Abstract: An implementation of Hough transform is described. According to one aspect, the Hough transform is used for line detection in an image based on a one-dimensional voting array.

Abstract: A system for spacing, storing and recognizing electronic representations of handwriting and printing comprises a central processing unit that couples with a display device, an input device, and a memory. The memory includes means for spacing ink stroke representations, means for recognizing outline elements, means for performing deferred recognition, means for highlighting drawing areas and means for storing and displaying ink stroke representations. These means operate together to provide a pen-based or graphics oriented computer system with improved performance and added functionality for manipulating the space between ink stroke representations.

Abstract: A method for rendering patterned lines in an environment that need not support rendering of patterned lines through a standard viewing mechanism, but that does support the rendering of graphical primitives in an off-screen imaging area. A solid line is rendered in the off-screen imaging area that has been filled with a background color by turning the pixels corresponding to the solid line from the background color to a line color. A pattern is next applied to the solid line to render a patterned line by use of a filtering mechanism that filters out pixels that correspond to off-segments of the pattern. The patterned line is next transferred from the off-screen imaging area to a viewing device where it is displayed.

Abstract: Methods and systems for tracking camera pose using dynamically calibrated line features for augmented reality applications are disclosed. The dynamic calibration of the line features affords an expanded tracking range within the real environment and into adjacent, un-calibrated areas. Line features within a real environment are modeled with a minimal representation, such that they can be efficiently dynamically calibrated as a camera pose changes within the environment. A known camera pose is used to initialize line feature calibration within the real environment. Parameters of dynamically calibrated line features are also used to calculate camera pose. The tracking of camera pose through a real environment allows insertion of virtual objects into the real environment without dependencies on pre-calibrated landmarks.

Abstract: The present invention is in the field of imaging and more specifically image analysis. An object of the invention is to provide for a method which restores a digitized image or a series of digitized images automatically, by eliminating straight or virtually straight scratches in the images. The method of the invention eliminates scratches in still or animated digital images by using the residual image information in the altered areas of the images. This method can be used in the field of the restoration of films and television documents and for retouching and restoring photographic images.

Abstract: Document type comparison and classification using layout classification is accomplished by first segmenting a document page into blocks of text and white space. A grid of rows and columns, forming bins, is created on the page to intersect the blocks. Layout information is identified using a unique fixed length interval vector, to represent each row on the segmented document. By computing the Manhattan distance between interval vectors of all rows of two document pages and performing a warping function to determine the row to row correspondence, two documents may be compared by their layout. Furthermore, interval vectors may be grouped into N clusters with a cluster center, defined as the median of the interval vectors of the cluster, replacing each interval vector in its cluster. Using Hidden Markov Models, documents can be compared to document type models comprising rows represented by cluster centers and identified as belonging to one or more document types.

Abstract: A method for finding a predefined object in an image includes selecting a line belonging to the object, the line having a known stroke width and defining a maximum width and a minimum width that together define a range of widths therebetween that contains the stroke width. A feature in the image is found having a feature width within the range, and the feature is processed to determine whether it is a part of the object.

Abstract: A technique which can generate stable virtual page marks even in a non-OCR form to stably identify the form. Key segments which can be stably detected within a form are predefined for the detection of virtual page marks, and the virtual page marks (circumscribed rectangle) are generated on the basis of the key segments. Redundancy is bestowed on the detection of the key segments, and substitute segments are defined so as to generate the virtual page marks on the basis of them even in a case where the key segment has failed to be detected due to a stain, a partial lack or the like.

Abstract: The invention improves the appearance of freehand drawn lines and shapes in an electronic document by first recognizing freehand drawn lines and shapes and generating a line made up of sequential straight line segments for the freehand drawn line when the line does not form a closed line and generating a multiple straight-line-segment shape when the line forms a closed line. If a multiple segment shape is being reshaped, a basic shape is selected from reference ideal shapes as the basic shape of the multiple segment shape. The basic shape is adjusted to provide a specific shape as an improved shape for the freehand drawn shape. The recognition of the freehand drawn lines and shapes is accomplished by comparing source segments of a source freehand drawn line to a straight line and substituting a straight line segment for a source segment if the deviation between the source segment and the straight line is below a predetermined value.

