Abstract: Real-time segmentation of foreground from background layers in binocular video sequences may be provided by a segmentation process which may be based on one or more factors including likelihoods for stereo-matching, color, and optionally contrast, which may be fused to infer foreground and/or background layers accurately and efficiently. In one example, the stereo image may be segmented into foreground, background, and/or occluded regions using stereo disparities. The stereo-match likelihood may be fused with a contrast sensitive color model that is initialized or learned from training data. Segmentation may then be solved by an optimization algorithm such as dynamic programming or graph cut. In a second example, the stereo-match likelihood may be marginalized over foreground and background hypotheses, and fused with a contrast-sensitive color model that is initialized or learned from training data. Segmentation may then be solved by an optimization algorithm such as a binary graph cut.

Abstract: Dropping out of color form backgrounds from images of completed forms to obtain color form dropout images retaining only the respondent information. In one embodiment, a color form image processing method (100) includes retrieving (102) a template image, retrieving (104) a respondent image, registering (106) the images against one another to establish correspondence between pixels in the respondent and template images, dilating (108) the template image, and performing (110) a color form dropout including comparing (112) corresponding pixels in the respondent and dilated template images, and determining (114) whether to keep corresponding pixels by applying (116) a geometric solid threshold comparison to assess both color similarity and relative darkness, and removing (118) pixels from the respondent image based on such comparison.

Abstract: Methods and systems for cropping images of book pages are disclosed according to one embodiment of the invention. A method may include identifying reference images and receiving cropping rectangles for the reference images. These cropping rectangles associated with reference images may then be used to generate cropping rectangles for images of book pages between the reference images. The cropping rectangles may be generated based on a linear interpolation of the cropping rectangles associated with the reference images and the number of pages between images. The method may also display one or more images of book pages with the associated one or more cropping rectangles superimposed thereon. A user may then have the opportunity to make adjustments to the position and/or size of the cropping rectangles.

Abstract: After markings have been placed on a pre-printed form by a user who interacted with an entity, the form is scanned to produce a scan file. The scan file is analyzed to identify whether user added markings are present on machine readable selection items. The method can take a number of automated actions, depending upon which pre-printed machine readable selection items were checked by the user. For example, in response to checkbox selections, the method can obtain (read) some form of electronically storable data relating to the entity based on which of the machine readable selection items the user checked. Alternatively, in response to other checkbox selections, the method can ignore the user added markings on the machine readable selection items. In addition, in response to the checkmarks, the system can maintain only an image of the user added handwritten text.

Abstract: A handwritten information management system, which includes a paper sheet document generation section that generates a paper sheet document in which plurality pieces of coded information including coordinate information indicating a position on a paper sheet and document identification information of the paper sheet document formed on the paper sheet are recorded; a paper sheet document registration section that registers the generated paper sheet document and an electronic document corresponding to the paper sheet document correlated with the document identification information; a handwritten information detection section that detects handwritten information handwriting on the generated paper sheet document; a handwritten information registration section that registers handwritten information identification information to identify the handwritten information correlated with the detected handwritten information; and an identification information management section that manages the document identification info

Abstract: There is described a system and method for automatically discriminating between different types of data with an image reader. In brief overview of one embodiment, the automatic discrimination feature of the present image reader allows a human operator to aim a hand-held image reader at a target that can contain a dataform and actuate the image reader. An autodiscrimination module in the image reader in one embodiment analyzes image data representative of the target and determines a type of data represented in the image data.

Abstract: A form reader includes a landmarks extractor configured to select textboxes of a converted document as form landmarks based on textual characteristics. A set of positional constraints constrain the form entries relative to the identified form landmarks. A constraints solver selects textboxes of the converted document as form entries by solving the set of positional constraints respective to a set of facts including the selected form landmarks and converted document. In some embodiments, the constraints solver includes a query engine configured to (i) construct a query in a logic programming language setting forth the set of positional constraints and the set of facts and to (ii) input said query to a logic programming language query solving engine and to (iii) receive a response from the query solving engine responsive to the input.

