Abstract: A system electronically registers an image on an input document. The system includes a scanner for generating an image data stream representing an electronic image of the image on the input document and an edge detecting circuit for detecting edge data within the image data stream. A circuit detects a first corner of a leading edge of the input document based on the detected edge data and for establishing a first coordinate value therefrom and a second corner of a leading edge of the input document based on the detected edge data and for establishing a second coordinate value therefrom. The system further determines a minimum and maximum location for a leading edge of the scanned document and determines a minimum and maximum location for a trailing edge of the scanned document.

Abstract: An image processing method includes the steps of: detecting an inclination of image data; storing the image data for one line into a first memory; changing a reading timing when the image data is read from the first memory, according to the detected inclination, whereby correcting a shift in a main scanning direction; storing the image data for not smaller than 2 lines into a second memory; and changing a line position in a sub-scanning direction when the image data is read from the second memory, according to the detected inclination, whereby correcting a shift in the sub-scanning direction.

Abstract: An image capturing apparatus capable of performing a suitable illuminance nonuniformity correction by setting a suitable &ggr;-characteristic every block even in the case that the size of a character image projected on a sensing surface changes according to an image capturing magnification, and capable of extracting a boundary area between a white board portion and a background portion and applying a suitable image processing to this boundary area during the illuminance nonuniformity correction for a picked image, and capable of image capturing a representation such as characters drawn on a white board in such a manner that an obtained image is clear and easy to see by suitably performing an illuminance nonuniformity correction even in the case of color image capturing, and capable of detecting a regularly reflected light with high accuracy and can thereby securely prevent an error of obtaining an image of low quality by image capturing, and capable of preventing an error in flash-image capturing a representa

Abstract: Characters read as images at different angles and as mirrored images in accordance with scanning directions are detected by an character recognition procedure, and are converted into erecting images. Such character images have low character recognition rates, therefore, a character image is extracted from the scanned image, a rotated angle and an existence of mirroring in which the recognition rate becomes maximum are detected. Then, the same rotation and mirroring are applied to the scanned image, thereby capable of converting the scanned image into an erecting image.

Abstract: A device and method for correcting a skew of an object in an image. The device and method comprises an input part for receiving an image; a binarization part for binarizing pixels of the image into pixels having brightness values for character pixels and background pixels; a candidate stripe generation part for generating candidate stripes by performing dilation on character regions of the binarized image. The device and method further comprises a stripe classification part for classifying candidate stripes having a predetermined eccentricity and blob size among the candidate stripes, as valid stripes; a skew angle decision part for calculating direction angles of the classified stripes, and determining a direction angle having the largest count value as a skew angle; and a skew correction part for correcting a skew of an image by rotating the mage by the skew angle.

Abstract: Labeling process unit groups a continuous black pixel area as a group by determining the sequence of black pixels from the binary image data read from the image input device, and extracts bounding rectangle information about each of the grouped continuous black pixel areas. Row extracting process unit extracts row rectangle information contained in an original image from the group bounding rectangle information extracted by the labeling process unit. Punctuation mark identification unit identifies a punctuation mark contained in the row rectangle extracted by the row extracting process unit. With the configuration, the direction of a row can be automatically determined by checking the relative position of the punctuation mark in a row based on the extracted row rectangle information and the extracted bounding rectangle information.

Abstract: A method and system for determining the orientation of an image of a picture within a scanned image, including the steps of locating in the scanned image the contour of the image of the picture, determining a plurality of bounding boxes confining the contour of the image of the picture, selecting one of the plurality of bounding boxes that is substantially aligned with the contour of the image of the picture, and calculating an angle of rotation of the picture based on the selected bounding box.

Abstract: A method of inserting additional information, such as a digital signature, in a digital image comprises the following steps: detecting (E10) the points of interest in said image (I); selecting (E11) a subset of points of interest adapted to define a geometric reference frame in said image (I); calculating (E12), for the geometric reference frame, reference information necessary for determining the geometric transformation applied to the image for a class of predetermined geometric transformations; coding (E13) said reference information in a reference key (Kr); and inserting (E16-E19) the additional information in said digital image (I). Use notably for geometrically re-fixing the image before extracting the signature.

