Abstract: In the multi-modal interface apparatus of the present invention, a gaze object detection section always detects a user's gaze object. The user inputs at least one medium of sound information, character information, image information and operation information through a media input section. In order to effectively input and output information between the user and the apparatus, a personified image presentation section presents a personified image to the user based on the user's gaze object. A control section controls a reception of the inputted media from the media input section based on the user's gaze object.
Abstract: A computer implemented method may be used in classifying and identifying finger prints. Dab prints are analyzed according to intersections with fiducial lines. A two-dimensional Hidden Markov Model is then used to correspond states of data based on the intersections.
November 12, 1997
Date of Patent:
September 12, 2000
International Business Machines Corporation
Abstract: The object is to improve the accuracy of signature input and signature verification and improve security by causing the handwriting of a signature to disappear by itself. The signature verification system includes a signature input apparatus 1 and a signature verifying apparatus 6. Signature input apparatus 1 has an externally charged type liquid crystal sheet 2 which has an underside being adhesive and is placed on a coordinate input unit 3. When a dedicated pen 5 having the function of applying static charge onto this liquid crystal sheet 2 is used to write a signature, the handwriting of the signature is liquid crystal representation on the display area of liquid crystal sheet 2 and this handwriting of the signature will disappear by itself after a predetermined period of time.
March 23, 1998
Date of Patent:
September 12, 2000
Kabushiki Kaisha Pilot
Akinobu Izuno, Yutaka Abe, Masahiko Ikeda, Morio Sato
Abstract: Methods and apparatus for compressing and decompressing multiple data sets (e.g., multiple color planes in a color image) in which interdependencies between the data sets, as well as spacial correlation within each data set, are used to create a plurality of compressed output data sets wherein one of the compressed output data sets includes data relating to all of the original uncompressed data sets and wherein each of the remaining output data sets includes data relating only to a particular one of the original uncompressed data sets. Advantageously, the disclosed techniques provide compression ratios approaching those of known vector compression algorithms while simultaneously providing decompression memory usage characteristics analogous to those found in planar compression systems. Thus, exemplary embodiments of the invention simultaneously provide relatively low peak and average memory requirements, corresponding to relatively low memory cost and relatively high data throughput, respectively.
Abstract: A DCT based lossy compression method for compressing a digitized image composed of a matrix of image samples to provide a compressed image which satisfies a predefined bit budget. The digitized image is first sub-divided into blocks (e.g. of size 8.times.8 pixels). A discrete cosine transform (DCT) comprising a set of DCT coefficients is then derived for each block. A quantization table is selected from a set of quantization tables and, using the selected table, the coefficients of each DCT are quantized. A zero-value index, corresponding to the average number of zero value quantized DCT coefficients per DCT, is determined. A predicted zero-value index is calculated using said predefined bit budget and a quantization table selected from said set of tables using the determined index and the predicted index. Using that selected table, the unquantized coefficients of the DCTs are re-quantized and the requantized coefficients compressed using run-length encoding and Huffman encoding.
Abstract: A computer implemented software device for estimating background tilt and offset is disclosed to preferably be adaptable to an optical measurement instrument wherein the estimated parameters are resolved to generate a reference plane to be subtracted from a height map of a phase profile. A set of histograms of heights is used to develop a merit function. A two-dimensional high speed iterative search is used to optimize the merit function to generate a reference plane coincident with the spatial tilt of the phase profile. The invention enables real-time measurement of substrate height for preferred use in high-speed image processing operations relating to circuit-board production lines.
Abstract: In a fingerprint detecting device, the skin surface of a finger is pressed against a transparent body. A light source irradiates incident light from an inside of the transparent body on the skin surface. A thin surface film layer is formed on the surface of the transparent body, on which a three-dimensional pattern of the skin surface is transferred by pressing the finger, to pass the incident light from the light source an area where the surface film layer contacts the surface of the transparent body. A fluid layer is formed by sealing one of a gas and a liquid between the surface of the transparent body and the surface film layer. The fluid layer is pressed by the surface film layer on which the three-dimensional pattern of the skin surface is transferred and totally reflects the incident light from the light source in an area where the fluid layer contacts the surface of the transparent body.
