Abbyy Development Inc.

Abbyy Software Limited

Abbyy Production LLC

Dmitry Deryagin

Konstantin Anisimovich

Yuri Chulinin

Mikhail Atroshchenko

Konstantin Zuev

Vladimir Rybkin

Andrey Anatolievich Isaev

Konstantin Statsenko

Atroshchenko Mikhail Yurievich

Systems and methods are disclosed to receive an image depicting at least a part of a document and identify a plurality of partition points dividing the image into potential segments; generate a linear partition graph (LPG) comprising a plurality of vertices using the plurality of partition points and a plurality of arcs connecting the plurality of vertices; identify a path of the LPG having a value of a quality metric above a threshold value, wherein the path is selected from a plurality of paths of the LPG and comprises one or more arcs and the value of the quality metric is derived using a neural network classifying each of a plurality of pixels of the image; and generate one or more blocks of the image wherein each of the one or more blocks corresponds to an arc of the identified path and represents a portion of the image associated with a type of an object.

Methods and systems of segmentation of a document

Disclosed are systems, computer-readable mediums, and methods for determining that text contains Chinese, Japanese, or Korean characters. One method includes determining a language hypothesis for each text fragment in a plurality of text fragments identified from connected components in a document image. The method further includes selecting a first subset of text fragments from the plurality of text fragments based on ratings for the language hypothesis of each text fragment in the plurality of text fragments. The method further includes verifying, by a processor, the language hypothesis of one or more text fragments in the first subset of text fragments based on optical character recognition of the one or more text fragments. The method further includes determining, by the processor, that Chinese, Japanese, or Korean (CJK) characters are present in the document image based on the verification of the language hypothesis of each of the one or more text fragments.

Chinese, Japanese, or Korean language detection

Systems and methods for identifying screenshots within document images. An example method comprises: receiving an image of at least a part of a document; identifying, within the image, a polygonal object having a visually distinct border comprising a plurality of edges of one or more intersecting rectangles; asserting a screenshot image hypothesis with respect to the identified polygonal object; and responsive to evaluating at least one condition associated with one or more attributes of the identified polygonal object, classifying the identified polygonal object as a screenshot image.

Identifying screenshots within document images

Disclosed are systems, computer-readable mediums, and methods for determining a text contains Chinese, Japanese, or Korean characters. A document image is received and binarized. The binarized document image is searched for connected components. A plurality of fragments is identified based on the connected components. A language hypothesis for each fragment of the plurality of fragments is determined. The language hypothesis has a probability rating. A subset of fragments from the plurality of fragments having the highest probability ratings is selected. The language hypothesis of each fragment in the subset of fragments is verified. A determination of the presence of Chinese, Japanese, or Korean characters is made based at least on the verification of the language hypothesis of the subset of fragments.

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.

Identifying picture areas based on gradient image analysis

The invention relates to methods for determining a logical structure of a document. The system stores a collection of models, each of which describes one or more possible logical structures. At least one document hypothesis is generated for the whole document. For each document hypothesis, the system verifies the document hypothesis on each page, for example, by generating at least one block hypothesis for each block in the document based on the document hypothesis, selecting a best block hypothesis for each block, selecting the model that corresponds to a best document hypothesis the document hypothesis that has a best degree of correspondence with the selected best block hypotheses for the document, and forming a representation of the document based on the best document hypothesis described.

Model-based methods of document logical structure recognition in OCR systems

Disclosed is a method of bit-mapped image analysis that comprises a whole image data representation via its component objects. The objects are assigned to different levels of complexity. The objects may be hierarchically connected by spatially-parametrical links. The method comprises preliminarily generating a classifier of image objects consisting of one or more levels differing in complexity; parsing the image into objects; attaching each object to one or more predetermined levels; establishing hierarchical links between objects of different levels; establishing links between objects within the same level; and performing an object feature analysis. Object feature analysis comprises generating and examining a hypothesis about object features and correcting the concerned object's features of the same and other levels in response to results of hypothesis examination. Object feature analysis may also comprise execution of a recursive X-Y cut within the same level.

Multilevel image analysis

Methods and system for processing document images in OCR systems, particularly for selecting a proper file name for a recognized document. The method comprises generating at least one document type hypothesis for the document; verifying each document type hypothesis; selecting a best document type hypothesis and saving the document with a proper name based on the best type hypothesis and unique features. The method further includes determining a logical structure of a document and selecting a best document model hypothesis that has the best degree of correspondence with the selected best block hypotheses for the document. On the basis of the best document model hypothesis the text document reflecting the logical structure of the source document in extended computer-editable format is formed and saved with a proper file name.

Automatic file name generation in OCR systems

In one embodiment, the invention provides a method for determining a logical structure of a document. The method comprises generating at least one document hypothesis for the whole document; for each document hypothesis, verifying said document hypothesis including (a) generating at least one block hypothesis for each block in the document based on the document hypothesis; and (b) selecting a best block hypothesis for each block; selecting as a best document hypothesis the document hypothesis that has the best degree of correspondence with the selected best block hypotheses for the document; and forming the document based on the best document hypothesis.

Growing community of inventors

Moscow, Russia

Dmitry Deryagin