Abstract: Some embodiments provide a for analyzing a document that includes a number of primitive elements. The method identifies boundaries between sets of primitive elements and identifies regions bounded by the boundaries. The method uses the identified regions to define structural elements for the document. The method defines a structured document based on the primitive elements and the structural elements.

Abstract: Candidate identification utilizing fingerprint identification is disclosed.

Abstract: Systems and methods analyze the physical structure of text rows in a document image, including the positions of one or more alignments of one or more character blocks in one or more text rows of the document image. The systems and methods determine one or more groups of text rows that are placed into a class based on the structures of the text rows, such as the positions of the one or more alignments of the one or more character blocks in each text row.

Abstract: A method comprises extracting a local identifier (130, 730a, 730b) from an image (100, 500, 700), the image (100, 500, 700) also having positional data (120) relating to the location at which the image (100, 500, 700) was captured; and associating the extracted local identifier (130, 730a, 730b) with the corresponding positional data (120) to allow for associating the extracted local identifier with a digital map (300, 600, 800).

Abstract: Aspects of the present invention are related to systems and methods for automatically extracting, from a document image, references to relevant external content and automatically retrieving the external content associated with the references.

Abstract: A method for automatically recognizing Arabic text includes building an Arabic corpus comprising Arabic text files written in different writing styles and ground truths corresponding to each of the Arabic text files, storing writing-style indices in association with the Arabic text files, digitizing an Arabic word to form an array of pixels, dividing the Arabic word into line images, forming a text feature vector from the line images, training a Hidden Markov Model using the Arabic text files and ground truths in the Arabic corpus in accordance with the writing-style indices, and feeding the text feature vector into a Hidden Markov Model to recognize the Arabic words.

Abstract: An electronic book can be paginated by reference to a print version of the same book. Pages of the print version are scanned to obtain text strings and page labels corresponding to each of the pages. The text strings are then compared to the electronic book to find the best matching positions within the electronic book. The matching positions within the electronic book are then associated with the page numbers of the pages from which the matching text strings were obtained. Autocorrelation can be used to determine matching positions.

Abstract: An imaging-based bar code reader that includes an imaging and decoding system. Focusing optics and a sensor array define a field of view. A data processor has a memory for storing a pattern definition of previously imaged OCR characters and comparing a format of said previously stored characters to a present image to determine a character content of the present image.

Abstract: A method of defining model characters of a font. The method includes receiving a string of characters, receiving an image that includes an occurrence of the string, identifying objects in the image, determining, for each respective object, which of the objects satisfies first criteria indicating that the respective object likely corresponds to a character in the string, determining, for each respective object satisfying the first criteria, which of the objects satisfies second criteria indicating that the respective object belongs to a sequence of objects likely to correspond to the string, and defining, for each respective object satisfying the second criteria, a model character for each character of the string based upon a corresponding object of the sequence of objects. The first criteria may include aspect ratio criterion, size criterion, or both, and the second criteria may include alignment criterion, spacing criterion contrast criterion, encompassment criterion, or combinations thereof.

Abstract: Disclosed is a character recognition preprocessing method and apparatus for correcting a nonlinear character string into a linear character string. A binarized character string region is divided into character regions on a character-by-character basis. Upper and lower feature points of each character region are derived, and an upper boundary line, which is a curve connecting the upper feature points of the character regions, and a lower boundary line, which is a curve connecting the lower feature points of the character regions, are generated by applying cubic spline interpolation. Nonlinearity is corrected through adaptive region enlargement by using the maximum horizontal length and the maximum height of the divided character regions.

Abstract: Recognition of numerical characters is disclosed, including: extracting a subimage from a received image comprising information pertaining to a plurality of numerical characters, wherein the extracted subimage is associated with one of the plurality of numerical characters; and performing recognition based at least in part on a set of topological information associated with the subimage, including: processing the subimage to obtain the set of topological information associated with the subimage; comparing the set of topological information associated with the subimage with a preset set of stored topological information; determining that in the event that the set of topological information associated with the subimage matches the preset set of stored topological information, the subimage is associated with a recognized numerical character associated with the preset set of stored topological information.

Abstract: A system for processing text captured from rendered documents is described. The system receives a sequence of one or more words optically or acoustically captured from a rendered document by a user. The system identifies among words of the sequence a word with which an action has been associated. The system then performs the associated action with respect to the user.

Abstract: A system including a data processing system, a network interface for communicating over a network, and a program memory storing instructions configured to cause the data processing system to implement a method for extracting textual information from images of a document containing text characters. The method includes receiving a plurality of digital images of the document over the network. Each of the captured digital images is automatically analyzed using an optical character recognition process to determine extracted textual data. The extracted textual data for the captured digital images are merged to determine the textual information for the document, wherein differences between the extracted textual data for the captured digital images are analyzed to determine the textual information for the document.