Abstract: The system of the present invention includes a form design component, a form description repository, and a forms processing component Each form used with the system has a layout including a form identifier field with a common location space for each given form of the plurality of different types of forms.

Abstract: A method for electronic analysis of handwriting samples. A digital bit-map of handwriting samples created, using a digital camera or electronic scanner. The bit-map is then marked at selected points using a cursor to obtain measurements, such as the slant angles of strokes and height measurements of major areas in the writing. The measurements are tabulated and may be used to authenticate the handwriting sample. Also, the tabulated results may be compared with predetermined standard profiles for determining certain characteristics relating to the person who produced the handwriting sample. The invention also provides a method for determining the pressure and angle of the writing instrument, by creating a three-dimensional representation of portions of the writing sample for measurement and analysis.

Abstract: A method for processing partial lines of image data from a detector, each partial line of data representing a portion of an image pixel matrix, includes: (a) communicating partial lines of image data over a network from an imaging system to a remote facility; (b) receiving partial lines of image data in a first sequence; (c) assigning to each partial line of image data in a first series a position in a second sequence by reference to a plurality of base addresses; (d) altering the base addresses; and (e) assigning to each partial line of image data in a second series a position in the second sequence by reference to the altered base addresses.

Abstract: A method of binarization used in an OCR system involves in determining text pixels by checking, for each pixel, that the difference between its value and the values of a plurality of pixels located at a predetermined distance therefrom is greater than a relative threshold corresponding to the difference in intensities between the text and the background of the image, subsampling the image at a rate corresponding to at least two pixels in order to detect kernels of text, and then binarizing the image pixels only in tiles of several stroke width sides containing text kernels by using in each tile, an absolute threshold estimated in that tile.

Abstract: A pattern extraction apparatus computes the convexity/concavity of an input pattern, regards a pattern having large convexity/concavity as a character, and regards a pattern having small convexity/concavity as a ruled line.

Abstract: An apparatus and a concomitant method for detecting and recognizing text information in a captured imagery. The present method transforms the image of the text to a normalized coordinate system before performing OCR, thereby yielding more robust recognition performance. The present invention also combines OCR results from multiple frames, in a manner that takes the best recognition results from each frame and forms a single result that can be more accurate than the results from any of the individual frames.

Abstract: A method of finding a Chinese character in an electronic dictionary. The method includes sorting the characters in the dictionary into three groups according to stroke type: horizontal, vertical and slant, identifying which group a character belongs to based on the first writing stroke of the character, locating an original root of the Chinese character from the identified group based on a first three writing strokes of the Chinese character and finding the Chinese character in the dictionary based on the first three writing strokes of the Chinese character that immediately follow the strokes of the located original root.

Abstract: A data input system is provided with an input pen. A sequence of characters are traced with the input pen, and image data on them are read from the image scanner incorporated in the input pen. Each character is recognized on the basis of the image data. If a recognition error occurs during the input processing, a correct character is written with the input pen. A locus of movement is traced on the basis of moving distance data, which is obtained by a moving distance sensor and includes a moving distance and a moving direction. A character represented by the traced locus is recognized.

Abstract: Systems and methods for reordering unconstrained handwriting data using both spatial and temporal interrelationships prior to recognition, and for spatially organizing and formatting machine recognized transcription results. The present invention allows a machine recognizer to generate and present a full and accurate transcription of unconstrained handwriting in its correct spatial context such that the transcription output can appear to “mirror” the corresponding handwriting.

Abstract: Ruled-line information is stored in a ROM. The ruled-line information relates to ruled lines present in images of originals having a plurality of different formats. Based upon consecutive pixels present in an entered image of an original, the format of this original image is identified as one format of the formats that have been stored in the ROM. Ruled lines present in the entered original image are extracted based upon the ruled-line information of the format identified.

Abstract: A management information extraction apparatus learns the structure of ruled lines of a document and the position of user-specified management information such as a title, etc. during a form learning process, and stores them in a layout dictionary. During the operation, the structure of the ruled lines extracted from an image of an input document is matched with that of the document in the layout dictionary. Then, position information in the layout dictionary is referred to, and the management information is extracted from the input document.