Abstract: A method and apparatus for processing a source raster representation of an image, where the source raster representation of the image has an orientation relative to the source raster. The method includes partitioning the source raster into blocks of pixels based on the orientation, and for each block of pixels, calculating a destination location in a destination raster for the block of pixels and transferring the block of pixels to the calculated destination location in the destination raster, the destination location being calculated from the orientation.

Abstract: The present invention is a method and apparatus for improving the appearance of electronic images, and more specifically, to the efficient use of template rotation within a template matching process to enhance such images. The invention reduces the number of entries that are stored in a look up table by eliminating identical patterns that differ only by the angle at which they are output. The appropriate templates are then rotated it by the required angle for output subsequent to their retrieval from the lookup table.

Abstract: The orientation of an image such as a radiographic image is deduced from the digital representation of the image.

Abstract: A method of detecting the orientation of a radiographic image represented by a digital signal representation wherein mathematical moments of the digital signal representation are calculated relative to different reference entities and wherein a decision on the orientation of the radiographic image, for example the position of the thorax edge in a mammographic image, is obtained on the basis of an extreme value (minimum, maximum) of the calculated moments.

Abstract: There is described image-processing methods, apparatus and programs, which make it possible to accurately determine an oriented direction of an image to be displayed as a single screen image, without requiring operator's assistances. The method includes the steps of: determining whether or not a personal image is included in the image, based on image signals representing the image; recognizing an oriented direction of the image, based on at least one of element-feature quantities with respect to constituent elements constituting the personal image, when determining that the personal image is included in the image in the determining step. The constituent elements include a shape of face, a positional relationship between both eyes, a positional relationship between the face and hair on a head, and a positional relationship between the face and a hand or a leg.

Abstract: A system and method for utilizing a search tool that registers transformation of a trained pattern by at least four degrees of freedom to register the instance of a pattern in an arbitrary six-degree-of-freedom pose is provided. The search tool is first trained to recognize a plurality of versions of a trained pattern/fiducial that are incrementally transposed to induce differing levels of known aspect and shear. An object having several instances of the trained pattern located at known spacings and orientations therebetween is imaged by the trained search tool, and the located instances of the trained pattern are compared to expected instances of the trained pattern to measure relative six-degree-of-freedom orientation for the underlying object.

Abstract: A method for evaluating an orientation of a molecular array having features arranged in a pattern. An image of the molecular array is obtained by scanning the molecular array to determine data signals emanating from discrete positions on a surface of the molecular array. An actual result of a function on pixels of the image which pixels lie in a second pattern, is calculated. This actual result is compared with an expected result which would be obtained if the second pattern had a predetermined orientation on the array. Array orientation can then be evaluated based on the result.

Abstract: An apparatus and method for rotating OSD fonts are disclosed. The OSD rotation device including a central processing unit, font addresses generator, memory, a font ROM, a decoder, a shift register, a output controller and output circuits, a synchronization signal generator, and a display device. In a manner, the OSD fonts corresponding to the OSD font addresses are produced. Next, the OSD fonts is read and delivered to the decoder. Further, the decoder performs a decoding process for the OSD fonts to form a plurality of rotated fonts. The OSD message, including the rotated fonts, is displayed on a display device so that the video display is synchronized with the OSD message.

Abstract: In a machine-fed scanner, orientation angles of edges of an image bearing substrate are obtained and used to calculate image shear and/or skew. A running weighted average of the image skew may be kept in a memory. When a skew value is obtained for a given image, it may be determined whether the skew value is within a predetermined range. If the skew value is within the predetermined range, the skew value is used to determine an image revision to compensate for the skew, and the skew value is incorporated into the running weighted average skew. If the skew value is not within the predetermined range, it is discarded and the running weighted average skew is used to determine an appropriate image skew revision. The running weighted average of the shear may also be kept in a memory. A shear value is obtained for each image, and incorporated into the running weighted average shear. Shear revision is performed based on the running weighted average shear.

