Abstract: A method and system for segmenting a text into a plurality of sections is provided. The text may be received in the form of an image. The method involves receiving one or more input labels from a user corresponding to one or more segmentation points of a plurality of segmentation points of the text. The plurality of segmentation points of the text are obtained by applying one or more segmentation heuristics over the text. The one or more input labels provided by the user are utilized to label the plurality of segmentation points of the text. In response to labeling, validation is performed to identify whether a segmentation point of the plurality of segmentation points is a valid segmentation point. Thereafter, based on the validation, a set of valid segmentation points is updated with one or more segmentation points of the plurality of segmentation points. The set of valid segmentation points facilitates segmentation of the text for recognizing the plurality of sections.

Abstract: A method for text character identification. The method acquires multiple connected components (CCs) in a binary image, and each CC has a pattern property value. The method determines at least one property limit based on the pattern property values, generates a filtering rule according to the property limit, and determines whether each of the CCs is a text character according to the filtering rule.

Abstract: The image processing apparatus and method, and the program and the recording medium according to the present invention can make the coefficient vector into high precision by noise elimination or correction utilizing the mutual correlation of the divided image areas in the intermediate eigenspace, and allows relaxation of the input condition and robustness. The high correlation in the divided image areas in the intermediate eigenspace can reduce the divided image areas to be processed, and actualize reduction in processing load and enhancement of the processing speed.

Abstract: Every time clustering processing for a predetermined number of pixels is complete, a small cluster having the number of allocated pixels, which is equal to or smaller than a pixel count threshold, is discriminated. The small cluster, which is discriminated to have the number of allocated pixels equal to or smaller than the pixel count threshold, is merged to a cluster having the nearest representative feature vector. With this arrangement, the number of clusters which are to undergo distance calculations of feature vectors is reduced. According to this arrangement, region segmentation of an image can be executed faster by the clustering processing.

Abstract: A method of processing uniform mailpieces referred to as a “run” of mailpieces, during which method OCR is performed for recognizing certain information in a zone of interest of an image of each mailpiece, and during which method the following steps are performed: a) initializing a matrix accumulator associated with said run and including unitary accumulation elements that correspond to the pixels of the image; b) consolidating said matrix accumulator by incrementing certain unitary accumulation elements by deriving an indication of the spatial position of a block of pixels in which said certain information has been recognized unambiguously, or by using construction and local graphical correlation of blocks of image pixels to derive an optical flow map indicating local graphical movements; and c) defining, in the OCR processing, said zone of interest on the basis of the unitary accumulation elements of the consolidated matrix accumulator that present extreme accumulation values.

Abstract: An image document processing device extracts a character sequence image having M number of characters in an image document, divides the image into individual character images, extracts features of the individual character images, and based on the features, selects N (N is an integer more than 1) character images in the order of degree of matching from a font-feature dictionary for storing features of all character images according to fonts, and generates an M×N index matrix for the extracted character sequence. In searching, the device searches an index-information storage section with respect to each search character included in a search keyword in an input search expression, and extracts an image document including an index matrix including the search keyword. This provides an image document processing device and an image document processing method each allowing indexing not requiring user's operation and each allowing highly precise searching without OCR recognition.

Abstract: A headline-region initial processing section clips a headline-region image in an image document, divides the image into individual character images, and extracts features of the individual character images. Based on the features, a candidate-character-sequence generating section selects N (N is an integer more than 1) character images as candidate characters in the order of degree of matching from a font-feature dictionary for storing features of individual character images, and generates M×N index matrix where M is the number of characters in an extracted character sequence. Based on the index matrix, a document-name generating section generates a meaningful document name according to the image document. An image-document-DB management section manages accumulated image documents using the document name.

Abstract: Methods and systems for segmenting printed media pages into individual articles quickly and efficiently. A printed media based image that may include a variety of columns, headlines, images, and text is input into the system which comprises a block segmenter and a article segmenter system. The block segmenter identifies and produces blocks of textual content from a printed media image while the article segmenter system determines which blocks of textual content belong to one or more articles in the printed media image based on a classifier algorithm. A method for segmenting printed media pages into individual articles is also presented.