Abstract: A device for coding picture data by compression, and a method thereof. By preprocessing, the device calculates a picture activity of one frame of picture data. A quantizing table circuit selects a particular quantizing step matching the picture activity and sets it in a quantizing circuit 32. On a first path, one frame of picture data is fed to a discrete cosine transform (DCT) circuit to be thereby subjected to orthogonal transform. The quantizing circuit sequentially quantizes DC components, low frequency components and high frequency components of the resulting output of the DCT circuit in ascending order of frequency block by block according to a quantizing step set therein. The picture data from the quantizing circuit are coded by Huffman coding while a code counter counts the coded data. The quantizing step set in the quantizing circuit is updated on the basis of the amount of coded data and a target amount of codes.
Abstract: A new photocopy machine capable of storing and transmitting image data for scanning an image or document to permit simultaneous facsimile transmitting, printing, and recording on various storage media of the scanned image or document. The inventive device includes a housing that contains a scanning means for scanning an image, various types of storage means for digitally storing the scanned image, a printing means for printing a copy of the scanned image, and a transmitting means for permitting transmission of the scanned image to remote outside sources.
Abstract: A method and apparatus for processing an image is characterized in that an interpolation processing circuit performs interpolation on an edit area signal output from a memory with reference to an image data which has not been subjected to the interpolation thereby reducing a gap (white gap) between an image edge and an edit area without having to increase the required memory capacity.
Abstract: Detection of a watermark in blocks of video or image data which has been subjected to affine geometric distortion is accomplished by approximating the distortion by a spatially varying translation of the blocks. The watermark is extracted by spatial translations of the blocks based on an assumed affine geometric distortion. The extracted watermarks are correlated with possible watermarks and the maximum correlator output is tested for statistical significance in order to determine whether a watermark is present in the image or video frame. The data may be compressed or uncompressed data. If the affine geometric distortion is known, a search for the maximum correlator output over a space of possible distortions can be performed.
Abstract: The invention relates to a method and an apparatus for the analysis and correction of colour casts in a digitally represented image. According to a preferred embodiment, the crominance contents of the image may be transformed by use of a Hough-transformation or the like transformation, wherein the line structures in the carthesic system of co-ordinates for the crominance content is transformed to a representation in a Hough-diagram in the form of a polar system of co-ordinates, since such line structure is hereby transformed into a point structure. The Hough-diagram is examined with a view to finding a cell in the diagram which has the highest value and the cell found is saved. subsequently cells are neutralized in the Hough-diagram within a predetermined angular distance, and the Hough-diagram is re-examined with a view to finding that diagram cell which, following neutralization, has the highest value.
Abstract: A plurality of alignment marks disposed on a substrate is photoelectrically detected, and detected photoelectric signals are added to phase-shifted signals produced by shifting the phase of the detection signals to produce a sum signal. The phase is shifted by quantities corresponding to the intervals between the marks to superpose the peaks of the detection signals on the peaks of the phase-shifted signals. A correlation is determined between the sum signal and a reference signal corresponding to the narrow width of detection. A position where the coefficient of correlation is a maximum value is taken as the mark position. Thus, it is possible to use a reference signal having less peaks and a narrow width in the detecting direction, so that the number of times of product-sum calculation for calculation of the correlation can be small as compared with that when a reference signal having as many peaks as the marks and a same periodicity.
Abstract: The present invention is a method for the color correction of digital images that have been compressed. In a preferred embodiment, aspects of the color correction are carried out on the compressed image data to improve computational efficiency. One of a number of alternative methods is employed to accomplish the color correction on lossy or losslessly compressed images. The color transformation process accomplishes color correction on compressed image data in conjunction with compression or decompression operations. A second, simplified phase of the color correction may be applied subsequently to the decompressed image data in certain embodiments.