Abstract: Disclosed is an image processing apparatus in which a marked image is extracted more accurately and classification based upon the type of mark is made possible. The apparatus includes an extraction unit adapted to extract information representing a region position of an object and information representing a region position of a mark as an object list and mark list, respectively, from the document image that has been scanned by a scanning unit; and a search unit adapted to search for information representing the region position of a mark in the mark list for which the distance is the shortest with respect to information representing a region position of each object in the object list.

Abstract: An image recording apparatus, method, and program corrects a red-eye portion of an image without requiring a coordinate input apparatus, properly corrects a red-eye portion of the image present at a position designated by a user, and detects a red-eye portion of the image, without utilizing a fluorescent ink. The apparatus detects a correction candidate position in a digital image designated by the user by a predetermined method, extracts a contour from the designated digital image, prepares a monochromatic schematic image based on the extracted contour, prepares a marked schematic image by synthesizing an indicator mark, indicates a correction candidate position, in the correction candidate position detected by correction candidate position detecting device, and prepares a mark sheet by printing the marked schematic image and mark sheet data.

Abstract: It is an object of the present invention to improve a rate of recognition and recognition speed, reduce a degree of a specification required for a device forming a photographing device or a system and lower consumed electric power and a cost. In a character recognition processing method according to the present invention, a character image as an object whose character is to be recognized is displayed on a display (12) together with a cursor of a character frame for recognizing a character. While the cursor is set to a character string as an object to be recognized, the character image is photographed by a photographing part (9). In an image fetching part (3), the image data of the photographed character image is fetched. In a layout analyzing part (5), the fetched image data of the character image and cursor position information from a cursor control part (4) corresponding to the character image are received.

Abstract: A method for deriving a representation of an image is described. The method involves processing signals corresponding to the image. At least one region of the image that is representative of the image is identified. In one embodiment, each identified region corresponds to a feature of the image. For each identified image region, a two dimensional function of at least part of the image region is derived and processed to obtain an intermediate representation of the at least part of the image region. A representation of the image is derived from the intermediate representation of at least one of the image regions.

Abstract: Embodiments of the present invention comprise systems, methods and devices for detection of image regions of various content types using a masking condition and an entropy measure.

Abstract: An encoded, reusable sticker is provided for temporary adhesion to a support surface. The sticker has coded data that is indicative of a sticker identity that is recorded in a computer system. The coded data is a two dimensional pattern and has an interactive element and a generation element. The interactive element is an area of the two dimensional pattern in which the coded data is also indicative of an action to be performed by the computer system. The generation interactive element is an area of the two dimensional pattern that is also indicative of an instruction for the computer system to generate an association between the sticker identity and an object stored in an object repository in the computer system. The sticker also has a visual representation of the action to be performed by the computer system and a visual representation of the generation of an association between the sticker identity and an object.

Abstract: An image processing device includes a weighted image generation section that generates a weighted image in which at least one of an object-of-interest area of an input image and an edge of a background area other than the object-of-interest area is weighted, a composition grid generation section that generates a composition grid that includes grid lines that are weighted, and a composition evaluation section that performs composition evaluation calculations on the input image based on the weighted image and the composition grid.

Abstract: A method and system of manual windowing within a displayed image is disclosed, which include specifying different windows within a displayed image, assigning tags to different windows, assigning processing channels to the windows, processing the different windows on the basis of the assigned tags and assembling the different windows. A scanner is disclosed which sends image data to a manual window construction module which specifies a plurality of windows within a displayed image and which assigns specific tags to each one of the plurality of windows. According to various exemplary embodiments, the windows are processed separately and re-assembled by a processor.

Abstract: A method for optimizing a source document comprising a plurality of pages of content, comprising each of the following, is presented. A source document is obtained. An optimized document is created corresponding to the source document. Thereafter, for each page in the source document, the following are applied. A page record is created for the page. Each page record comprises a word table comprising a list of the page's words in the order that they appear in the page's content. Each page record further comprises a paragraph entry list for the page including a paragraph entry for each paragraph in the page. Each paragraph entry includes a reference to the first and last word of that paragraph in the word table. The page record is compressed using a compression technique. Thereafter, the compressed page record is stored in the optimized document.