Abstract: An edge image generator in an image processing determining apparatus extracts multiple edges from an image included in display data output from an external device, such as a terminal unit, a navigation unit, or an imaging unit. Then, the edge image generator selects certain edges from the extracted multiple edges by a certain selection method matched with characteristics of the external device to generate an edge image.

Abstract: Embodiments of the present invention comprise systems and methods for refining text-detection results for a digital image.

Abstract: The watermarked information embedding apparatus which inputs an image and embeds watermarked information in the input image, comprises: picture element determining means which determines whether it is a picture element constituting a background image for each of picture elements which constitute the input image; background picture element removing means which removes all of background picture elements determined as picture elements constituting the background image by the picture element determining means; and watermarked information embedding means which embeds the watermarked information in an image constituted by a picture element from which the background picture element constituting the input image is removed by the background picture element removing means.

Abstract: A method for extracting a character string from print data rasterizes the print data into a raster image. Then, the method divides the raster image into a character region and non-character region and determines character data used for metadata based on the raster image of the character region and character data extracted from the print data and drawn at approximately the same position as the character region.

Abstract: A document processing system for accurately and efficiently analyzing documents and methods for making and using same. Each incoming document includes at least one section of textual content and is provided in an electronic form or as a paper-based document that is converted into an electronic form. Since many categories of documents, such as legal and accounting documents, often include one or more common text sections with similar textual content, the document processing system compares the documents to identify and classify the common text sections. The document comparison can be further enhanced by dividing the document into document segments and comparing the document segments; whereas, the conversion of paper-based documents likewise can be improved by comparing the resultant electronic document with a library of standard phrases, sentences, and paragraphs. The document processing system thereby enables an image of the document to be manipulated, as desired, to facilitate its review.

Abstract: An application folder associated with a client PC and an application software of the client PC is generated in a storage section of a station PC. Scan data stored in the application folder is then moved to an application data folder of the client PC, which folder corresponds to the client PC and application software associated with the application folder. As a result, in a network scanner system in which a scanner apparatus is connected to the client PC over a network, it is possible to efficiently store scan data read out by the scanner apparatus and perform data processing to the scan data by an application software.

Abstract: An interactive system provides for increasing retrieval performance of images depicting text by allowing users to provide relevance feedback on words contained in the images. The system includes a user interface through which the user queries the system with query terms for images contained in the system. Word image suggestions are displayed to the user through the user interface, where each word image suggestion contains the same or slightly variant text as recognized from the word image by the system than the particular query terms. Word image suggestions can be included in the system by the user to increase system recall of images for the one or more query terms and can be excluded from the system by the user to increase precision of image retrieval results for particular query terms.

Abstract: The present invention discloses an on-line identifying method of hand-written Arabic letter. The advantage of the present invention is that the multilayer coarse classification algorithm based on the local characteristic of Arabic letter fully utilize the various local characteristics of Arabic letter, obtain the first candidate letter aggregation matching with the inputted hand-written Arabic letter according to the first level coarse classification formed by the stroke number of letter, and then obtain the second candidate letter aggregation matching with inputted hand-written Arabic letter according to the other local characteristics and the first candidate letter aggregation. The application of the algorithm enables that the inputted hand-written Arabic letter only need to match with the standard letter stored in the predetermined letter library and the corresponding standard letters of the second candidate letter aggregation.

Abstract: Systems and methods for replacing non-image text are provided. One method for replacing non-image text includes padding a first data representing an image of text to create an image segment. The method includes replacing a second data representing non-image text with the image segment.

Abstract: Disclosed herein is a method, computer system and computer program product for identifying a writing system associated with a document image containing one or more words written in the writing system. Initially, a document image fragment is identified based on the document image, wherein the document image fragment contains one or more pixels from one or more of the words in the document image. A set of sequential features associated with the document image fragment is generated, wherein each sequential feature describes one dimensional graphic information derived from the one or more pixels in the document image fragment. A classification score for the document image fragment is generated responsive at least in part to the set of sequential features, the classification score indicating a likelihood that the document image fragment is written in the writing system.

