: 6486909
    Abstract: The purpose of this invention is a process and a device operating in real time for identifying and localizing an area in relative movement in a scene and determining the speed and oriented direction of this relative movement in real time. The process carries out temporal processing of the digital video input signal S(PI) that consists in deducing, from the variations between the value of each pixel in a frame and in the immediately preceding frame, a binary signal DP identifying whether or not there was a significant variation and a digital signal CO representing the magnitude of this variation, and a spatial processing consisting in distributing in a matrix these two signals successively for a single frame that is scanned through the matrix, and deducing the required relative movement and its parameters from this matrix distribution.
    Type: Grant
    Filed: September 13, 1999
    Date of Patent: November 26, 2002
    Assignee: Holding B.E.V.
    Inventor: Patrick Pirim
  • Publication number: 20020159648
    Abstract: A method and apparatus for processing image data is described, comprising acquiring a frame of image data and compressing the dynamic range of the frame of image data using a dynamic range compression (DRC) algorithm that utilizes down-sampling, median filtering, and up-sampling. The DRC algorithm comprises down-sampling a frame of image data comprising a first array of pixels to generate a second array of pixels, applying a first median filter to the second array of pixels to generate a blurred array of pixels, up-sampling the blurred array of pixels, and removing at least a portion of low-frequency gradient data generated by previous steps from the frame of image data.
    Type: Application
    Filed: April 25, 2001
    Publication date: October 31, 2002
    Inventors: Timothy Alderson, Gene D. Tener, Valete C. Hopkins, Charlie W. Stuart, Chinh X. Nguyen
  • Publication number: 20020159649
    Abstract: A method and system for reducing correlated noise in imagers is disclosed. Methods are described for determining a pixel correction value. Correlated noise in data generated by imagers can be reduced by applying the pixel correction value to adjust image data before being displayed.
    Type: Application
    Filed: February 22, 2001
    Publication date: October 31, 2002
    Applicant: Varian Medical Systems, Inc.
    Inventors: Ivan P. Mollov, Richard E. Colbeth, Christopher Webb, Jiann M. Yu
  • Publication number: 20020136453
    Abstract: The present invention relates to a method and system for modifying images characterized in that it includes the ability to assign colour intensity values to pixels exposed during image manipulation operations comprising:
    Type: Application
    Filed: February 13, 2001
    Publication date: September 26, 2002
    Applicant: International Business Machines Corporation
    Inventors: Gaurav Aggarwal, Sugata Ghosal
  • Patent number: 6438270
    Abstract: Image sharpening methods and systems sharpen the edges of objects within images by comparing and replacing the pixel value for a pixel with the median of the original value of the pixel and two values obtained by linear extrapolation of one or more neighboring values on each side of the pixel. By performing the process for all or a plurality of the pixels of a stored image, objects within the image, when output, on a video monitor, paper or other display media, will be provided with sharpened edge characteristics. Thus, the displayed image will include a reduced quantity of edge ringing and a reduced quantity of noise amplification.
    Type: Grant
    Filed: March 3, 1999
    Date of Patent: August 20, 2002
    Assignee: Xerox Corporation
    Inventor: Steven J. Harrington
  • Publication number: 20020094130
    Abstract: Noise filtering an image sequence (V1) is provided wherein statistics (S) in at least one image of the image sequence (V1) is determined (11) and at least one filtered pixel value (Pt′) is calculated from a set of original pixel values (Pt, Mi) obtained from the at least one image, wherein the original pixel values (Pt, Mi) are weighted (13) under control (12, &agr;) of the statistics (11).
    Type: Application
    Filed: June 13, 2001
    Publication date: July 18, 2002
    Inventors: Wilhelmus Hendrikus Alfonsus Bruls, Leonardo Camiciotti, Gerard De Haan, Richard Petrus Kleihorst, Albert Van Der Werf
  • Patent number: 6384864
    Abstract: A letter-box filtering circuit and a method of using the same that includes a preprocessing filter circuit and a filtering circuit to reduce or correct the round-off errors generated when displaying an image of 16:9 on a screen of 4:3 in a digital image processing. The letter-box filtering circuit can display the down sampled image based on the 16:9 image without any distortion or reduced distortion.
    Type: Grant
    Filed: July 24, 1998
    Date of Patent: May 7, 2002
    Assignee: Hyundai Electronics Industries Co., Ltd.