Abstract: Certain characters and symbols are often erroneously recognized as non-solid periodic lines such as dotted lines and single chain lines. The erroneous recognition stems out from recognition criteria which are generally based upon the characteristics of the line elements such as height, width and a distance between the elements. To substantially eliminate the erroneous recognition, one preferred embodiment according to the current invention substantially eliminates erroneous candidates for non-solid periodic lines based upon a distance to adjacent text characters.

Abstract: A stroke data editing device, for editing stroke data, indicating at least one stroke of a coordinate input device, comprising, a stroke data storage unit that stores stroke data, each piece of the stroke data corresponding to one stroke of a coordinate input device, a stroke data retrieving unit that retrieves, according to a predetermined condition, at least one piece of the stroke data from the stroke data storage unit so that the retrieved stroke data corresponds to at least one stroke included in a predetermined area, and a stroke data editing unit that edits at least one piece of the stroke data retrieved by the stroke retrieving unit on a stroke basis.

Abstract: A ruled line extracting apparatus obtains circumscribed rectangles of pixel concatenation regions included in an input pattern, and calculates the most frequent value of their heights. Additionally, the apparatus integrates segments by ignoring a wild card segment, and calculates the most frequent value of height/width of extracted straight lines and segments structuring the straight line. Next, it performs a process for integrating/deleting straight lines using each threshold value based on the highest frequency value. Then, it checks/deletes a straight line according to a distribution of black pixels around the straight line, and recognizes the remaining straight lines as ruled line candidates.

Abstract: A handwriting stroke information encoder capable of reducing costs resulting from communication and storage of encoded data by reducing the amount of the encoded data. The handwriting stroke information encoder is provided with a processor which calculates forecast coordinates from sampled coordinates to be encoded and coordinates sampled before the sampling of the sampled coordinates, and an encoder which encodes the differences between the forecast coordinates and sampled coordinates of an actual input. The coordinates are forecast as; e.g., coordinates that are in line with the extension of vector, which are connecting together the previously-sampled coordinates and coordinates sampled prior to the previously-sampled coordinates, and are spaced the magnitude of the vector away from the previously-sampled coordinates.

Abstract: A method of representing handwriting which includes receiving a pen input, for example at a digitizer (10), segmenting the input into strokes (13), grouping the strokes into stroke sub-structures (14) and quantizing the stroke sub-structures according to a predefined set (the “alphabet”) of stroke sub-structures. For handwriting recognition, distance measurements are computed between the input stroke sub-structures and members of a predefined set or “alphabet” (19) of stroke sub-structures.

Abstract: The present invention intends to generate non-intersecting entire outline data from intersection-involving stroke data at a high speed, thereby outputting characters and graphics of higher quality at a higher speed. Every pair of the strokes are extracted from all the strokes of a registered character, then the curve data of the extracted strokes are straightened, and there is discriminated whether the line segments of the every extracted pair of the strokes mutually intersect. If an intersection is present, the coordinate of such intersection is calculated and stored. Then the outline is re-connected according to a predetermined rule, based on the stored coordinate of intersection.

Abstract: An apparatus including a method of digital image processing which allows a user to remove narrow, elongated areas of image distortion from a digitized image. According to the method, the user selects a portion of an image to be edited and, once the portion has been identified, the image distortion within the portion is automatically removed. The procedure for removing the distortion includes the steps of delineating the distortion within the portion of the image selected and replacing the delineated area with a correction image which is created as a function of the attributes of the neighboring portions of the image.

Abstract: A method of compressing a bitmap of a symbol includes dividing up the symbol into one or more strokes which include a number of parallel, laterally adjacent, continuous line segments, run-length encoding each stroke to form a stream of line codes for that stroke, where the stream of line codes provides absolute values for position and length of one line segment and relative values of position and length for the other line segments; and then presenting the streams of the line codes in sequence, as a set representing the symbol.