Abstract: A method of characterizing a word image includes traversing the word image in steps with a window and at each of a plurality of the steps, identifying a window image. For each of the plurality of window images, a feature is extracted. The word image is characterized, based on the features extracted from the plurality of window images, wherein the features are considered as a loose collection with associated sequential information.

Abstract: A method and system for preprocessing an image for Optical Character Recognition (OCR), wherein the image includes a plurality of columns is disclosed. Each column includes one or more of Arabic text and non-text items. The method includes determining a plurality of components associated with one or more of the Arabic text and the non-text items, wherein a component includes a set of connected pixels. On determining the plurality of components, a line height and a column spacing is determined for the plurality of components. The plurality of components are then associated with a column of the plurality of columns based on the line height and the column spacing. Subsequently, a set of characteristic parameters are calculated for each column and the plurality of components of each column are merged based on the set of characteristic parameters to form sub-words and words.

Abstract: When a list of file names is to be displayed on a display device, a comparison is made between a necessary display width of each of the file names and a width of a display area of the display device. For each of the file names having a necessary display width greater than the width of the display area, it is checked whether the file name contains a particular character string portion of a predetermined type, and, if so, the file name is displayed in the list in a partly-omitted display style where a leading end portion, particular character string portion and extension of the file name are left in the list with the other part of the character string omitted. The particular character string portion can function as an important element for identifying the data item in question.

Abstract: The invention discloses a method for moving targets tracking and number counting, comprising the steps of: a). acquiring continuously the video images comprising moving targets; b). acquiring the video image of a current frame, and pre-processing the video image of the current frame; c). segmenting the target region of the processed image, and extracting the target region; d). matching the target region of the current frame obtained in step c) with that of the previous frame based on an online feature selection to establish a match tracking link; and e). determining the number of the targets corresponding to each match tracking link based on the target region tracks recorded by the match tracking link.

Abstract: The present disclosure provides a computer-implemented method of translating an image-based electronic document into a text-based electronic document. The method includes electronically scanning an image-based document to determine positions of word images in the image-based document. The method also includes extracting the word images from the image-based document and storing the word images to an electronic storage device. The method also includes grouping a subset of the word images into a word cluster based on a similarity of the word images, wherein the word images in the word cluster correspond to a same actual word. The method also includes generating a character-encoded transcription for the word cluster based on the word images in the word cluster. The method also includes adding the character-encoded transcription to a text-based electronic document at locations corresponding to the positions of the word images in the image-based document.

Abstract: Aspects of the present invention relate to systems and methods for determining text orientation in a digital image.

Abstract: The present invention aims to prevent a problem that an image on a document sheet is erased due to misdetection of a line-shaped noise. A copy machine 1 compares RGB values of a target pixel with averaged RGB values (Step S103). If only one of the RGB values has a difference that is greater than a prescribed value Ref2 (Step S103: YES), the copy machine 1 extracts the target pixel as a line-shaped noise pixel, and moves to a line-shaped noise correction (Step S108) while holding the address of the target pixel in a line-shaped noise address storing area 49b. If two of the RGB values have differences (Step S103: NO, Step S104: YES) and a difference between these two of the RGB values is no greater than a prescribed value Ref3 (Step S105: YES), the copy machine 1 extracts the target pixel as a line-shaped noise pixel, and moves to the line-shaped noise correction (Step S108) while holding the address of the target pixel in the line-shaped noise address storing area 49b.

Abstract: A method and system is provided for segmenting scanned image data in accordance with mixed raster content processing for more efficient processing of non-uniform color touching objects. The scanned data is segmented to background and foreground layers wherein the foreground layer is comprised of a plurality of objects such as text characters. At least one of the plurality of objects is identified as being non-uniform in color. The non-uniform color object is partitioned into a plurality of sub-objects of predetermined size pixel blocks. The sub-objects are then clustered by uniform color and coded with a binary compression algorithm as a foreground layer segment. Non-uniform color sub-objects are alternatively discarded for compression with the background layer algorithm, or processed for determination of a particular color based upon the color of a plurality of pixels within the sub-object.