Abstract: Systems and methods are provided for challenge/response animation. In one implementation, a request for protected content may be received from a client, and the protected content may comprise data. A challenge phrase comprising a plurality of characters may be determined, and a computer processor may divide the challenge phrase into at least two character subsets selected from the characters comprising the challenge phrase. Each of the at least two character subsets may include less than all of the characters comprising the challenge phrase. The at least two character subsets may be sent to the client in response to the request; and an answer to the challenge phrase may be received from the client in response to the at least two character subsets. Access to the protected content may be limited based on whether the answer correctly solves the challenge phrase.

Abstract: Product images are used in conjunction with textual descriptions to improve classifications of product offerings. By combining cues from both text and image descriptions associated with products, implementations enhance both the precision and recall of product description classifications within the context of web-based commerce search. Several implementations are directed to improving those areas where text-only approaches are most unreliable. For example, several implementations use image signals to complement text classifiers and improve overall product classification in situations where brief textual product descriptions use vocabulary that overlaps with multiple diverse categories. Other implementations are directed to using text and images “training sets” to improve automated classifiers including text-only classifiers.

Abstract: An image processing apparatus segments Western and hieroglyphic portions of textual lines. The apparatus includes an input component that receives an input image having at least one textual line. The apparatus also includes an inter-character break identifier component that identifies candidate inter-character breaks along a textual line and an inter-character break classifier component. The inter-character break classifier component classifies each of the candidate inter-character breaks as an actual break, a non-break or an indeterminate break based at least in part on the geometrical properties of each respective candidate inter-character break and the bounding boxes adjacent thereto. A character recognition component recognizes the candidate characters based at least in part on a feature set extracted from each respective candidate character that can be histogram features, Gabor features or any other feature set applicable to character recognition.

Abstract: A character recognition apparatus and method based on a character orientation are provided, in which an input image is binarized, at least one character area is extracted from the binarized image, a slope value of the extracted at least one character area is calculated, the calculated slope value is set as a character feature value, and a character is recognized by using a neural network for recognizing a plurality of characters by receiving the set character feature value. Accordingly, the probability of wrongly recognizing a similar character decreases, and a recognition ratio of each character increases.

Abstract: An information processing apparatus includes an identifying unit, a character recognition unit, an obtaining unit, a correcting unit, and an output unit. The identifying unit identifies a still image included in a moving image. The character recognition unit performs character recognition on the still image identified by the identifying unit. The obtaining unit obtains information about the moving image. The correcting unit corrects, on the basis of the information obtained by the obtaining unit, a character recognition result generated by the character recognition unit. The output unit outputs the character recognition result corrected by the correcting unit in association with the moving image.

Abstract: In an extracting step, the extracting portion obtains a linked component composed of a plurality of mutually linking pixels from a character string region composed of a plurality of characters, and extracts section elements from the character string region, the section elements each being surrounded by a circumscribing figure circumscribing to the linked component. In the first altering step, the first altering portion combines section elements at least having a mutually overlapping part among the extracted section elements so as to prepare a new section element. In the first selecting step, the first selecting portion determines a reference size in advance and selects section elements having a size greater than the reference size, from among the section elements altered in the first altering step.

Abstract: A character region extracting apparatus and method which extract a character region through the calculation of character stroke widths are provided. The method includes producing a binary image including a candidate character region from an original image; extracting a character outline from the candidate character region; acquires character outline information for the extracted outline; setting a representative character stroke width and a representative character angle in each of the pixels forming the outline, based on the character outline information; and determining a character existing region in the candidate character region by confirming the ratio of effective representative stroke widths and effective angles as compared to the entire length of the outline. Accordingly, it is possible to efficiently determine whether one or more characters exist in the candidate character region.

Abstract: An example method of entering calendar events into an electronic calendar involves capturing a digital image of a document that contains a written calendar event; analyzing the digital image of the document containing the written calendar event to extract text information appearing on the digital image of the document; matching the extracted text information in the digital image of the written calendar event document to a date in the electronic calendar; and populating the extracted text information to at least one field of the electronic calendar associated with the date.

Abstract: A device includes a character-data rotating section that rotates a regular-position character by a predetermined angle with respect to a reference point that is the center point of the background area of the regular-position character by using regular-position character data having a rotation angle of 0° and a center-point matching processing section that horizontally and/or vertically enlarges the background area of the rotated character data to cause the center point of the rotated character and the center point of BMP data to match each other even with respect to rotated character data. Thus, when multiple pieces of character data are arranged so that the center points thereof lie on a reference line, not only are the center points of the characters aligned along the reference line, but also bottom portions of the characters aligned with respect to the reference line.