Abstract: A data processing system 10 is illustrated including an HPR input device 12 coupled to an HPR accelerator system 14. A general purpose processor 16 receives conventional character data from the HPR accelerator system 14 which translates raw input data received from the HPR input device 12. The general purpose processor 16 may be further coupled to a memory 18 and display 20 and other I/O systems 22. The incorporation of an HPR accelerator system 14 in data processing system 10 allows for the handprint recognition function to be accomplished in the background of the overall operation of the data processing system 10.

Abstract: The disclosure relates to techniques for determining the presence, orientation and location of features in an image of a two dimensional optical code. The techniques are adapted for use in mapping data in an image pixel plane with grid locations in a grid-based two dimensional code to account for size, rotation, tilt, warping and distortion of the code symbol. Where such a code is a MaxiCode, techniques are disclosed for determining the presence and location of the MaxiCode bulls-eye, orientation modules, primary data modules and secondary data modules.

Abstract: Font attributes of characters or letters are estimated by an optical character recognition system using character, run or stroke-based techniques with or without character identification information.

Abstract: A method of 3-dimensional structure estimation of the invention, making use of a plurality of stereo-pictures, repeats, for each pixel of a first picture, a step of extracting corresponding small regions (4) corresponding to the concerning pixel according to a depth estimation (68); a step of calculating a neighboring correspondence value for each of the corresponding small regions representing correspondence among neighboring corresponding small regions of picures taken by neighboring cameras; a step of obtaining a sum (61) of the neighboring correspondence values; and selecting a value of the depth estimation (68) which gives a singular value of the sum (61) representing correspondence among the corresponding small regions.

Abstract: A method for encoding handwritten symbols operates upon penstroke data received from a device capable of sampling a stylus position at discrete intervals. Each handwritten symbol is segmented into an ordered sequence of discrete strokes. An arc length and initial and final tangent angles are evaluated for each of these strokes. Each stroke is encoded in the form of a parameter set comprising position coordinates of the initial and final endpoints of the stroke, the arc length, and the initial and final tangent angles. In specific embodiments of the invention, the segmentation is based, in part, on properties of the handwritten symbol when it is expressed as a curve .phi.(s), wherein s represents arc length and .phi. represents the net turning angle.

Abstract: A method for automatically verifying the accuracy of an electronic map by an algorism which is adapted to verify irregular unit objects on the electronic map, thereby reducing the verification time and costs while obtaining improved verification accuracy. The method includes the steps of calculating a desired number of unit straight lines to be included in a broken line, overlapping an image, which is indicative of raster data transformed from vector data obtained based on original raster data of an original map, with an image indicative of the original raster data, thereby calculating the number of overlapping unit straight lines, and verifying the accuracy of the electronic map, based on the calculated results along with desired limitations associated with those results.

Abstract: An optical character recognition method and system adjusts positional, rotational and scaling differences between an input form image and a standard form image based upon selected corresponding portions of the two images prior to optically recognizing characters in predetermined areas of the input form. The corresponding portions include a predetermined mark such as a cross-line portions and or predetermined characters, and a selection process of these corresponding portions eliminates unlikely pairs of the corresponding portions based upon a distribution histogram analysis.

Abstract: In an image extraction system, an extracting part for extracting wide lines, an extracting part for extracting narrow lines and a frame detector detect a frame from a pattern which is extracted by a connected pattern extracting part. An attribute adder adds attributes of a character (graphic and symbol inclusive), frame, and a contact pattern of the character and frame to a partial pattern, and a separating part separates the frame from the contact pattern. An intersection calculator calculates intersections of the character and frame, and the calculated intersections are associated by an intersection associating part. An interpolator obtains a character region within the frame and interpolates this region based on the associated intersections. A connection confirming part confirms a connection of the pattern with respect to the extracted character pattern, and patterns confirmed of their connection are integrated in a connected pattern integrating part to thereby extract the character.