Abstract: A method for character string recognition may include processing image data into black-and-white binary image data, calculating vertical projection data of the binary image data in a vertical direction perpendicular to a direction of the character string while shifting the binary image data, detecting positions exceeding a prescribed border judgment threshold value in the vertical projection data, judging validity of the border judgment threshold value, and deciding whether to segment characters out of the character string based on whether the border judgment threshold value is valid.

Abstract: A preprocessing section binarizes input image data and calculates a total black pixel ratio. A feature extracting section detects connected components contained in the binarized image data and detects circumscribing bounding boxes that circumscribe these connected components, respectively. Based on sizes of the circumscribing bounding boxes detected and numbers of black pixels contained therein, predetermined connected components are removed. A determining section generates an edge map by using the residual connected components, and performs two-dimensional fast Fourier transform thereon to generate spectral data. The determining section performs two-dimensional fast Fourier transform on template images to generate spectral data. The determining section determines, based on these pieces of spectral data, whether or not a circular shape is contained in the input image data.

Abstract: A document processing system for accurately and efficiently analyzing documents and methods for making and using same. Each incoming document includes at least one section of textual content and is provided in an electronic form or as a paper-based document that is converted into an electronic form. Since many categories of documents, such as legal and accounting documents, often include one or more common text sections with similar textual content, the document processing system compares the documents to identify and classify the common text sections. The document comparison can be further enhanced by dividing the document into document segments and comparing the document segments; whereas, the conversion of paper-based documents likewise can be improved by comparing the resultant electronic document with a library of standard phrases, sentences, and paragraphs. The document processing system thereby enables an image of the document to be manipulated, as desired, to facilitate its review.

Abstract: The present invention firstly roughly classifies an analysis range specified by the operator in the color image data of a form into background, a character frame and a character, precisely specifies a character frame on the basis of the classification result, eliminates the character from the color image data from which the background is eliminated and recognizes the remaining character.

Abstract: An image processing apparatus includes: an imaging information calculation unit acquiring a first image and higher-resolution second images, and calculating coordinate positions of the second images to the first image and differences in imaging direction between second cameras and a first camera; an eyepoint conversion unit generating eyepoint conversion images obtained by converting the second images based on the differences in imaging direction so that eyepoints of the second cameras coincide with an eyepoint of the first camera and matching the first image with the eyepoint conversion images to calculate phase deviations of the eyepoint conversion images from the first image; and an image synthesizing unit extracting high-frequency images, having frequency components higher than or equal to a predetermined frequency band, from the second images, and pasting the high-frequency images at the coordinate positions in correspondence with the first image to eliminate the phase deviations to generate a synthesize

Abstract: Script-agnostic text reflow technique embodiments are presented that generally reflow text found in an image of a document in a manner that functions across multiple scripts, multiple fonts of a script and multiple languages using the same script. This generally involves segmenting regions of text in a document image into individual words and doing this without relying on any script-specific characteristics or requiring any form of character recognition. While segmenting text, the possible presence of accents, diacritics and punctuation marks is considered.

Abstract: An overall processing time to rasterize, at the first device, the electronic document to be rendered is computed. Also, a rendering time to render, at the first device, the electronic document to be rendered is computed. When the overall processing time to rasterize at the first device is greater than the rendering time to render at the first device, the electronic document to be rendered is parsed into a first document and sub-documents. A productivity capacity of each node is determined, the productivity capacity being a measured of the processing power of the node and the communication cost of exchanging information between the first device and the node. A sub-document is rasterized at a node when a productivity capacity of the node reduces the processing time to rasterize the electronic document to be rendered to be less than the computed overall processing time.