Abstract: Structural content filtration to reduce the number of hypotheses for the location of an active object in a recorded image generated by a graphical user interface (GUI) of an application program may be accomplished by transforming the recorded image and a corresponding playback image, determining a sub-image from the transformed recorded image which corresponds to an object to be searched for in the transformed playback image, determining a set of points on the transformed playback image which have appropriate values for matching the sub-image, and filtering hypotheses on the playback image which are more than a selected distance from any one of the subset of points.

Abstract: Methods and apparatus for detecting the presence of combs, determining their shape and removing the combs from a scanned form in an automated manner are described. Horizontal and vertical line feature analysis is combined with knowledge of the usual size, shape, and spacing characteristics of lines which form a comb. Vertical and horizontal lines failing to meet certain characteristics, e.g., size or shape characteristics, are eliminated from consideration. Vertical lines which do not intersect a horizontal line are also eliminated from consideration. Confidence measures for different possible comb shapes are generated and the most probable comb shapes as indicated by the confidence measures are included in a comb list. The comb list may be output for use in further processing, e.g., comb removal and/or data extraction processing.

Abstract: An edge detecting method detects edge segments by searching all of the search lines forming an image from an end of the image in a direction perpendicular to the edges. If a line whose edge segment cannot be detected exists, a search is made in all of the search lines from the vicinity of the center of the image toward the end of the image, whereby edge segments are detected. A linear edge is determined from edge segments. A plurality of edge candidates are obtained from the edge segments for all of the search lines, and an optimum candidate is selected from among the edge candidates. Ruled lines are extracted from the source document in the image, and an optimum candidate is selected based on a comparison with a ruled line. As a result, an edge of the source document can be detected with high accuracy even if an image on a background side is unstable, or if materials of the background and the source document are similar.

Abstract: Traditional graph cuts provides a powerful method for producing semi-automatic image/volume segmentation, but often exceed available computer power. Multi-level methods for coarsening images/volumes are disclosed that allow for faster processing and creation of graph cuts segmentation without sacrificing quality of image. The disclosed graph cuts methods can be used at interactive speeds.

Abstract: Real-time segmentation of foreground from background layers in binocular video sequences may be provided by a segmentation process which may be based on one or more factors including likelihoods for stereo-matching, color, and optionally contrast, which may be fused to infer foreground and/or background layers accurately and efficiently. In one example, the stereo image may be segmented into foreground, background, and/or occluded regions using stereo disparities. The stereo-match likelihood may be fused with a contrast sensitive color model that is initialized or learned from training data. Segmentation may then be solved by an optimization algorithm such as dynamic programming or graph cut. In a second example, the stereo-match likelihood may be marginalized over foreground and background hypotheses, and fused with a contrast-sensitive color model that is initialized or learned from training data. Segmentation may then be solved by an optimization algorithm such as a binary graph cut.

Abstract: Embodiments of the present invention comprise methods and systems for context-based adaptive image processing wherein print job elements are processed according to context, which may be determined by segmentation and analysis of print job elements.

Abstract: In certain embodiments, a method for generating fees using a receiving device, involves distributing censored video from a distributor video to a receiving device; and uncensoring the censored video using the receiving device upon payment of a fee. The receiving device uses overlay data received from the distributor to uncensor the censored video by overlaying the overlay data over the censored video using a video overlay frame to overlay a video frame containing the censored video data in accordance with boundaries determined by an alpha plane within the receiving device. This abstract should not be considered limiting since embodiments consistent with the present invention may involve more, different or fewer elements.

Abstract: In one embodiment, a method includes receiving a portion of text from a document. A document type is associated with the document based on at least one of the portion of text or an identifier associated with the document. The method also includes selecting, based on the document type, a document template having a plurality of sections. Each section from the plurality of sections being associated with a document category. At least one section from the plurality of sections including at least one policy preference.