Abstract: An apparatus according to the present invention comprises: a region extraction unit configured to extract region data for each object from document image data including tables; a table structure analysis unit configured to analyze the region data relating to table objects out of the extracted region data and extract table structure information on each of the table objects; a sheet generation unit configured to generate a display sheet for reproducing a layout of the object in the document image data and an edit sheet for each table for editing the table, by using the region data and the table structure information on each object; and an electronic-document generation unit configured to generate an electronic document which associated the display sheet with the edit sheet.

Abstract: A method of identifying a printed page from a scan of the printed page is disclosed. The method comprises the steps of generating a page key of the printed page on the basis of the scan (710), searching a database (199) for a similar page key (730). For each found similar page key (740), the method further comprises; retrieving from the database an instance key location (750), generating an instance key for the printed page (530), based on the retrieved instance key location of the referenced page instance; and comparing the generated instance key for the printed page with the retrieved instance key of the referenced page instance (770). A match between the instance keys indicates that the printed page is the referenced page instance.

Abstract: A computer-implemented method for selecting at least one segmentation parameter for optical character recognition is provided. The method can include receiving an image having a character string that includes one or more characters. The method can also include receiving a character string identifying each of the one or more characters. The method can also include automatically generating at least one segmentation parameter. The method can also include performing segmentation on the image having the character string using the at least one segmentation parameter. The method can also include determining if a resultant segmentation satisfies one or more criteria and if the resultant segmentation satisfies the one or more criteria, selecting the at least one segmentation parameter.

Abstract: A computer-implemented method for selecting at least one segmentation parameter for optical character recognition is provided. The method can include receiving an image having a character string that includes one or more characters. The method can also include receiving a character string identifying each of the one or more characters. The method can also include automatically generating at least one segmentation parameter. The method can also include performing segmentation on the image having the character string using the at least one segmentation parameter. The method can also include determining if a resultant segmentation satisfies one or more criteria and if the resultant segmentation satisfies the one or more criteria, selecting the at least one segmentation parameter.

Abstract: Using methods, computer-readable storage media, and apparatuses for computer-implemented processing, a passage of text may be variably rendered. For each glyph in the passage of text, a glyph representation is varied according to a geometric transformation that was determined from statistical measurements of at least one geometric property from an ensemble of representations of the current glyph. Each varied glyph representation is included in renderable output data, such that when the passage of text is rendered to an output device, a given rendered representation of a given glyph subtly differs from other rendered representations of the given glyph.

Abstract: The invention is a method for omnidirectional recognition of recognizable characters in a captured two-dimensional image. An optical reader configured in accordance with the invention searches for pixel groupings in a starburst pattern, and subjects located pixel groupings to a preliminary edge crawling process which records the pixel position of the grouping's edge and records the count of edge pixels. If two similar-sized pixel groupings are located that are of sizes sufficient to potentially represent recognizable characters, then the reader launches “alignment rails” at pixel positions substantially parallel to a centerline connecting the center points of the two similarly sized groupings. A reader according to the invention searches for additional recognizable characters within the rail area, and subjects each located pixel grouping within the rail area to a shape-characterizing edge crawling process for developing data that characterizes the shape of a pixel grouping's edge.

Abstract: In a method for computer-aided transfer of data from a document application into a data application having a set of data fields, a document is displayed in the document application opened on a computer with a display device, and wherein from the document data are to be transferred into the data application also opened on the computer. A name of a data field into which data are to be transferred is displayed on the display device. Via identification of a corresponding data value in the document on the display device, a character string representing the data value is automatically read out from the document and entered into the data field corresponding to the data field name in the data application via actuation of a predetermined button.

Abstract: Systems, methods and computer program products on storage devices for shape clustering and applications in processing various documents, including an output of an optical character recognition (OCR) process. The output of an OCR process is classified into a plurality of clusters of clip images and a representative image for each cluster is generated to identify clusters whose clip images were incorrectly assigned character codes by the OCR process.

Abstract: In one embodiment, there is provided a method for an Optical Character Recognition (OCR) system. The method comprises: recognizing an input character based on a plurality of classifiers, wherein each classifier generates an output by comparing the input character with a plurality of trained patterns; grouping the plurality of classifiers based on a classifier grouping criterion; and combining the output of each of the plurality of classifiers based on the grouping.