Abstract: The invention relates to a process for the formation of an image obtained from an array of detector pixels comprising at least one detector pixel, the image formed constituted by at least one set of P image pixels for each detector pixel, characterized in that it successively comprises: a stage (E2) making it possible to create N elementary pixels from one detector pixel, N being an integer equal to or lower than P and a stage (E3) of the random distribution of events received by a detector pixel in the N elementary pixels corresponding thereto. The invention applies to the field of medical imaging.

Abstract: A system, method, and corresponding computer data product, generates a high resolution reference image, generates a set of rotated reference images from the high resolution reference image, cross-correlates the set of rotated reference images with the high resolution reference image to generate a set of cross-correlation values, and determines that the high resolution reference image is suitable for use in a normalized gray scale pattern find process when the difference between any two correlation values generated by the image cross correlation module is greater than a predetermined threshold value.

Abstract: A fingerprint verification system for verifying input fingerprint data with pre-registered reference fingerprint data is disclosed, the reference fingerprint data being data containing the fingerprint data, the pre-designated rotated angles, and the rotated order thereof, the system comprising a rotation data and reference fingerprint data storing portion for storing the reference fingerprint data.

Abstract: A method and system for producing digital orthophotos from imagery acquired as full or sparse stereo. The orthophotos can be produced in a variety of map coordinate systems without the need to convert or recompute DEM or photogrammetric solution data. In one embodiment, a two dimensional, planimetric free-network solution, utilizing arbitrary datum definition constraints, is used to provide a transitory coordinate system that is used to facilitate the image measurement process. It is utilized as a preliminary step to refine apriori block layout information to facilitate point picking and to provide general quality control capabilities before undertaking a rigorous 3D photogrammetric adjustment. In place of a general map conversion transformation, an identity transformation can be used, so that map coordinates and world coordinates are identical.

Abstract: Skew angle in a document image is estimated using operators known from mathematical morphology. Skew angle in a document image (A) is estimated by run-length smoothing the image and then producing a plurality of eroded run-length-smoothed images. The run-length-smoothed image (RLSA(A)) is eroded using a linear structuring element (k2L&agr;) oriented at each of a plurality of different angles (&agr;). The angle of the linear structuring element which produces an eroded image having the greatest surface area is designated as the skew angle. A plurality of run-length-smoothed images (RLSA&agr;(A)) may be produced, each generated by smoothing the document image using a linear structuring element (k1L&agr;) oriented at a respective different angle (&agr;i). Then each run-length smoothed image (RLSA&agr;(A)) is eroded using a linear structuring element oriented at the corresponding angle (&agr;i).

Abstract: An image reading apparatus, comprises an image reader to read images of a document; a rotation processor to conduct a rotation processing for image data obtained through reading by the image reader in accordance with inclination of the document; a memory in which image data subjected to rotation processing by the rotation processor are stored; and an external inputting device to input arbitrary rotation angles. When the rotation angles are inputted from the external inputting device, the image data obtained through reading by the image reader are subjected to the rotation processing in accordance with the inputted rotation angle and are stored in the memory.

Abstract: A method to be used within a document image capture device, such as a document scanner (10) of detecting and measuring the document skew (25) while the document is still entering the imaging area and then effecting a rotation of the camera to accommodate the document skew so that the resulting image taken is not skewed. This can be accomplished by one of several methods, such as rotating the entire camera, rotating only the sensor element, or rotating another element in the optical path such as a prism. Use of the methods incorporated in this invention does not require the document motion to be stopped to measure and accommodate skew or to take the image.

Abstract: A method of recognizing a character specifies a character in an image. Furthermore, the method of recognizing a character evaluates a symmetrical-ness of a shape of the character. Furthermore, the method of recognizing a character carries out an inclination correction process for the shape when the symmetricalness of the shape is detected.

Abstract: A method of registering images is provided which includes capturing a first original image of a scene under surveillance, capturing a second original image of the scene under surveillance, coarsely matching the first original image to the second original image to give approximate angle and offset parameters of the first original image to the second original image, and finely matching the first original image to the second original image based on the approximate angle and offset parameters of the coarsely matching step. The coarsely matching step involves applying a plurality of rotations to a reduced version of the first original image to form a rotated and reduced first original image and performing a correlation match between the rotated and reduced first original image and a reduced version of the second original image.