Abstract: A method and system for identifying multiple questionnaire pages. An embodiment of a method includes receiving a set of coordinates from a capture device, the set of coordinates indicating shapes made on multiple pages of a questionnaire without the use of a graphical user interface, and mapping each shape to a questionnaire answer on a different page. The method enables a processor to accurately and quickly determine which answers were entered on each questionnaire page attached to the capture device, without relying on the user to do so. Exemplary applications include a field survey and inventory.

Abstract: In a method and system for collection of data from documents present in machine-readable form, at least one already-processed document stored as a template and designated as a template document is associated with a document to be processed designated as a read document. Fields for data to be extracted are defined in the template document. Data contained in the read document are already extracted from regions that correspond to the fields in the template document. Should an error have occurred or no suitable template document having been associated given the automatic extraction of the data, the read document is shown on a screen and fields are manually inputted in the read document from which the data are extracted. After the manual input of the fields in the read document, the read document with field specifications is stored as a new template document or the previous template document is corrected corresponding to the newly input fields.

Abstract: In a method for generating a dimensionally tractable set of image layout constraints, a selected constraint is applied to limit a dimension of a highly dimensional solution space. Through random selection one or more unconstrained dimensions of the highly dimensional solution space are identified which, when constrained, will further limit the highly dimensional solution space to a dimensionally tractable solution space. The tractability of the dimensionally tractable solution space is associated with a capability of a computational resource.

Abstract: The present invention provides a document management system including a webpage server, a file receiving server for receiving a document through the webpage server; an optical identification device for performing optical identification on a non-textural content in the received document; a feature mark identifier for setting up a feature mark for the document; and a database for storing the document, which can be output from the database directly or through the file receiving device. The present invention also discloses a remote document management method. The device and method of the present invention have functions of identification, classification, search, and save.

Abstract: In preparation for rendering respective portions of a document via a respective plurality of engines, objects within the document are identified and characterized. A determination is made as to whether gamut variations between the engines might result in objectionable variations in the appearance of rendered versions of identified objects having similar characteristics. For those objects within the document for which the determination is made that variations might be objectionable, a target gamut is selected to be an intersection gamut of the engines to be used to render the document. For those objects within the document for which the determination is made that variations would be unobjectionable, the target gamut is selected to be that of selected individual engines. A system for selecting target gamuts for objects within a document can include an object identifier, a characteristic identifier and a gamut selector.

Abstract: A method of performing an action in relation to a first association between a reusable sticker and an object, the reusable sticker having disposed therein or thereon coded data indicative of a sticker identity and of a first interactive element representing the action, the method including: receiving first indicating data from a sensing device, the first indicating data being at least partially indicative of the sticker identity and the first interactive element, the sensing device, when placed in an operative position relative to the first interactive element, sensing at least some of the coded data to generate the first indicating data; identifying, using the first indicating data, the action and the first association; and, performing the action in relation to the first association.

Abstract: One system to which the present invention is applied obtains the digitized form image of a form, recognizes a character string existing in the obtained form image, extracts a headline wording being a predetermined character string from the recognized character strings, determines a table structure existing in the form image, on the basis of the extracted headline wording and the arrangement of headline wordings in the form image and specifies a correspondence relationship between a headline wording and a character string other than the headline wording that is recognized, using the determination result.

Abstract: Disclosed are systems for and methods of registering (i.e., aligning) a deformable image with a reference image subject to a spatially variant flexibility model and/or a non-Gaussian smoothing model. These systems and methods allow for the consideration of differences between the ways in which structures of interest may move within a patient. The registration includes modifying the deformable image to match similar features in the reference image. The systems include a deformation engine configured for performing a deformation algorithm subject to the spatially variant flexibility and/or smoothing models.

Abstract: A translation device analyses image data that includes characters, to separate and extract a character region, and another region whose structural elements differ from those of the character region; recognizes characters in the region determined to be the character region; extracts a character string related to the other region, from a result of the character recognition, to associate the character string with the character region from which the character string is extracted and the other region; translates the resulting recognized characters; and, based on the association between the character string and the other character region, pastes the resulting translation that includes the character string onto the same page as the other region that is associated with the character string.