Abstract: An image processing apparatus includes: a memory; an obtaining unit that obtains image data representing an image including concatenated pixels; an isolating unit that isolates a rendering element, the rendering element being an image surrounded by border lines of a color in an image represented by the image data; and a classifying unit that, in a case where a plurality of rendering elements has been isolated by the isolating unit, and in a case where color difference between two of the plural rendering elements or the distance between the two rendering elements is less than a threshold, classifies the two rendering elements into the same group, associates pieces of image data that represent rendering elements belonging to the same group with one another, and stores the pieces of image data in the memory.

Abstract: An apparatus for analyzing a video capture screen includes: a video frame extracting unit extracting at least one frame from a video having a plurality of frames; an extracted frame digitizing unit digitizing features of each of the at least one frame extracted by the video frame extracting unit; an image digitizing unit digitizing features of at least one collected search target image; an image comparing and searching unit comparing the search target image with the at least one frame extracted from the plurality of frames by digitized values of the collected search target image and the at least one frame; and a search result processing unit mapping related information of the collected search target image to a frame coinciding with the search target image and storing the related information in a database, when the extracted at least one frame coinciding with the search target image is present in a comparison result.

Abstract: Systems and methods for character recognition by performing lateral view-based analysis on the character data and generating a feature vector based on the lateral view-based analysis.

Abstract: Systems, apparatuses, and methods to relate images of words to a list of words are provided. A trellis based word decoder analyses a set of OCR characters and probabilities using a forward pass across a forward trellis and a reverse pass across a reverse trellis. Multiple paths may result, however, the most likely path from the trellises has the highest probability with valid links. A valid link is determined from the trellis by some dictionary word traversing the link. The most likely path is compared with a list of words to find the word closest to the most.

Abstract: A system and method to identify pictures in documents. An image representing a page of a document is received. The image is analyzed to identify text objects in the page. A masked image is generated by masking out regions of the image including the text objects in the page. Groups of pixels in the masked image are identified, wherein a respective group of pixels corresponds to at least one picture in the page. When there is one or more groups of pixels, regions for pictures are identified based on the one or more groups of pixels. Metadata tags for the pictures are stored, wherein a respective metadata tag for a respective picture includes information about a respective bounding box for the respective picture.

Abstract: The technology is directed to determining a class associated with an image. In some examples, a method determines the class associated with an image. The method can include determining a segmentation score for an image segment based on a comparison of the image segment and a region of an image. The region of the image can be associated with the image segment. The method further includes determining a confidence score for the image segment based on the segmentation score and a classification score. The classification score can be indicative of a similarity between the image segment and at least one class. The method further includes determining a class associated with the image based on the confidence score. The method further includes outputting the class associated with the image.

Abstract: Processing video for utilization in second language learning is described herein. A video file includes spoken words in a source language, subtitles in the source language, and subtitles in a native language of an end user (a target language). The subtitles in the source language are synchronized with the spoken words in the video, and the subtitles in the source language are mapped to the subtitles in the target language. Both sets of subtitles are displayed simultaneously as the video is played by the end user.

Abstract: In some embodiments, provided are procedures for processing images that may have different font sizes. In some embodiments, it involves OCR'ing with multiple passes at different resolutions.

Abstract: A digital image is converted to a multiple level image, and multiple scale sets are formed from connected components of the multiple level image such that different ones of the scale sets define different size spatial bins. For each of the multiple scale sets there is generated a count of connected components extracted from the respective scale set for each spatial bin; and adjacent spatial bins which represent connected components are linked. Then the connected components from the different scale sets are merged and text line detection is performed on the merged connected components. In one embodiment each of the scale sets is a histogram, and prior to linking all bins with less than a predetermined count are filtered out; and each histogram is extended such that counts of adjacent horizontal and vertical bins are added (single region bins are filtered out) and the linking is on the extended histograms.

Abstract: An image processing apparatus includes a line information reception unit, a line extraction unit, an inversion unit and a determination unit. The line information reception unit receives a set of information indicating (i) information on an image having a possibly of being a line and (ii) line elements being a rectangular pixel lump which constitutes a line. The line extraction unit extracts a line by tracing from a first start point to an end point of the line, based on the received information indicating the line elements and a tracing direction of the line. The inversion unit inverts the tracing direction of the line, sets the extracted end point of the line as a second start point and sends the second start point and the inverted tracing direction to the line extraction unit. The determination unit determines whether or not to cause the inversion unit to perform a process.