    Inventor: Young-No Kim
  • Patent number: 6360025
    Abstract: This invention relates to an image processing method for filtering pixels of an image in a temporal sequence of successive images. At each pixel of the images, spatially smoothed intensities are determined, and the spatially smoothed intensities of a center image are tested relative to the smoothed intensities of its preceding and its succeeding images. If a spatially smoothed central intensity at a pixel substantially deviates from the intensities in the preceding and succeeding images at that pixel, the filtered intensity of that central pixel is assigned as the spatially smoothed intensity of the center image. If not, the filtered intensity of that pixel is assigned as a further temporal filtering of the three spatially smoothed intensities. The invention also relates to an apparatus, in particular a medical x-ray imaging apparatus, for carrying out the image processing method.
    Type: Grant
    Filed: December 22, 1998
    Date of Patent: March 19, 2002
    Assignee: U.S. Philips Corporation
    Inventor: Raoul Florent
  • Publication number: 20010051006
    Abstract: A system to minimize the number of comparison and rearrangement operations and increase the processing speed with a small circuit scale utilizing a median filter in the form of hardware. A signal processing apparatus for performing, when a region to be subjected to signal processing has shifted, signal processing of pixel signals in a region to be subjected to signal processing after shifting the region, by utilizing old pixel signals already subjected to signal processing before the shift of the region. The apparatus includes a new-data-order determination circuit for determining an order of magnitude of pixel signals newly added to a region to be subjected signal processing after the shift of the region. A new-data selection circuit rearranges the new pixel signals based on the determined order. An all-data-order determination circuit determines the order of magnitude of all signals by comparing the rearranged new pixel signals with the old pixel signals.
    Type: Application
    Filed: December 6, 2000
    Publication date: December 13, 2001
    Inventor: Akihiko Yushiya
  • Patent number: 6330372
    Abstract: In one embodiment, a method comprises the detection of an edge pixel in an unsharpened image. The edge pixel is sharpened to form a sharpened image. Additionally, a region of the sharpened image is selected such that the region includes the edge pixel. The region of the sharpened image is compared to a corresponding region of the unsharpened image to form a difference value. The difference value is then processed to form a final image.
    Type: Grant
    Filed: September 13, 1999
    Date of Patent: December 11, 2001
    Assignee: Intel Corporation
    Inventors: Judith A. Goldstein, Thomas P. O'Rourke, Rajeeb Hazra
  • Publication number: 20010048771
    Abstract: An image processing method is provided, which makes it possible to enlarge an input or original image in such a way as to suppress or eliminate jaggies at the contours in the image, and which is applicable to a dot-matrix type display device with a fixed resolution. In the step (a) a local area is defined in an input image. The local area includes a target pixel and neighboring pixels surrounding the target pixel. In the step (b), a filtering operation is applied to the target pixel and the neighboring pixels in the local area while all the pixels in the local area are successively assigned to the target pixel, thereby outputting a filtered image. In the step (c), the filtered image and the input image are mixed together at a specific mixing ratio, thereby forming an output image. In the step (b), the median or average value is preferably used for forming the filtered image.
    Type: Application
    Filed: May 24, 2001
    Publication date: December 6, 2001
    Applicant: NEC CORPORATION
    Inventor: Daigo Sasaki
  • Patent number: 6295091
    Abstract: Methods of interpolating missing pixels between interlaced scan lines is disclosed. A first method interpolates a desired pixel by selecting the median pixel from a first pixel above and to the left of the desired pixel, a second pixel above the desired pixel, a third pixel above and to the right of the desired pixel, a fourth pixel below and to the left of the desired pixel, a fifth pixel below the desired pixel, a sixth pixel below and to the right of the desired pixel, and a seventh pixel from the previous video field having same position of the desired pixel.
    Type: Grant
    Filed: November 7, 1997
    Date of Patent: September 25, 2001
    Assignee: Intel Corporation
    Inventor: Samson Huang
  • Publication number: 20010019633
    Abstract: The invention provides noise filtering (3) of a signal (x) by estimating (30) a type of noise in the signal (x) and enabling (30) one of at least two noise filters (310,311,312), the enabled noise filter (310,311,312) being a most suitable filter for the estimated type of noise.
    Type: Application
    Filed: January 11, 2001
    Publication date: September 6, 2001
    Inventors: Livio Tenze, Sergio Carrato, Stefano Olivieri
  • Publication number: 20010017943
    Abstract: An image filter circuit which realizes a large-scale median filter as a digital circuit which has not been conventionally constructed without difficulty. The filter circuit has a histogram A update unit 102 which generates a histogram A with upper-bit data of pixels within a matrix, a cumulative adder A104 which searches for a class including a pixel value corresponding to a predetermined rank in ascending order as a target class based on the histogram A, a histogram B update unit 106 which generates a histogram B with lower-bit data of pixels included in the target class, and a cumulative adder B108 which searches for a pixel value of the predetermined rank in ascending order in the matrix. The filter circuit having these constituent elements outputs a median pixel value of the matrix.