Abstract: Document distribution system 10 receives user record documents 10R containing record data 10D for distribution and storage throughout a user network 10N. Controller 10C controls the operation of the document distribution system. A transmittal sheet 10T carrying transmittal symbol accompanies each record document for directing the distribution. Document scanner 10S has a document port 10P for receiving the user documents and transmittal sheets. The document scanner is responsive to the controller for scanning the transmittal symbols on the transmittal sheet and the record data on the documents. The scanner provides an electronic pixel image of the record data for distribution and storage throughout the user network. The scanner also provides a pixel image of the transmittal symbols for directing the distribution and storage. Controller display 12 is responsive to the controller for displaying operational information about the document distribution system to the user.

Abstract: Line-by-line dataform readers and methods are enabled to operate with reduced buffer memory capacity. Storage and retention of image data are controlled on an interactive basis by concurrent analysis while lines of image data are coupled from a linear scanner. In a first embodiment lines of image data are analyzed for recognition of a feature unique to a type of dataform. Upon successful feature recognition a series of earlier and later lines of image data adequate to include a complete dataform image are retained in memory for decoding purposes. Prior to successful feature recognition, as lines of image data exceeding a limited memory storage capacity are received the earliest stored lines are discarded on a first-in-first-out basis. In a second embodiment, a buffer memory has a storage capacity adequate to store only a partial dataform image.

Abstract: A processor based method for recognizing, capturing and storing tabular data receives digital-computer data representing a document either as a pixel-format document-image, or as formatted text. Within the digital computer, either form of the digital-computer data is processed to locate tabular data present therein. After a table has been located, tabular data is extracted from cells present in either form of the digital-computer data. The extracted tabular data is stored into a database present on the digital computer.

Abstract: It is an object of this invention to provide an image processing device which would be able to extract a specified pattern accurately even if there are fine lines in its periphery. The invention is configured as follows in order to achieve this object. The density data for R, G and B are transmitted to fine line processor. There a count is made of the number of pixels among the eight surrounding a center object pixel, which have density values within a fixed margin of the center object pixel. If there are three or fewer, the center object pixel is judged to be part of a line segment, and the determination flag 0 is output. If there are four or more, the center pixel is judged to be not part of a line segment, but a part of pattern. Once the object pixel is judged to be a part of a line segment, then a line removal device remove the object pixel from the image data in order to extract a specified pattern.

Abstract: An information inputting method, as well as an information inputting device, extracts effective characteristic points from inputted stroke information and outputs coordinate information relating to the extracted characteristic points. First point coordinates values are input in a sampling cycle at a predetermined sampling rate. Second point coordinates values are input in the sampling cycle delayed a predetermined time behind the inputting of the first point coordinate values. A first vector is formed based on the first and second point coordinate values. First and second point coordinate values are also input in the subsequent sampling cycle, and a second vector is determined based thereon. The first point coordinate values inputted earlier are selected when these values cause the angle formed between the first and second vectors to exceed a predetermined threshold value. Information concerning the selected first point coordinates values are output to, for example, a host device.

Abstract: A hierarchical database system for on-line Chinese character recognition is disclosed. The hierarchical database system comprises: (1) the database of character description; (b) the database of stroke correspondence rules of components; (3) the database of character structures; (4) the database of standard component patterns; and (5) the database of spatial relationships between strokes of components. The database of character description stores the rule code(s) of the constituent component(s) and the character structure for all the characters included in the vocabulary. The database of stroke correspondence rules of components stores stroke correspondence rules for all the components denoted by the above mentioned rule codes. The database of character structure stores the synthesis rules of character patterns, the decomposition rules of character structures, and the spatial relationships between components, for all the 208 character structures utilized in a preferred embodiment.

Abstract: There is provided a method for detecting horizontal and vertical lines of an encoded run length image to identify cells of the image that are formed by the lines. Initially, the method comprises the steps of filtering excess details, such as text, from the image and grouping a portion of the image for analysis. A top profile of the image is generated and, then, points along the top profile are determined where substantial changes in the profile position occur. A horizontal section of the top profile which corresponds to a line completely visible from the top is then extracted from between the points. By repeating the above steps, each horizontal line of the image is detected and extracted from the image. In addition, in order to detect the vertical lines of the image, the image is simply rotated 90.degree. and processed with the vertical lines being positioned horizontally.