Abstract: A system that can convert content and structure of a document from an original format into a target format irrespective of the functional specifics of the original format. The system can automatically infer the content and structure of a document via a rendered format thereby restoring the programmatic functionality of the original file (or generating programmatic functionality of a desired target format) through the novel conversion/import process. The system can extract the document structure (e.g., layout) together with the content in order to effectuate the conversion. Heuristics (e.g., logic and/or reasoning) can be employed to make decisions with respect to importing the document into a target format and/or formats.

Abstract: One embodiment of the present invention provides a system that facilitates detecting white space tables within a document, wherein a white space table is comprised of text arranged in rows and columns, wherein at least some of the rows and columns are separated by bands of white space rather than by lines. The system operates by identifying an area that includes consecutive lines of text objects with an amount of white space between text objects greater than a specified value. Note that a text object is a string of text without an amount of white space greater than the specified value. The system then determines if the text objects on consecutive lines have widths that are within a specified tolerance of each other. If so, the system checks the spaces between the consecutive lines of text objects to determine if they belong to a single white space table or multiple white space tables.

Abstract: A system and a method for publishing a newspaper page or other data through a Web page, to make it available more easily through a network such as the Internet. The data is automatically converted by first rendering the newspaper page into a digital format; the converting the digital format to a basic internal publishing format; and then publishing the data in a number of different possible publishing formats, including a mark-up language document such as a Web page. Features include arrangement of content according to relationships within the information by analyzing the page as distinct objects. Object types include titles, articles, pictures etc. Objects may be categorized, and objects in each category are preferably compressed according to a different image format.

Abstract: A preprocessing section binarizes input image data and calculates a total black pixel ratio. A feature extracting section detects connected components included in the binary image data and detects circumscribing bounding boxes of the connected components. Predetermined connected components are removed from all of the connected components based on the sizes of the detected circumscribing bounding boxes and bounding box black pixel ratios. By using the connected components that remain after removing the unnecessary connected components, a histogram is generated by specifying the sizes of the circumscribing bounding boxes as classes and numbers of the connected components as the frequencies of occurrence. A determining section determines whether the input image data is document image data or non-document image data based on information related to the generated histogram and the total black pixel ratio.

Abstract: An apparatus and method for processing a captured image and, more particularly, for processing a captured image comprising a document. In one embodiment, an apparatus comprising a camera to capture documents is described. In another embodiment, a method for processing a captured image that includes a document comprises the steps of distinguishing an imaged document from its background, adjusting the captured image to reduce distortions created from use of a camera and properly orienting the document is described.

Abstract: Embodiments herein print documents comprising magnetic ink characters, capture overlapping partial images of the MICR line of magnetic ink characters, and stitch these partial images together to form a complete image of the MICR line. This allows the embodiments to accumulate the routing number, the account number, the check number, and the amount data in the magnetic ink characters on each document to produce image-based data including accumulated amounts (e.g. ledger). Then, the image-based data including the accumulated amounts can be compared to the anticipated data from the original source of information to verify the accuracy of the printing operations.

Abstract: Methods and system for automatic identification of repeating patterns of slanted stripe features (marks) on an item.

Abstract: A system and method are provided for processing a document. The method comprises: marking an area of a document, using at least one predetermined markup symbol; identifying a markup symbol; using the identified markup symbol to locate the document area; determining characters in the document area; and, processing the determined characters. Fields that can be processed include title, keywords, author, signature, document name, category, subject, or document status, to name but a few examples. In some aspects, marking an area of a document, using at least one predetermined markup symbol includes: electronically marking using a medium such as a touchscreen, pressure pad, or a mouse. Alternately, the paper document can be physically marked (with a pen or pencil) before scanning. Typically, marking an area of a document includes forming a markup character proximate to an enclosing figure. A wide range of markup characters can be defined cross-referenced to document processes.