Abstract: A method for determining component misalignment in a multi-modality imaging system including a first modality unit, a second modality unit, and a table, includes imaging an object with the first modality unit to generate a first image, imaging the object with the second modality unit to generate a second image, and determining a table alignment status using the first and second images.

Abstract: Method and apparatus for observing a specimen image on a scanning charged-particle beam instrument in such a way that the original observational position can be automatically resumed after movement of a specimen or its image. When an image is observed at the original position after a specimen or its image is rotated or moved, a keyboard or a pointing device is operated to command reconstruction of the image. In response to this, the CPU of the instrument reads data from a memory that indicates the history of rotations and movements of the specimen and image. For example, if the specimen has been mechanically rotated, the CPU controls a rotational drive circuit according to the data read from the memory, the data indicating amounts and directions of rotations. The specimen stage is rotated through a given angle in a direction opposite to the previous direction. As a result, the specimen can be returned to the position assumed prior to the rotation.

Abstract: A method for printing a human-readable document from a composite image data file is disclosed. The method comprises obtaining a scanned image of the composite printed document to extract a spatial pattern of the document and other image property information. Next, the machine-readable content in the composite image is decoded to obtain reference information describing nominal spatial relationships of the human-readable and machine-readable content. The method then compares observed spatial relationships with reference spatial relationships to obtain a spatial deviation value. The spatial deviation value is then applied to the scanned image, and then a corrected image is outputted.

Abstract: A system and method are provided for aligning a scanned document in a digital scanner. The method comprises: scanning a document to create a matrix of pixel information; averaging pixel intensity information into a first plurality of row-mean buffer elements, where each row strip includes information from a second plurality of column elements; and, averaging pixel information into a third plurality of column-mean buffer elements, where each column strip includes information from a fourth plurality of row elements; comparing adjacent elements of matrix row information to generate a first set of edge positions; comparing adjacent elements of matrix column information to generate a second set of edge positions; generating a rotation angle in response to analyzing the first and second edge positions; and, rotating the scanned document by the rotation angle.

Abstract: The invention provides an image reading apparatus and method capable of performing skew error correction at high speed. In an image reading apparatus that conveys a stacked medium along a conveyance path, reads an image of the medium being conveyed, and discharges the medium after reading the image, the apparatus comprises calculation means for calculating the amount of skew of the medium while being conveyed, and correction means for correcting the amount of skew of the medium in accordance with the amount of skew calculated by the calculation means. The correction means comprise determination means for determining whether correction for the amount of skew should be performed or not, and control means for correcting the amount of skew of the medium. The calculation means comprise medium detection means for detecting the state of the medium conveyed along the conveyance path, and calculate the amount of skew from the detected result of the state of the medium.

Abstract: A subject inclination detector provided with: a display for displaying on a screen a subject image comprising a plurality of pixels; a pixel specifier for a user to specify a first pixel from among the pixels; a calculator for defining a pixel group constituting a circumference of an area including the specified first pixel, and calculating deviations in pixel value between adjoining pixels of the pixel group; and a detector for selecting a second pixel and a third pixel based on the calculated deviations, and detecting an inclination of a straight part of the subject based on the selected second and third pixels.

Abstract: A method for determining skew angle (5) and location of a document in an over-scanned digital image is disclosed. A document is scanned to capture a gray scale image. The outline pixel (1) of a document from the captured gray scale image are connected (2). An initial angle is set at zero and a x-minimum, y-minimum, x-maximum, and y-maximum of the polygon are determined (3). A rectangular boundary (4) area based on the x-minimum, y-minimum, x-maximum, and y-maximum is calculated and recorded. The angle is recorded and the polygon is rotated z-degree. The process is repeated and a minimum boundary area is selected. An angle corresponding to the minimum boundary is the skew angle of the document.

Abstract: 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.