May 6, 1999
Date of Patent:
August 1, 2000
R. Victor Klassen, Steven J. Harrington, Thyagarajan Balasubramanian, Ricardo L. de Queiroz
Abstract: A method and apparatus is provided for efficiently and automatically removing unwanted temporal lighting variations from image sequences. Specifically, temporal light variations from a sequence of images are removed by a process of selecting a reference image (R) and designating a portion of the reference image as a reference image region (M), designating a first portion of at least one base image (F) of the sequence of images as a first selected image region (M') and a second portion as a second selected image region (M"), generating a color mapping by comparing the selected image region of the base image (M') with the reference image region of the reference image frame (M), and applying the color mapping to the second selected image region (M") of the base image (F) to generated a corrected image (F'). Three separate methods of performing the color mapping are disclosed.
Abstract: A method for encoding mode signals of a target block of M.times.N pixels having a binary value `0` or `255` is provided. If the target block is determined as neither "all.sub.-- 0" nor "all.sub.-- 255", either a frame-based coding or a field-based coding is selected to generate a coding mode signal, wherein the frame-based coding represents the target block is encoded on an M.times.N pixels basis and the field-based coding represents the target block is encoded on an M/2.times.N pixels basis, and a base mode of the target block based on the "all.sub.-- 0", "all.sub.-- 255" and/or the coding mode signal. If the frame-based coding is selected, a frame mode is generated.
Abstract: A method to identify the location and orientation of an article is provided, the method comprising the steps of: obtaining an image of the article with the article having a known orientation and location relative to a camera; creating a X and Y template edge matrix from the image of the article; creating a plurality of sets of modified template edge matrices, each of the sets of modified template edge matrices being a X and Y template edge matrix with the article in a different orientation; capturing an digital visual image containing the article, the digital image being a matrix of pixels; creating X and Y article edge matrices from the matrix of pixels; quantifying difference between each of the sets of modified template edge matrices and the X and Y article edge matrices with the modified template edge matrices placed at a plurality of locations within the bounds of the article edge matrices; and identifying the location and orientation of the article as the orientation of the article represented by the se
Abstract: A method and apparatus is provided for measuring patterns in an at least partially heat conducting surface. The method includes the steps of bringing a plurality of surface sensor elements into thermal contact with a substantial part of a surface to be examined, heating the plurality of surface sensor elements with a supplied heat, measuring a temperature or a change in temperature of each surface sensor element one or more times, or continuously, comparing the measured temperature or the change in temperature in each surface sensor element to the supplied heat to provide a measure of a loss of heat from each surface sensor element to the surface to be examined, and collocating the loss of heat at each surface sensor element to provide a segmented picture of the surface to be examined based upon a variation in the loss of heat from the plurality of surface sensor elements.
Abstract: A reliable method and apparatus for illuminating and imaging eyes uses multiple light sources producing multiple images of a subject each created under illumination by different illuminators. A composite image of the subject is formed by selecting pixels based upon their gray scale values or using pyramid image processing. A composite image can be created which is free of bright spots commonly caused by reflection of illumination from eyeglasses that may be worn by a subject or for which is free of dark shadows or which is free of both bright spots and dark shadows.
January 27, 1998
Date of Patent:
July 11, 2000
Sensar, Inc., Sarnoff Corporation
Theodore A. Camus, Marcus Salganicoff, Thomas A. Chmielewski, Jr., Keith James Hanna
Abstract: A printer managing apparatus includes a physical printer managing unit for managing the outputting of the attribute information of individual physical printers and controlling the outputting of a job to the physical printers, a logical printer attribute information generating unit for logically selecting the attribute information of the physical printer from the physical printer managing unit for managing attribute information of physical printers to generate attribute information of a logical printer, and a logical-printer attribute-information managing unit for managing the generated attribute information of the logical printers. With such an arrangement, the attribute information set in the logical printer is obtained from the attribute information of the individual physical printers that form the logical printer. Therefore, the attribute information of the logical printer may more properly be set than by a manager.