    Type: Application
    Filed: February 26, 2001
    Publication date: August 30, 2001
    Inventor: Katsumi Otsuka
  • Patent number: 6219101
    Abstract: A VGA to analog video converter is useful e.g. for displaying video and/or graphics data from a computer onto a large screen television or television monitor. The RGB video signals output from the personal computer are first converted to digital form. The analog-to-digital converter which does this is clocked by a clock signal generated by a phase-locked loop using the horizontal synchronizing signal from the personal computer. The digital RGB signals are then converted to a YCbCR format. A flicker filter eliminates the flickering appearing on the TV monitor by operating on the luminance (Y) component. The YCbCr signals are encoded into NTSC or PAL Standard, and output in composite analog video or S-VHS format. A color subcarrier synthesizer generates the color subcarrier signal to generate an accurate subcarrier frequency for the video output signals. An analog-to-digital clock phase adjustment is used to ensure that the input RGB signals are sampled at the proper instant by the analog-to-digital converters.
    Type: Grant
    Filed: May 15, 1998
    Date of Patent: April 17, 2001
    Assignee: Fairchild Semiconductor Corporation
    Inventors: Mehdi H. Sani, De Dzwo Hsu, Willard K. Bucklen
  • Patent number: 6208764
    Abstract: A rank filter suitable for use in real-time signal processing and image processing applications, for example, for improving an image. The rank filter can be implemented in software or hardware, employed in single- and multi-dimensions, and within any specified window. A link-list algorithm is combined with a memory array for storing and ordering the data.
    Type: Grant
    Filed: August 20, 1998
    Date of Patent: March 27, 2001
    Assignee: Eastman Kodak Company
    Inventors: Greg L. Archer, Larry W. Wilkinson
  • Patent number: 6199084
    Abstract: Methods and apparatus for efficiently implementing weighted median filters are described. In a first hardware embodiment, the median filter is implemented using a number of storage locations which is equal to or greater than the sum of the weight values used to implement the filtering operation. In other embodiments, the number of storage locations required to store data values is reduced to equal or approximately equal the number of data values, used in the filtering operation. In such embodiments, the data values are sorted according to size and a filter weight is associated with each data value, e.g., as part of a data record created for each one of the data values. The data value to output, as the result of the filtering operation, is determined from the stored weight values.
    Type: Grant
    Filed: September 9, 1998
    Date of Patent: March 6, 2001
    Assignee: Hitachi America, Ltd.
    Inventor: John Wiseman
  • Patent number: 6195132
    Abstract: A noise reduction signal processing apparatus for reducing noises accurately, such as for use in a display apparatus for displaying a video signal, has a median filter (40) which receives the video signal, and which executes a filter processing on the inputted video signal and outputs a reference signal; a subtracter (50) which is connected with the median filter (40), and which outputs a difference signal that indicates a difference between a reference signal outputted from the median filter (40) and the video signal; a minimum value detection circuit (70) which outputs the difference signal from the subtracter (50) or a limitation value, whichever is smaller, as a minimum value signal; and an adder (35) which adds a noise reduction signal obtained on the basis of the minimum value signal output from the minimum value detection circuit (70) and the video signal.
    Type: Grant
    Filed: July 1, 1999
    Date of Patent: February 27, 2001
    Assignee: Hitachi, Ltd.
    Inventors: Katsunobu Kimura, Takaaki Matono, Haruki Takata, Tatsuo Nagata, Masato Sugiyama, Yasutaka Tsuru, Koichi Sudo
  • Patent number: 6163324
    Abstract: A method of finding median values from different sets of values includes a step of forming a plurality of bit registers. Each bit register has bit groups corresponding respectively to the different sets of values, and each bit group has bit positions corresponding respectively to different individual ones of the values of the corresponding set. Bits in different bit registers at bit positions corresponding to a particular one of the given values indicate the truth or falsity of different possible relationships between the particular given value and other individual ones of the given values in the same set. The method includes a further step of logically ANDing different combinations of the bit registers to form a plurality of parallel condition registers having bit positions corresponding respectively to different individual ones of the given values of the different sets.