Abstract: A system is disclosed for determining the rotational orientation of an object. An image of the object is digitally captured. The digital image is filtered through an edge detection operation to enhance the edge information contained in the image. The filtered image is rotated through a series of incremental angles to produce a series of rotated images. Each rotated image is projected onto an x-axis and y-axis defined by pixel grid axes defined by the original image. The projection of the rotated images produces projected pixel counts formed by the summation of pixel value differences of the x-axis and y-axis on the projected image. For each rotated image, a score corresponding to the sum of the difference of gray-scale values for adjacent projected pixels is computed. The scores for the projections of each rotated image are plotted on a score-angle graph. A curve which includes the highest score and the neighboring next highest scores is interpolated to determine the peak score.

Abstract: A simple and robust method for determining the orientation angle of a scanned image or the angular orientation of an object in an electronic image based on using descriptors of the image and systems that implement the method are disclosed. A system including means for acquiring the electronic image, means for determining image centroid coordinates, means for obtaining second order moments corresponding to the image, and, means for determining an orientation angle of a principal axis of the image implements the method.

Abstract: A method for deskewing text is disclosed, in which a noisy digitized gray scale image is preprocessed by binarization, noise removal, and then by smearing in three patterns. A selected smeared image having minimal runs is sampled, and the resulting sampled image subjected to the Hough process using a search technique at two angular resolutions. A dominant text line is chosen from the output of the Hough process, and the image deskewed accordingly. The method is suitable for detecting skew at any angle, and is capable of detecting multiple skew text lines.

Abstract: An improved orientation system includes a lens pre-distorting an image generated by imaging apparatus such that a non-linearity, the degree of which is dependent on the off-axis distance of an imaged object, is generated so as to increase the strength of a signal indicating said non-linearity.

Abstract: An image-processing system comprising an image-acquisition device configured to generate a representation of a source object, the representation comprising a plurality of pixels, a processor communicatively coupled to the image-acquisition device, and a memory element communicatively coupled to the processor, the memory element comprising logic configured to determine when the source object is skewed in the representation, and logic configured to identify the borders of an orthogonal portion of the representation responsive to the logic configured to determine, wherein the orthogonal portion comprises unaltered pixels that contain information corresponding to the source object.

Abstract: An angular orientation of a lattice image pattern is found by forming a composite image of superimposed portions of a lattice image by aligning centroids of a lattice element in each portion, and determining a lattice axis from a line fit through centroids of lattice elements in the composite image. The composite image is formed by selecting a seed pixel, finding a local minimum near the seed pixel and then finding a centroid based on the local minimum. From the centroid, it is determined whether the glyph centroid has sufficient contrast to be included in the composite image. The composite image is formed by superimposing subsamples of the lattice image based on glyph centroids. The composite image is then analyzed to determine a lattice axis through centroids in the composite image, and determining a quadrant image based on the lattice axis.

Abstract: Cameras (10, 12) produce first and second images (14, 16) of an object (18) from different viewpoints. An image partitioning module (54) partitions the images (14, 16) into a plurality of vertically striped regions (70), such that each region (70) of the first image (14) corresponds to a region (70) of the second image (16). A region alignment module (55) vertically shifts a region (70) of the first image (14) in a direction calculated to vertically align a portion (30) of the region (70) with a substantially matching portion (32) of the corresponding region (70). A disparity calculation module (62) calculates at least one disparity value between at least a portion (30) of the vertically shifted region (70) and at least a portion (32) of the corresponding region (70).

Abstract: A method of calculating a skew angle for a two-dimensional barcode, in which the horizontal or vertical edges within the barcode are located, preferably using a finite state recognizer, and an edge array identifying the horizontal or vertical edges is generated. Next, the edge lines within the edge array are identified and traced, and any segments of edge lines within the edge array which are perpendicularly connected are separated. The slope for each edge line within the edge array is calculated, preferably using linear regression techniques. In the preferred embodiment, a discrete histogram of the slopes is generated and the skew angle is then set as the highest value within the discrete histogram. In a preferred embodiment, edge lines within the edge array which are located within a first predetermined threshold of each other are merged with each other and edge lines within the edge array having a length less than a second predetermined threshold are eliminated from the edge array to speed processing.