Abstract: A method and system for performing image mark recognition for a document is disclosed. A document is scanned into a digital image. Reference image marks are sensed from the digital image. The reference image marks may include trigger row marks and/or corner/crop marks. Coordinates denoting the location of cells within the digital image are determined base on the locations of the reference image marks. Response marks are evaluated for darkness, opacity, and/or grayness on a pixel-by-pixel basis. The response marks are each assigned a percentage value based on the ratio between a total color value for a cell and a maximum color value for a cell.

Abstract: This disclosure relates to an image processing apparatus or an image processing method, and takes as its technical problem the fact that conventionally it is difficult to perform a process of carving out a rectangle with high precision in a noisy environment. Therefore, as to an image made by binarizing an original image, histograms of the numbers of horizontal and vertical black pixels thereof are created and trapezoidal approximation is performed on the histograms so as to detect vertex coordinates indicating an extent of a rectangular area in the binary image. An inclination direction of the rectangular area in the extent is detected so that the rectangular area can be detected from the binary image with high precision.

Abstract: A system for learning an attention model for an image based on user navigation actions while viewing the image is provided. An attention model learning system generates an initial attention model based on static features derived from the image. The learning system logs the actions of users as they view the image. The learning system identifies from the actions of the users those areas of the image that may be of user interest: After the learning system identifies areas that may be of user interest, the learning system attempts to identify attention objects based on those identified areas. After identifying the user attention objects, the learning system updates the attention model for the image to reflect the additional information provided by the user attention objects.

Abstract: An image processing apparatus can perform highly efficient coding that generates less visible distortion. The image processing apparatus includes a pre-processing unit configured to reduce the data amount of an input image and output the image after reducing its data amount and a coding unit configured to receive an output from the pre-processing unit and output coded data. The pre-processing unit includes a region segmentation unit configured to divide a frame image into regions, a filter setting unit configured to set filter strength, an intermediate region setting unit configured to generate an intermediate filter region in accordance with the filter strength, and a filter processing unit configured to perform filter processing at the set filter strength. Filter regions and filter strength can be set in accordance with the code length.

Abstract: A device that performs automatic red-eye correction on a digital image saves the original image data and a location indicator from each portion of the digital image on which red-eye correction is performed to facilitate the subsequent undoing of red-eye correction. Red-eye correction may be undone at any time after red-eye correction has been performed by replacing the corrected image data with the original image data that is stored with the corrected digital image, and the device on which red-eye correction is undone may be distinct from the device that captured the digital image and from the device that performed red-eye correction on the digital image.

Abstract: An OCR percentage matching algorithm achieves a significant reduction in false mismatches accounting for combinations of unprocessed spaces, missing characters, extra characters and character substitution errors during the OCR scanning processing and allows for a specified percentage of the OCR character scan rather the entire OCR character scan to be the same as the expected character string to declare a match.

Abstract: A method of utilizing data extracted from an electronic image includes receiving an electronic image associated with a subclient and receiving a plurality of data extracted from the electronic image and organized into predefined fields associated with the subclient. The method also includes displaying the electronic image in a first portion of a user interface and, in a second portion of the user interface, a field table containing the plurality of data and associated field names.

Abstract: The definition of regions or volumes of interest and the delineation of objects of interest are important and frequently performed tasks in clinical imaging. However, today's solutions for this task are often time-consuming and cumbersome for the clinician. The present invention provides an alternative approach that is intuitive and works even for noisy images where automatic segmentation approaches usually fail. The disclosed systems and methods automatically translate mouse movements relative to an imaginary x-axis and y-axis into modifications in the threshold and scale (i.e., shape and size) of a boundary delineation, thereby permitting a system user/clinician to rapidly arrive at a desired ROI and/or VOI.

Abstract: In one embodiment, the invention discloses, a method for processing a document image. The method comprises segmenting the document image into a picture component and a non-picture component; compressing the non-picture component; and saving the uncompressed picture component and the compressed non-picture component in memory so that the document image may be recomposed to form a recomposed image based on the uncompressed picture component and the compressed non-picture component.