Abstract: The present invention discloses a multilevel method of bitmapped image analysis that comprises a whole image data representation via its components—objects of different levels of complexity—hierarchically connected therebetween by spatially-parametrical links. The said method comprises preliminarily generating a classifier of the objects that possibly may be present in the image consisting of one or more levels differing in complexity; parsing the image into objects; attaching each object to one of predetermined levels; establishing hierarchical links between objects of different levels; establishing links between objects within the same level; and performing an object feature analysis. The objects feature analysis comprises at least generating and examining a hypothesis about object features and correcting the object's features of the same and other levels in response to the hypothesis examination results.

Abstract: Embodiments of the present invention disclose a method and system for restoring a motion-blurred image. The method comprises determining parameters for a one-dimensional Optical Transfer Function (OTF) for the motion-blurred image in Fourier space; determining a signal-to-noise ratio for the motion-blurred image in the Fourier space; and correcting for motion blur based on the parameters of the OTF. Determining the parameters comprises calculating a function ?(p,q) which is based on the square of the modulus of the Fourier transform |G(p,q)|2 of the motion-blurred image. The parameters include the absolute value of the one-dimensional OTF, and the phase and sign of the OTF.

Abstract: In one embodiment, the invention provides a method for binarizing an image. The method comprises establishing boundaries of image objects of the image and classifying each image object as either suspect or non-suspect. The method further comprises creating a local binarization threshold map comprising threshold binarization values associated with image objects classified as non-suspect and then expanding the local binarization threshold map to cover the entire image thereby to create a global binarization threshold map for the entire image.

Abstract: In one embodiment, there is provided a computer-implemented method and system for implementing the method. The method comprises: preliminarily analyzing at least one corpus of natural language text comprising for each sentence of each natural language text of the corpus, performing syntactic analysis using linguistic descriptions to generate at least one syntactic structure for the sentence; building a semantic structure for the sentence; associating each generated syntactic and semantic structure with the sentence; and saving each generated syntactic and semantic structure; for each corpus of natural language text that was preliminarily analyzed, performing an indexing operation to index lexical meanings and values of linguistic parameters of each syntactic structure and each semantic structure associated with sentences in the corpus; and searching in at least one preliminarily analyzed corpora for sentences comprising searched values for the linguistic parameters.

Abstract: A method and computer system for analyzing sentences of various languages and constructing a language-independent semantic structure are provided. On the basis of comprehensive knowledge about languages and semantics, exhaustive linguistic descriptions are created, and lexical, morphological, syntactic, and semantic analyses for one or more sentences of a natural or artificial language are performed. A computer system is also provided to implement, analyze and store various linguistic structures and to perform lexical, morphological, syntactic, and semantic analyses. As result, a generalized data structure, such as a semantic structure, is generated and used to describe the meaning of one or more sentences in language-independent form, applicable to automated abstracting, machine translation, control systems, Internet information retrieval, etc.

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: In one embodiment, the invention provides a method for a machine to perform machine-readable form pre-recognition analysis. The method comprises preliminarily assigning at least one graphic image in a form for identification of form type, preliminarily creating at least one model of the said graphic image for identification of the form type, parsing a form image into regions, determining an image form type for the form image, comprising: (a) detecting on the form image at least one of said graphic images for identification of the form type, (b) performing a primary identification of the form image type based on a comparison of the detected graphic image with the said model, and(c) performing a profound analysis using a supplementary data said-primary identification results in multiple possibilities for the form image type.

Abstract: A device for obtaining graphical information from a single- or multi-page document printed on a hard media where reading out of the position of the document elements is performed by using a method of volumetric scanning of a document (even closed) is described. Processing of scanning results, comprises joining up the separate scanning layers scanning results, removing noise, correction of document image orientation, dividing information into portions relating to separate pages, is performed after reading the information. Then text information recognition contained in the graphical file is performed. Information may be read out by using methods of magnetic resonance scanning, supersonic scanning, X-ray scanning etc. The results of scanning in electronic form may be stored for further transmission thereof on a medium or via communication channels to a distant location for recognition.

Abstract: A method of obtaining graphical information from a single-or multi-page document printed on a hard media where reading out of the position of the document elements is performed by using a method of volumetric scanning of a document (even closed) is described. Processing of scanning results, comprises joining up the separate scanning layers scanning results, removing noise, correction of document image orientation, dividing information into portions relating to separate pages, is performed after reading the information. Then text information recognition contained in the graphical file is performed. Information may be read out by using methods of magnetic resonance scanning, supersonic scanning, X-ray scanning etc. The results of scanning in electronic form may be stored for further transmission thereof on a medium or via communication channels to a distant location for recognition. A device for realization of the described method is also disclosed.

Abstract: The present invention relates generally to an optical character recognition of machine-readable forms, and in particular to a verification of a direction of spatial orientation and a definition of a form type of the document electronic image. The goals of the invention are achieved by preliminarily assigning one or more form objects as elements composing a graphic image unambiguously defining its direction of spatial orientation. Similarly, one or more form objects are preliminarily assigned as elements composing a graphic image unambiguously defining its type. The direction of spatial orientation and the type of the form are verified via identification of said images. The models of graphic images either for verification the direction of spatial orientation or for defining the form type are stored in a special data storage means, one of the embodiment of which is form model description.

Abstract: In one embodiment, a method for identifying areas in a document image is provided. The method comprises generating binarized and gradient images based on the document image; and performing a classification operation to classify areas in the document image into one of a noise area and a picture area based on attributes computed on the binarized and gradient images.