Abstract: An image forming device includes a backing and is arranged for a media to be positioned against the backing. The backing includes an embedded backing pattern. The media is skewed with respect to the scanning, thus defining a media skew. The media and the backing are scanned to form a pixel pattern. For each pixel, an included pattern recognition algorithm determines when the pixel represents the backing. When the pixel represents the backing, it is replaced with a replacement backing symbol, otherwise it is replaced with a replacement media symbol. The pixel determining and symbol replacing process is continued for remaining pattern pixels to form a replacement symbol pattern comprised of the replacement backing symbols and the replacement media symbols. The media is then processed based on the replacement symbol pattern, the processing including detecting the media skew, registering the media, or both.

Abstract: A reference line detecting step 4 assumes two reference lines dividing a character row image into three, i.e., upper, intermediate and lower areas to be quadratic curves independent of one another, and obtains parameters determining the two reference lines such as to best separate the length distributions of white runs in the three areas from one another. A reference line correcting step 5 corrects the character row image from the parameters and the same image such as to obtain two horizontal reference lines and predetermined values of area (i.e., height) ratios of the three areas, and feeds the corrected image back to a preprocessing step 2. A character row reading step 3 executes character segmentation, feature extraction and character recognition with the character row image from the preprocessing means 2. The means 3 executes selection of the character recognition results, and outputs a most likely read-out result.

Abstract: According to an image data correction conversion circuit and a skew quantity correction method respectively according to the present invention, image data for predetermined lines is stored, the quantity of skew of the stored image data is measured, the number of blocks into which the image data is divided is operated based upon the quantity of the skew and a control signal for correcting and converting the image data based upon the operated data is output. A selector acquires image data the quantity of skew of which is corrected by selecting and outputting the stored image data for predetermined lines according to the control signal.

Abstract: A method of measuring of the present invention includes the steps of photographing a rotating body with the curved surface having marks to obtain two-dimensional images thereof; generating a three-dimensional imaginary body with the curved surface having marks; setting an arbitrary posture of the imaginary body as a reference posture, performing a posture displacement operation in such a way that the marks of the imaginary body are coincident with the marks of the body; specifying a three-dimensional posture of the body on the basis of an amount of the posture displacement operation relative to the reference posture; and determining the rotational amount of the body and the direction of the rotational axis thereof by computing a rotation matrix relating to a rotation operation to make the three-dimensional posture of the body at one time coincident with that at another time.

Abstract: An image processing apparatus is provided with a binarization processing unit separately from a pseudo gradation processing unit that binarizes image data of a document image according to an error diffusion method. The binarization processing unit binarizes the image data of the document image using a threshold that is set higher than a threshold used by the pseudo gradation processing unit, with each threshold representing a different predetermined density level. Histograms showing the density distribution of the document image are generated from the image data binarized by the binarization processing unit. The orientation of the document image is recognized by the histograms.

Abstract: An image forming apparatus for inputting an original image and forming an image on a recording medium, includes an image direction detection unit for detecting the image direction of an original image, and a control unit for controlling image forming device to perform predetermined image forming processing in accordance with the detection result of the image direction detection unit. The image direction detection unit outputs a degree of determination as the detection result of the image direction, and a level for determining whether or not the predetermined image processing is done is set for the degree of determination.

Abstract: A system and method for digital x-ray imaging may correct distortion in a digital image by determining the cell size of a grid. A rotational correction may be applied to the cell to correct for distortion that might result from improper alignment between an imaging sensor used to capture the image and the image itself. Stretch factors for the pixels are calculated and applied to the cells of the grid to provide cells of consistent size. The calculated stretch factors may be applied to subsequent images to provide an efficient method for optical distortion correction.

Abstract: A method for determining the orientation of a digital image, includes the steps of: employing a semantic object detection method to detect the presence and orientation of a semantic object; employing a scene layout detection method to detect the orientation of a scene layout; and employing an arbitration method to produce an estimate of the image orientation from the orientation of the detected semantic object and the detected orientation of the scene layout.