Abstract: A system and method, which enables precise identification of characters contained in vehicle license plates, container I.D, chassis I.D, aircraft serial number and other such identification markings. The system can process these identified characters and operate devices, such as access control operations, traffic systems and vehicle and container tracking and management systems, and provide records of all markings together with their images.

Abstract: Systems and methods for character-by-character alignment of two character sequences (such as OCR output from a scanned document and an electronic version of the same document) using a Hidden Markov Model (HMM) in a hierarchical fashion are disclosed. The method may include aligning two character sequences utilizing multiple hierarchical levels. For each hierarchical level above a final hierarchical level, the aligning may include parsing character subsequences from the two character sequences, performing an alignment of the character subsequences, and designating aligned character subsequences as the anchors, the parsing and performing the alignment being between the anchors generated from an immediately previous hierarchical level if the current hierarchical level is below the first hierarchical level. For the final hierarchical level, the aligning includes performing a character-by-character alignment of characters between anchors generated from the immediately previous hierarchical level.

Abstract: A method for correcting distortions in a scanned image of a page, paragraph, sentence or other portion of text is disclosed. The method comprises identifying at least one set of collinear elements in the scanned image; and generating a corrected image based on the scanned image including for at least some of the collinear elements in each set applying a spatial location correction to position all collinear elements in the set on a common horizontal rectilinear base line in the corrected image.

Abstract: The present invention relates to an image processing method, an image processing apparatus and an image processing program for dealing with inverted characters (outlined characters) constituted by white pixels on a black ground in a tree structure same as that of normal characters constituted by black pixels on a white ground. In the present invention, black pixel blocks and white pixel blocks are sampled recursively from a binary image, tree structure data indicating a positional relation between the sampled black pixel blocks and white pixel blocks is created, an inverted image is created by white-black-inverting the insides of black pixel blocks that can include inverted characters, of black pixel blocks included in the tree structure data, white pixel blocks and black pixel blacks are sampled from the created inverted image, and data regarding the sampled white pixel blocks and black pixel blocs is added to corresponding nodes of the tree structure data.

Abstract: A method and a system are disclosed for labeling an anatomical point associated with a lesion in an organ such as a lung. The method includes: a segmentation of a vessel tree anatomical structure starting from an autonomously determined initial image point; labeling the vessel segments of the vessel tree segmentation with segment labels based on a priori anatomical knowledge, thereby creating an individualized anatomical model; receiving a user-specified image point having a location from a user and locating a nearby vessel structure; tracking along the vessel structure in a direction towards a root of a parent vessel tree until a prior labeled vessel segment is encountered in the anatomical model, and assigning the label of the encountered prior labeled vessel segment from the anatomical model as an anatomical location label of the user-specified image point.

Abstract: The invention relates to the authentication of users for a multi-function peripheral (MFP) device using handwritten signatures. Systems and methods are disclosed which relate to a MFP that conditions access to MFP operations based on an authenticating process that compares a prospective user's signature to previously saved signatures. The signatures are communicated to the MFP using the MFP's native scanning function.

Abstract: Aspects of the present invention relate to systems and methods for determining text orientation in a digital image.

Abstract: The present invention provides a technique of accurately extracting areas of characters included in a captured image even in a case where noise or dirt of a relatively large area occurs in a background image. An integrated pixel value is obtained by integrating pixel values in a character extracting direction B for pixel positions in a character string direction A of an image including a character string. A standard deviation value is calculated along the character extracting direction for pixel positions in a character string direction A. The integrated pixel value and the standard deviation value are combined for pixel positions in a character string direction A. A threshold is set automatically or manually. A part of pixel positions in a character string direction A having the combined value of the integrated pixel value and the standard deviation value higher than the threshold is recognized as a character area to be extracted.

Abstract: The present invention provides a technique of accurately extracting areas of characters included in a captured image even in a case where noise or dirt of a relatively large area occurs in a background image. A pixel value integration evaluation value is obtained by integrating pixel values in a character extracting direction B at each of the pixel positions in a character string direction A of an image including a character string. A waveform of the value is expressed as waveform data. A first threshold and a second threshold are set for the waveform data. An area in which the waveform data exceeds the first threshold is set as a character candidate area. In a case where an area in which the pixel value integration evaluation value exceeds the second threshold exists in the character candidate areas, the character candidate area is regarded as a true character area and the characters are extracted.