    Type: Grant
    Filed: June 30, 1998
    Date of Patent: December 19, 2000
    Assignee: Microsoft Corporation
    Inventor: Marc Holder
  • Patent number: 6137917
    Abstract: An apparatus and method of reducing impulse noise using a median filter involve selecting a pixel of median.sub.-- value from a plurality of pixels, comparing a reference range, determined by adding or subtracting an offset value to or from the selected median.sub.-- value, with a mid.sub.-- pixel among the plurality of pixels, to determine whether the input image signal has impulse noise and generate a noise coefficient depending on the presence of impulse noise, calculating a correlation between the luminance component of a pixel and the color component thereof depending on the noise coefficient, and cross-fading the pixel of median.sub.-- value and the mid.sub.-- pixel depending on the correlation.
    Type: Grant
    Filed: August 13, 1998
    Date of Patent: October 24, 2000
    Assignee: Samsung Electronics Co., Ltd.
    Inventor: Se-woong Park
  • Patent number: 6122404
    Abstract: A visible Stokes polarimetric imager (80) that separately and contemporaneously measures each of the four separate Stokes polarization parameters of a visible light beam reflected from a scene. The imager (80) includes a compound prism assembly (50) including five prism elements (52-58). The reflected beam from the scene is collected by an input lens (84) in the imager (80), and is directed to the prism assembly (50) at a certain location. The prism elements (52-58) split the beam (82) into four separate beams (62-68) of substantially equal intensity that are emitted from the prism assembly (50) in different directions. Four separate polarimetric filters (86) and imaging devices (88) are positioned relative to the prism assembly (50) to separately receive each of the four beams (62-68) emitted from the prism assembly (50).
    Type: Grant
    Filed: May 28, 1998
    Date of Patent: September 19, 2000
    Assignee: TRW Inc.
    Inventors: James D. Barter, Peter H. Y. Lee
  • Patent number: 6078213
    Abstract: A hardware apparatus for determining a median sample from n input samples is disclosed. The hardware apparatus includes n(n-1)/2 comparator circuits. The n(n-1)/2 comparator circuits compare each input sample with every other input sample. The comparator circuits are coupled to n-1 median selector circuits. Each median selector circuits selects a particular input sample as the median sample if the number of greater-than or equal-to comparisons for that particular input sample equals the number of less-than comparisons for that particular input sample.
    Type: Grant
    Filed: November 7, 1997
    Date of Patent: June 20, 2000
    Assignee: Intel Corporation
    Inventor: Samson Huang
  • Patent number: 6075903
    Abstract: The invention concerns a process to correct the intensity of images from a digital infrared camera with a two-dimensional detector. First a stationary correction is made at a reference source to determine correction coefficients. For the stationary correction, an average intensity characteristic curve U.sub.av (T) is determined for the detector and compared with the intensity characteristic curve U.sub.j (T) for each image point j of the two-dimensional detector to determine correction coefficients for each image point j. A non-linear formulation is used to adapt the intensity characteristic curve U.sub.j (T) of each image point to the average intensity characteristic curve U.sub.av (T). Correction coefficients are stored in a memory to correct the individual intensity U.sub.j of each image point j during operation. During operation, the correction coefficients are improved dynamically. The intensity Uj of an image is recorded and the intensity Uj of the image is filtered by a locally acting adaptive filter.
    Type: Grant
    Filed: April 16, 1998
    Date of Patent: June 13, 2000
    Assignee: AEG Infrarot-Module GmbH
    Inventors: Rainer Breiter, Wolfgang Cabanski
  • Patent number: 6058405
    Abstract: One way of filtering graphics images to remove pixel dropout and shot noise while preserving overall image quality is to use non-linear filters based on rank order of an M.times.N grid of pixel values surrounding a pixel to be filtered. This rank order filter method for M.times.N grids first sorts columns, then sorts rows, and finally sorts diagonals to obtain either just a median filter value, a rank-based filter value, or a complete sorted order of the M.times.N grid. All of the sort operations are control-flow-free allowing implementation on SIMD processors processing multiple overlapping M.times.N grids in parallel. Advantage is taken of columns and rows that are shared among the horizontally and vertically overlapping M.times.N grids in the digital image to reduce the total number of sort operations for the image.
    Type: Grant
    Filed: November 6, 1997
    Date of Patent: May 2, 2000
    Assignee: Motorola Inc.