Abstract: A method and apparatus to recognize hand-printed characters represented as an electronic ink stream. A first filtering box (300) is defined around a center of a first output point (324) in the electronic ink stream and the first output point is added to a filtered ink stream. A second output point (328) is added to the filtered ink stream only when the electronic ink stream moves outside the filtering box (300). When the second output point (328) is added, the filtering box is re-centered on the second output point (328) only in a dimension that is exceeded after adding the second output point (328) to the filtered ink stream to create a second filtering box (310). Moreover, extraction points (630, 640, 650 and 660) are provided approximately equally spaced across a finite set of points of electronic ink to create a compact representation of a finite set of points of electronic ink.

Abstract: An input carrier sheet 12C for document distribution system 10 carries input symbols hand entered by the user into pre-existing constraint grids 12. The constraint grids may be printed in continuous tone or halftone. The print only partially covers the underlying carrier, permitting the exposed carrier to reflect light. The grids have sufficient pigment to be visible to the user, but insufficient pigment to form foreground pixels along with the hand-entered stroke when detected during the scanning. The signal (symbol)-to-noise (carrier) ratio is enhanced by reducing the pigment content of the constraint grids which increases the reflectivity of the grids. The S/N may be further enhanced by placing the strokes of the hand-entered symbols on top of the grid which occults some of the grid pigment. The S/N is further enhanced by highly reflective brightening agents in the grid print, and by aperture effect during scanning.

Abstract: An efficient system and method for reliably identifying guidelines, ruled lines, and the like, in images of text, and distinguishing those portions of such lines which touch or intersect character strokes in the image. The system provides for removal of guideline segments between character strokes without deleting character strokes. The system operates effectively on both machine printed and hand formed text. An image of text has most characters separated for more effective subsequent OCR processing, despite the presence of guidelines connecting the characters.

Abstract: An image processing device for extracting connected components of image data includes a first image memory for storing image data that contains connected components (e.g., object image data). A processor searches the image data stored in the first image memory to detect a part of the object image data and determines a watching area of the image data in the main rasterizing scan direction on the basis of the detected part of the object image data. The processor then searches the image data to detect a line segment of continuous pixels that is coupled to and includes the watching area. Once, the line segment is detected, the processor cancels the detected line segment from the image data stored in the first image memory and outputs data corresponding to the cancelled line segment. The device also includes a second image memory for storing the detected and cancelled line segment outputted by the processor as a connected component of the image data.

Abstract: A line image detection apparatus having an image input function for inputting an image on a document. The image includes a rectangular character string and a linear line drawn along the character string. The line detection apparatus has a function for detecting a character string image from the image, a function for forming a projection image on the basis of the distance from a lower end of the character string image to each first black pixel found in the character string image in the vertical direction, and a function for comparing the distance from the lower end of the character string image to each first black pixel, with a predetermined threshold value, thereby to define a plurality of "slice 0" areas included in the projection image. The distance, in each of the "slice 0" areas, from the lower end of the character string image to each first black pixel is lower than the predetermined threshold value.

Abstract: Image processing method and apparatus are provided. A histogram of image information is obtained. A peak position is detected from the histogram obtained. A small area is decided by using the detected peak position as a reference. A histogram of the decided small area of the image information is obtained. A specific point is determined from the peak position of the histogram of the small area obtained. The size of small area is decided in accordance with a peak width of the histogram of the image information. The image information is inputted by a scanner. A table can be used as image information and the specific point which is decided is an intersection of ruled lines of the table. A cell area of the table is judged from the intersection decided.

Abstract: The orientation of strands relative to a selected axis is determined by analyzing a digitized grey scale image to select edge pixels therefrom that may define the edges of strands segmenting the image into a plurality of subimages and processing one of the subimages by defining pixel lines merging the pixel lines to form lines and culling selected one of the lines in the subimage. The subimage is then combined to co-terminus other of the subimages to form a combined subimage which is processed in the same manner by merging pixel lines to form lines and culling selected line in the combined subimage. This process is repeated by combining further subimages and processing until substantially all of the image has been processed to provide the plurality of strand lines defining strand edges and then determining the orientation of the strand lines relative to the axis.