Abstract: Embodiments of the invention disclose techniques for processing of machine-readable forms of unfixed or flexible format. An auxiliary brief description may be optionally specified to determine the spatial orientation of the image. A method of searching for elements of a document comprises the following main operations in addition to the operations of preliminary image processing: selecting the varieties of structural description from several available variants, determining the orientation of the image, selecting the text objects, where the text must be recognized, and determining the minimal required volume of recognition, recognizing the text objects, searching for elements of the form. Searching for elements of the form comprises the following actions: selecting a searched element in the structural description, gaining the algorithm of search constraints from the structural description, searching for the element, testing the obtained variants.

Abstract: Embodiments disclose a technique to recognize text in a current frame of an image in a view finder of a digital camera. In accordance with the technique, text at a marker (e.g. a cursor or cross hairs) associated with the view finder is recognized and a lookup is performed based on the recognized text. Advantageously, the lookup yields useful information e.g. a translation of a recognized word that is displayed in the viewfinder adjacent to the text. The current frame is not captured by a user. As the user moves the camera to position a new word at the marker, the view finder is updated to provide lookup results associated with the new word. Lookups may be performed of a bilingual dictionary, a monolingual dictionary, a reference book, a travel guide, etc. Embodiments of the invention also cover digital cameras or mobile devices that implement the aforementioned technique.

Abstract: A method of obtaining graphical information from a single- or multi-page document printed on a hard media where reading out of the position of the document elements is performed by using a method of volumetric scanning of a document (even closed) is described. Processing of scanning results, comprises joining up the separate scanning layers scanning results, removing noise, correction of document image orientation, dividing information into portions relating to separate pages, is performed after reading the information. Then text information recognition contained in the graphical file is performed. Information may be read out by using methods of magnetic resonance scanning, supersonic scanning, X-ray scanning etc. The results of scanning in electronic form may be stored for further transmission thereof on a medium or via communication channels to a distant location for recognition.

Abstract: A method is claimed for processing a vector-raster image file which contains a text image. The method comprises the steps of: fragmenting the image to obtain regions containing non-separable, logically connected fragments of text of the maximum possible size; processing text, vector, and raster objects; discarding excessive information; analyzing each object with the help of all available information. The step of processing text objects includes the steps of: dividing into separate characters and character groups according to supposed locations of blank spaces or other non-indicated symbols, and analyzing and assembling character groups into words and verifying and correcting characters encoding based on recognition of assembled words as raster objects. The step of processing vector objects includes the step of identifying separators, background, and substrates of blocks.

Abstract: The present invention discloses a method of character and text recognition of a bit-mapped graphic file received from an optical scanning device. The method comprises a trainable template cache, a preliminarily trained feature analysis means, and a context analysis means. The present invention discloses the way to use said means for achieving the best results in recognition. The method supposes that the template cache along with the context analysis means are used as the main shape characteristic analyzing means. The feature analysis means along with the context analysis means are used as subsidiary shape characteristic analyzing means and as a training means for the template cache. The method comprises applying the main shape characteristic analyzing means and optionally applying the subsidiary shape characteristic analyzing means if no or not enough reliability of recognition is achieved after the template cache analyzing.

Abstract: The present invention deals with text comprising image parsed to graphemes. A result of character recognition is creation of one or more versions of characters for each grapheme. All possible words versions are obtained using all characters versions, and all parsing versions are examined. A supplementary data of several types is applied successively in the preliminarily prescribed order to the examined words. The processing with the use of supplemental data may be represented as a three times repeated processing of the same text fragment with the use of supplementary information becoming available at each time. The examination comprises three steps. 1) A set of chains LPG is built using all obtained recognized grapheme-to-character versions. 2) All obtained versions are analyzed with the successive application of subsequent supplemental data types in connection with the preliminarily assigned order or with a joint application thereof. 3) A supplementary space recognition correction.

Abstract: In one embodiment, a method for correcting distortions in a scanned image of a page 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: An adjustment method of a machine-readable form model and a filled form scanned image thereof in the presence of distortion comprising model form free of distortion, image form, containing distortion, obtained by optical input device from paper media. One embodiment of the method comprises the steps of assigning of one of the forms as changeable, marking on the form regions containing distortion, computing the consolidated distortion correction factor of the spatial parameters for the said changeable form, parameters correction on the base of the said consolidated factor. Another embodiment of the method setting the correspondence between identical objects of image form and model form is followed by steps of computing distortion correction factors of the spatial parameters for each object of selected level, spatial parameters correction of the changeable form, on the base of the said factors, affecting objects of the same and lower identification reliability levels.

Abstract: A method is described of bit-mapped image analysis comprising division of all analysis means at one's disposal into several groups differing in accuracy and further processing multi-stage analysis. The analysis comprises a primary analysis stage and at least one profound analysis stage, with supplemental data collected at both stages. The primary analysis, includes preliminary recognition of objects with distortion and detection of objects that require more precise analysis means to overcome the distortion. At the primary analysis stage, the analysis means from the group of the most inaccurate group are used. The profound analysis stage includes repeating recognition of objects with a distortion taking into account the supplemental data obtained at the previous stage, detecting objects that require more precise analysis means to overcome the distortion, and collection of newly appeared supplemental data. Each subsequent profound analysis stage uses analysis means from the group of more accurate means.