Abstract: A filter unit performs an analysis filtering operation by segmenting a frequency component to generate a plurality of subbands composed of coefficient data segmented on a per frequency band basis. A coefficient storage unit stores the coefficient data on a subband by subband basis, with each subband corresponding to a respective area of the coefficient storage unit. The filter unit writes the coefficient data onto the coefficient storage unit via a write buffer in response to a determination that the analysis filtering operation has not reached a final segmentation level. An entropy encoding unit entropy encodes the coefficient data. The filter unit writes the coefficient data on an area of the coefficient storage unit corresponding to the subband to which the coefficient data belongs. The entropy encoding unit reads the coefficient data, stored on different areas on a subband by subband basis, from the coefficient storage unit.

Abstract: The present invention relates to a method of authenticating and/or identifying an article containing a chemical marking agent, which is substantially inseparably enclosed in a marker as a carrier and contains selected chemical elements and/or compounds in the form of marker elements, in concentrations based on a predetermined encryption code, which method comprises the steps of: i) qualitatively and/or quantitatively identifying the marker elements of the chemical marking agent, and ii) comparing the values identified in step (i) with the predetermined encryption code.

Abstract: A system and a method for identification of alphanumeric characters present in a series in an image are disclosed. The system and method captures the image and further processes it for binarization by computing a pattern of the image. The generated binarized images are then filtered for removing unwanted components. Candidate images are identified out of the filtered binarized images. All the obtained candidate images are combined to generate a final candidate image which is further segmented in order to recognize a valid alphanumeric character present in the series.

Abstract: A set of ordered characters is received in association with information specifying the locations of the characters within the image of the document. Language-conditional character probabilities for each character are determined based on a set of language models and the ordering of the characters. Neighbor characters associated with a target character are identified based on the locations of the characters. Language-conditional character probabilities associated with the neighbor characters and language-conditional character probabilities associated with the target character are combined to generate a local language-conditional likelihood associated with the target character, the local language-conditional likelihood representing a concordance of the target character to a language model.

Abstract: A method and system for preprocessing an image, wherein the image includes a plurality of columns, or regions, of text is disclosed. A plurality of components associated with the text is determined. On determining the plurality of components, a line height and a column spacing is determined for the components. The components are then associated with a column based on the line height and the column spacing. A set of characteristic parameters are calculated for each column and the plurality of components of each column are merged based on the characteristic parameters to form sub-words and words. A first plurality of words and/or subwords is merged and processed as a first region and a second plurality of words and/or subwords is merged and processed as a second region wherein at least a portion of the second region vertically overlaps at least a portion of the first region.

Abstract: The invention is a method for omnidirectional recognition of recognizable characters in a captured two-dimensional image. An optical reader configured in accordance with the invention searches for pixel groupings in a starburst pattern, and subjects located pixel groupings to a preliminary edge crawling process which records the pixel position of the grouping's edge and records the count of edge pixels. If two similar-sized pixel groupings are located that are of sizes sufficient to potentially represent recognizable characters, then the reader launches “alignment rails” at pixel positions substantially parallel to a centerline connecting the center points of the two similarly sized groupings. A reader according to the invention searches for additional recognizable characters within the rail area, and subjects each located pixel grouping within the rail area to a shape-characterizing edge crawling process for developing data that characterizes the shape of a pixel grouping's edge.

Abstract: Using methods, computer-readable storage media, and apparatuses for computer-implemented processing, a passage of text may be variably rendered. For each glyph in the passage of text, a glyph representation is varied according to a geometric transformation that was determined from statistical measurements of at least one geometric property from an ensemble of representations of the current glyph. Each varied glyph representation is included in renderable output data, such that when the passage of text is rendered to an output device, a given rendered representation of a given glyph subtly differs from other rendered representations of the given glyph.

Abstract: Disclosed is a processor-implemented method for processing image data using an image processing apparatus. The processor is configured to receive a PDL file of image data and raster image process (RIP) the PDL file to determine pixels representing text. The ripped file is then segmented to determine at least any pixels representing text that were not initially indicated or identified by the ripped file. The results are combined to determine locations for marking onto a substrate by the output device using substantially colorless marking material over text pixels (to coat marked text pixels). In some instances, locations for covering text pixels with substantially colorless marking material can be tagged during segmenting image data, using a tag plane.

Abstract: Disclosed is a processor-implemented method for processing image data using an image processing apparatus. The processor is configured to receive a PDL file of image data and raster image process (RIP) the PDL file to determine at least pixels representing text of a predetermined colorant. The ripped file is then segmented to determine at least any text pixels of the predetermined colorant not initially indicated by the ripped file. The results are combined to determine text pixels of the predetermined colorant for marking onto a substrate using marking material (e.g., ink). In some instances, pixels of the predetermined colorant can be tagged during segmenting using a tag plane to determine text pixels for marking by the output device.