    Inventors: Priyadarshan Kolte, Roger Smith
  • Patent number: 6041145
    Abstract: A great decrease of a step, i.e., block distortion, of such as unnatural, abnormal or artifact brightness and color, on a block boundary, leads to much visual improvement of picture quality. Against a picture that has been encoded and decoded by block, the smoothing filter characteristics is changed according to quantizing parameter, block activity of a decoded picture, the activity of each pixel of the decoded picture, and the position with respect to a block boundary of a pixel. A smoothing filter utilizes a weighted mean whose weight is the inverse number of a difference between a target pixel and its surrounding pixels. The weighted mean is obtained by adding a certain number to a weight before taking its inverse number. The value to be added is controlled by the parameter. Similarly, the mixing ratio of a pixel value after performing the smoothing filter and a decoded pixel value is changed depending on the parameter.
    Type: Grant
    Filed: July 15, 1998
    Date of Patent: March 21, 2000
    Assignee: Matsushita Electric Industrial Co., Ltd.
    Inventors: Takaaki Hayashi, Minoru Etoh
  • Patent number: 6028966
    Abstract: An image reading apparatus wherein the original image to be read by main scanning is prescanned first to obtain reading parameters optimum for the main scanning, is characterized in that the apparatus processes image data picked up at predetermined dots from among all image data by a median filter during the pre-scanning, and processes all image data by the median filter during the main scanning.
    Type: Grant
    Filed: March 28, 1996
    Date of Patent: February 22, 2000
    Assignee: Minolta Co., Ltd.
    Inventor: Yoshiyuki Hashimoto
  • Patent number: 5974194
    Abstract: The present invention is directed to a method of scratch removal from digitized images wherein a "scratched" portion of a digitized image which requires repair or noise removal, is identified and a binary mask is generated which distinguishes the defined "scratched" portion from the other portions of the digitized image. A repair window is defined which delineates the portion of the digitized image desired to be repaired of scratches or other noise. A sample window having the same dimensional attributes as the repair window, and preferably free of any scratch or other noise, is then defined. The sample window is preferably selected so as to be as close in resemblance to the repair window as possible.
    Type: Grant
    Filed: October 21, 1996
    Date of Patent: October 26, 1999
    Assignee: Sony Corporation
    Inventors: Anil Hirani, Takashi Totsuka
  • Patent number: 5974192
    Abstract: Blocks in frames of a video are matched to similar blocks across frames using qualitative characteristics of the pixels in the block. Pixels in a frame are categorized into two or more categories. In one embodiment, three pixel categories are: 1) quasi-constant (QC) pixels having intensity values that vary slowly (e.g., pictorial portions), 2) textured pixels characterized by many small edges and thin-line signals (e.g., texture portions), and 3) dominant-edge (DE) pixels characterized by few sharp and dominant edges (e.g., edge portions). Blocks in different frames are matched by minimizing a cost function that measures the mismatch between a block in a first frame and a plurality of candidate blocks in a second frame. In one embodiment, a cost function comprises a contrast component and a brightness component. The brightness component is determined as a function of the pixel categories. Block matching in sequential frames of a video enables a calculation of displacement or motion estimation in the video.
    Type: Grant
    Filed: May 6, 1996
    Date of Patent: October 26, 1999
    Assignees: U S West, Inc., MediaOne Group, Inc.
    Inventor: Amlan Kundu
  • Patent number: 5953464
    Abstract: A method and apparatus for achieving an ultra-small or compressed image buffer images at half the resolution and then scales by two to achieve the device resolution. Acceptable quality can be maintained by identifying edge and interior portions of the page image and using this information to scale intelligently. A split-level frame buffer provides this identification of the image components. Further, an extension of block-truncation coding can be used with the split-level frame buffer to provide up to a 16 to 1 compression for an overall compression of up to 64 to 1. Actual techniques to scale these encodings are described along with new encodings for graphics and text designed this high compression of the image.
    Type: Grant
    Filed: September 23, 1997
    Date of Patent: September 14, 1999
    Assignee: Xerox Corporation
    Inventor: Steven J. Harrington
  • Patent number: 5949919
    Abstract: A precompression extrapolation method extrapolates image features of arbitrary configuration to a predefined configuration to facilitate compression of the image features. This extrapolation method allows compression of arbitrary image features to be performed in a conventional manner, such as discrete cosine transform or lattice wavelet compression. This method includes defining an extrapolation block boundary about an object or image feature that is bounded by a feature boundary. Pixels within the block, but not included in the image feature, are identified as non-feature pixels. The feature pixels in the feature boundary include pixel values that correspond to the image characteristics of the pixels. The non-feature pixels adjacent the pixels of the feature boundary are assigned pixel values that include the pixel values of the adjacent feature boundary pixels.