Abstract: A method and apparatus are provided for generating a character collage message. A character is recognized from an image. A region is extracted from the image to create a character image. The region includes the recognized character. The created character image is stored in a memory. At least the character image is output to an output unit as the character collage message in accordance with input of one or more characters through an input unit. At least one of the one or more characters corresponds to the character image, and the character image is output to the output unit as a substitute for the at least one of the one or more characters.

Abstract: To handle static text and logos in stabilized images without destabilizing the static text and logos, a method of handling overlay subpictures in stabilized images includes detecting an overlay subpicture in an input image, separating the overlay subpicture from the input image, stabilizing the input image to form a stabilized image, and merging the overlay subpicture with the stabilized image to obtain an output image.

Abstract: The present disclosure provides a computer-implemented method of translating an image-based electronic document into a text-based electronic document. The method includes electronically scanning an image-based document to determine positions of word images in the image-based document. The method also includes extracting the word images from the image-based document and storing the word images to an electronic storage device. The method also includes grouping a subset of the word images into a word cluster based on a similarity of the word images, wherein the word images in the word cluster correspond to a same actual word. The method also includes generating a character-encoded transcription for the word cluster based on the word images in the word cluster. The method also includes adding the character-encoded transcription to a text-based electronic document at locations corresponding to the positions of the word images in the image-based document.

Abstract: Methods, systems, and apparatus including computer program products for using extracted image text are provided. In one implementation, a computer-implemented method is provided. The method includes receiving an input of one or more image search terms and identifying keywords from the received one or more image search terms. The method also includes searching a collection of keywords including keywords extracted from image text, retrieving an image associated with extracted image text corresponding to one or more of the image search terms, and presenting the image.

Abstract: A method and system generates fine-grained fingerprints for identifying content in a rendered document. It includes applying image-based techniques to identify patterns in a document rendered by an electronic document rendering system, irrespective of a file format in which the rendered document was electronically created. The applying of the image-based technique includes identifying candidate keypoints at locations in a local image neighborhood of the document, and combining the locations of the candidate keypoints to form a fine-grained fingerprint identifying patterns representing content in the document.

Abstract: An apparatus includes a content acquisition unit configured to acquire content data contained in image data, an extraction unit configured to extract a keyword from the image data, a setting unit configured to set acceptance or rejection of modification of the keyword according to a keyword extracted by the extraction unit, and a storage unit configured to store the data of the content, the keyword, and the setting of acceptance or rejection of modification in association with each other.

Abstract: A first aspect of the invention relates to a method for creating a binary mask image from an a inputted digital image of a scanned document, comprising the steps of creating a binarized image by binarizing the inputted digital image, detecting first text regions representing light text on a dark background, and inverting the first text regions, such that the inverted first text regions are interpretable in the same way as dark text on a light background. A second aspect of the invention relates to a method for comparing in a binary image a first pixel blob with a second pixel blob to determine whether they represent matching symbols, comprising the steps of detecting a line in one blob not present in the other and/or determining if one of the blobs represents an italicized symbol where the other does not.

Abstract: A method for segmenting a small feature in a multidimensional digital array of intensity values in a data processor computes an edge metric along each ray of a plurality of multidimensional rays originating at a local intensity extreme (local maximum or minimum). A multidimensional point corresponding to a maximum edge metric on each said ray is identified as a ray edge point. Every point on each ray from the local extreme to the ray edge point is labeled as part of the small object. Further points on the feature are grown by labeling an unlabeled point if the unlabeled point is adjacent to a labeled point, and the unlabeled point has a more extreme intensity than the labeled point, and the unlabeled point is closer than the labeled point to the local extreme. The resulting segmentation is quick, and identifies boundaries of small features analogous to boundaries identified by human analysts, and does not require statistical parameterizations or thresholds manually determined by a user.