    Type: Grant
    Filed: June 4, 1996
    Date of Patent: September 7, 1999
    Assignee: Microsoft Corporation
    Inventor: Wei-ge Chen
  • Patent number: 5917733
    Abstract: An image and pulse analysis system and method including a datasieve comprising a succession of ordinal filtering means receiving at least a selected part of a signal conglomerate and filtering the at least a selected part of the signal conglomerate to produce a bandpass filter output, the ordinal filtering means having window lengths increasing from N (less than M) up to M, whereby an input signal to each successive filtering means is formed from the output of the previous filtering means; selection means for selecting a predetermined subset of bandpass filter outputs from outputs of the succession of filtering means; and arithmetic means for adding together the signals of the subset to produce an output signal containing only wanted signals determined by the selection means; data storage means for storing data relating to a sought-for-feature or object which may exist in a field from which data is derived for presenting to the datasieve; comparison means for matching or comparing the stored data with the out
    Type: Grant
    Filed: June 30, 1995
    Date of Patent: June 29, 1999
    Assignee: Cambridge Consultants Limited
    Inventor: James Andrew Bangham
  • Patent number: 5896468
    Abstract: Digital maps digitized by scanning paper maps are smoothed by identifying regions which should be of a uniform color, empirically selecting sets of colors which the pixels of a given region might have in the raw map, empirically determining the statistical probability of occurrence of pixels of each color of the selected set in the region, defining groups of connected pixels having colors within the selected set, and resetting the color of all the pixels in a group to a predetermined region color if, and only if, the pixels of the group satisfy the determined statistical probability of occurrence. The inventive method provides uniform coloring of the map where desired, without affecting the sharpness of small-print map annotations.
    Type: Grant
    Filed: November 27, 1995
    Date of Patent: April 20, 1999
    Assignee: Nortrhop Grumman Corporation
    Inventor: Stephen V. Kowalski
  • Patent number: 5880767
    Abstract: The present invention provides an effective, low cost method and system for enhancing various types of images including photograph, CD, video, and graphic art images. The method of enhancing the input image, includes the steps of: filtering the input image to extract m different frequency components r.sub.k ; adaptively sharpening the m different frequency components r.sub.k, where the amount of sharpening for each component r.sub.k corresponds to a sharpening function g.sub.k ?r.sub.k !; and adding the adaptively sharpened m different frequency components g.sub.k ?r.sub.k ! to the input image. Because the sharpening function is typically nonlinear, the step of determining the value of the adaptive frequency component corresponding to the sharpening function is achieved by mapping the filtered component by the corresponding sharpening function.
    Type: Grant
    Filed: September 11, 1996
    Date of Patent: March 9, 1999
    Assignee: Hewlett-Packard Company
    Inventor: Samson J. Liu
  • Patent number: 5832118
    Abstract: An apparatus for classifying a textured image based on pattern coarseness and directivity comprises: a quantizer for obtaining a quantized image from the textured image, the quantized image containing a plurality of pixels each of which is represented by one of N quantized values, N being a positive integer; a scanning block for scanning the quantized image along M scanning directions, M being a positive integer, to thereby provide M scanned images; a grey level mean runlengths configuration block for providing a set of runlengths by counting runlengths of pixels having a same quantized value, for each of the M scanned images and each of the N quantized values, to thereby provide M x N sets of runlengths, providing M.times.N mean runlengths by averaging each set of runlengths, and forming an M.times.
    Type: Grant
    Filed: October 21, 1996
    Date of Patent: November 3, 1998
    Assignee: Daewoo Electronics Co., Ltd.
    Inventor: Jin-Hun Kim
  • Patent number: 5832134
    Abstract: The present invention provides a system for performing fast segmentation and image processing for enhanced three dimensional (3-D) visualization of a subject. The present invention automatically extracts unwanted dominant features from images while preserving the desired information. The present invention works especially well with Computed Tomography Angiograms (CTA) for viewing vasculature of a subject's head. Segmenting and removing dominant structures from image data permits visualization techniques such as maximum intensity projection (MIP), surface rendering and volume rendering, to provide enhanced vessel visualization.
    Type: Grant
    Filed: November 27, 1996
    Date of Patent: November 3, 1998
    Assignee: General Electric Company
    Inventors: Gopal Biligeri Avinash, Abdalmajeid Musa Alyassin
  • Patent number: 5798846
    Abstract: An image processing arrangement is described that minimizes moire patterns in scanned halftone images of a scanned image while maintaining the resolution of the scanned image. The arrangement includes a filtering processor that has a modified median filter with a specially shaped (e.g., cross-shaped) filter window. The filter window allows the modified median filter to preserve scanned text images of the scanned image when processing the scanned halftone images to minimize the moire patterns. In addition, the filtering processor determines if an image pixel of the scanned image appears to belong to scanned continuous tone images of the scanned image prior to processing the image pixel. The filtering processor does not process that image pixel and the pixel value of that image pixel is left unchanged if the image pixel appears to belong to the scanned continuous tone images.
    Type: Grant
    Filed: May 28, 1996
    Date of Patent: August 25, 1998
    Assignee: Hewlett-Packard Company
    Inventor: Daniel R. Tretter
  • Patent number: 5793899
    Abstract: A coprocessor in an image processing system is coupled to the bus to which a CPU and RAM holding image data are also coupled. The coprocessor extracts an input pixel stream corresponding to input images from selected bus transactions, performs computations on the input stream to produce output pixel streams corresponding to output images, and inserts the output pixel streams into selected CPU-to-memory bus transactions so that the memory stores the data. The CPU generates the selected bus transactions with specially marked address and/or control signals. The coprocessor includes a lookup table, and a first row delay. The row delay accumulates the three most recent rows of input pixels, which are sent to Sobel and rank processing sections for neighborhood processing. The results are thresholded and formatted, and are either output directly or passed through an additional pair of row delays to accumulate three rows of result data for neighborhood peak detection.
    Type: Grant
    Filed: December 5, 1995
    Date of Patent: August 11, 1998
    Assignee: Cognex Corporation
    Inventors: Robert Anthony Wolff, Steven Mark Rosenthal, William Michael Silver, Jean-Pierre Schott
  • Patent number: 5793900
    Abstract: A method is disclosed for generating a categorical depth map of a scene using passive defocus sensing. In a preferred embodiment three synchronized CCD cameras focused at different distances detect three images of the same scene. An image processor partitions the images into an array of regions and calculates a sharpness value for each region. The sharpness value for a region is calculated by summing over all pixels (x,y) in the region the absolute difference in the intensity value of a pixel (x,y)( with pixel (x-k,y-l), where k and l are constants. The image processor then constructs a depth map of the scene by determining for each region the image with the greatest sharpness in that region. An application of the invention to a mobile robot control system is described in detail. Among other applications, the method may be used for collision avoidance, object detection, and speed measurement.
    Type: Grant
    Filed: December 29, 1995
    Date of Patent: August 11, 1998
    Assignee: Stanford University
    Inventors: Illah Reza Nourbakhsh, David Andre
  • Patent number: 5784499
    Abstract: In an image data including tone information, a level of an attentional picture element in an image and a level of an adjacent picture element located adjacent to the attentional picture element are smoothed to create a first substitution level, and by internally dividing the level of the attentional picture element and the first substitution level with a predetermined internal division ratio to obtain a second substitution level which finally substitutes the level of the attentional picture element.
    Type: Grant
    Filed: June 18, 1997
    Date of Patent: July 21, 1998
    Assignee: Matsushita Electric Industrial Co., Ltd.
    Inventors: Yasuhiro Kuwahara, Haruo Yamashita, Tsumoru Fukushima
  • Patent number: 5732159
    Abstract: A blocking-effect eliminating circuit in an image coder/decoder employs a weighted value according to the location of each pixel and the quantization step size in a block constituting an input image, to thereby reduce the blocking effect generated in a received image and enhance image quality.
    Type: Grant
    Filed: April 24, 1996
    Date of Patent: March 24, 1998
    Assignee: Samsung Electronics Co., Ltd.
    Inventor: Seok-yoon Jung
  • Patent number: 5712807
    Abstract: A pulse analysing system for separating broadband wanted pulses of duration to include a predetermined plurality of data samples from a signal which also contains unwanted short signals, in which a data-sieve comprising a succession of ordinal signals of increasing value is employed in conjunction with pulse selection and addition of signals of a selected subset obtained from the data-sieve.
    Type: Grant
    Filed: December 13, 1994
    Date of Patent: January 27, 1998
    Inventor: James Andrew Bangham
  • Patent number: 5701163
    Abstract: A video processing method and apparatus for processing video frames comprising animated images so as to blur movement of the animated images between successive frames, the method or apparatus comparing a first input frame with the following input frame; identifying moving areas in the first input frame by the comparison; calculating the velocity of the movement of the moving areas between the first and following input frame; summing, only for each input pixel site within the identified moving areas of the input frame, the pixel value of each input pixel site and adjacent pixel values determined according to the calculated velocities and a predetermined aperture type for the input frame; and providing an output frame from at least the summed pixel values and pixel values of areas of the input frame other than the moving area, the adjacent pixel values either being the pixel values of pixel sites in the input frame which are spatially adjacent in the direction of the velocity or predicted temporally adjacent pi
    Type: Grant
    Filed: December 18, 1995
    Date of Patent: December 23, 1997
    Assignees: Sony Corporation, Sony United Kingdom Limited
    Inventors: John William Richards, Martin Rex Dorricott
  • Patent number: 5694492
    Abstract: A post-processing method and apparatus is capable of improving the image quality of a decoded image signal by effectively reducing or removing a blocking effect in the decoded image signal. Decoded image data is first stored in a memory and target pixel data is filtered. Thereafter, an absolute difference value between original target pixel data and the filtered target pixel data is calculated and compared with a predetermined threshold value.
    Type: Grant
    Filed: May 1, 1995
    Date of Patent: December 2, 1997
    Assignee: Daewoo Electronics Co., Ltd
    Inventor: Sang-Ho Kim
  • Patent number: 5559902
    Abstract: The present invention provides a method and apparatus for enhancing and recognizing connected and degraded text. The enhancement process comprises filtering a scanned image to determine whether a binary image value of an image pixel should be complemented, determining whether complementing the value of the pixel reduces the sharpness of wedge-like figures in the image, and complementing the binary value of the pixel when doing so does not reduce sharpness. The recognition process may comprise determining primitive strokes in a scanned image, segmenting the scanned image into sub-character segments based on the primitive strokes, identifying features which characterize the sub-character segments, and comparing identified features to stochastic models of known characters and determining an optimum sequence of known characters based on the comparisons through the use of Viterbi scoring and level building procedures.
    Type: Grant
    Filed: May 31, 1994
    Date of Patent: September 24, 1996
    Assignee: Lucent Technologies Inc.
    Inventors: Chinmoy B. Bose, Shyh-Shiaw Kuo
  • Patent number: 5555028
    Abstract: A post-processing filter apparatus for use in an image signal decoding system, capable of improving the image quality of the system, comprises a memory for storing decoded image data of a current frame; a filter for filtering target pixel data in the stored decoded image data to thereby produce filtered target pixel data, wherein the target pixel data represents the value of a pixel to be filtered; a calculator for calculating an absolute difference value between an original target pixel data and the filtered target pixel data; a first circuit for selectively generating a first or a second threshold value based on the decoded image data and the target pixel data; a second circuit for updating the stored target pixel data with the filtered target pixel data if the absolute difference value is smaller than the generated threshold value; and a third circuit for sequentially updating the stored target pixel value with corrected target pixel data T times if the absolute difference value is equal to or larger than
    Type: Grant
    Filed: June 29, 1995
    Date of Patent: September 10, 1996
    Assignee: Daewoo Electronics Co., Ltd.
    Inventor: Sang-Ho Kim
  • Patent number: 5555029
    Abstract: A post-processing filter apparatus for use in an image signal decoding system, capable of improving the image quality of the system, comprises a memory for storing decoded image data of a current frame; a controller for assigning each pixel value included in the stored decoded image data as a target pixel value in sequence, the target pixel value representing the value of a target pixel to be filtered, and for producing information representing the position of the target pixel; a pixel region determinator for generating a selection signal representing whether or not the target pixel belongs to a boundary region of a block in the stored decoded image data by using the position information of the target pixel, wherein the boundary region denotes a region which includes pixels placed along the boundary of the block; a filtering device for selectively filtering the target pixel value in response to the selection signal to thereby produce a filtered target pixel value; a updating circuit for updating the target pi
    Type: Grant
    Filed: July 28, 1995
    Date of Patent: September 10, 1996
    Assignee: Daewoo Electronics Co., Ltd.
    Inventor: Sang-Ho Kim
  • Patent number: 5495538
    Abstract: A method for reducing ringing and blocking artifacts in a decompressed image models an image in a relatively small area as several smooth regions separated by edges. The method uses JPEG MxM pixel blocks and is compatible with JPEG decompression. To reduce ringing, a block is examined for uniformity, segmented and smoothed. Then, after a DCT transform, a projection is performed to guarantee that the DCT coefficients of the resulting image block will be within the initial quantization interval. The resultant image is produced by an inverse DCT. To reduce blocking, the method is modified to employ a large outer window for uniformity checking, segmentation and smoothing and a small inner window for DCT projection.
    Type: Grant
    Filed: March 8, 1995
    Date of Patent: February 27, 1996
    Assignee: Xerox Corporation
    Inventor: Zhigang Fan