Xerox Corporation

Other

Ashok Chhabedia Popat

Daniel H Greene

Dan S Bloomberg

Tze-Lei Poo

Justin Romberg

Thomas P Minka

Francine R Chen

Eric Saund

Daniel Lynn Larner

Thomas P Moran

James V Mahoney

Eric Gaussier

William C Janssen

Thomas M Breuel

David J Fleet

Henry S Baird

Daniel S Bloomberg

The invention converts a document originating in a page-image format into a form suitable for an arbitrarily sized display, by reformatting or 're-flowing' of the document to fit an arbitrarily sized display device. A two-stage system analyzes, or 'deconstructs,' page image layout. The deconstruction includes both physical (geometric) and logical (functional) segmentation of page images. The segment that image elements may include blocks, lines, and/or words of text, and other segmented image elements. The segment that image elements are synthesized and converted into an intermediate structure. The intermediate data structure is then distilled or converted or redisplayed into any number of standard print formats.

Method and system for document image layout deconstruction and redisplay

Methods, systems, and articles of manufacture consistent with certain principles related to the present invention enable a computing system to perform hierarchical topical clustering of text data based on statistical modeling of co-occurrences of (document, word) pairs. The computing system may be configured to receive a collection of documents, each document including a plurality of words, and perform a modified deterministic annealing Expectation-Maximization (EM) process on the collection to produce a softly assigned hierarchy of nodes. The process may involve assigning documents and document fragments to multiple nodes in the hierarchy based on words included in the documents, such that a document may be assigned to any ancestor node included in the hierarchy, thus eliminating the hard assignment of documents in the hierarchy.

Methods, systems, and articles of manufacture for soft hierarchical clustering of co-occurring objects

Methods and systems for document image decoding incorporating a Stack algorithm improve document image decoding. The application of the Stack algorithm is iterated to improved decoding. A provisional weight is determined for a partial path to reduce template matching. In addition, semantically equivalent hypotheses are identified to reduce redundant hypotheses.

Document image decoding systems and methods using modified stack algorithm

Methods and systems for document image decoding incorporating a Stack algorithm improve document image decoding. The application of the Stack algorithm is iterated to improve decoding. A provisional weight is determined for a partial path to reduce template matching. In addition, semantically equivalent hypotheses are identified to reduce redundant hypotheses.

A system encodes a separate assist channel that carries only a small amount of additional information in a hardcopy document to compensate for failure of an OCR system to accurately reconstruct a scanned electronic version of the hardcopy document. After the OCR system produces an interpretation of the scanned electronic version of the hardcopy document, a rewrite system that is guided by the assist channel operates independent from the OCR system to correct decoding errors produced by the OCR system.

Assist channel coding using a rewrite model

A graphical input and display system for creating and manipulating electronic images includes input devices permitting a user to manipulate elements of electronic images received from various image input sources. A processor, connected to the system, receives requests for various image editing operations and also accesses a memory structure. The system memory structure includes a user interaction module, which allows a user to enter new image material or select and modify existing image material to form primary image objects, as well as a grouping module, which maintains an unrestricted grouping structure, an output module, and data memory.

System and method for editing electronic images

In a text recognition system that uses a stochastic finite state network to model a document image layout, the computational efficiency of text line decoding is improved. In a typical implementation, the dynamic programming operation that accomplishes decoding uses actual scores computed between two-dimensional (2D) bitmapped character template images and the (2D) bitmapped observed image. Scoring measures the degree of a match between a character template and the observed image. Computation of these actual scores is replaced with the simpler computation of column-based (i.e., one-dimensional) heuristic scores. Because the column-based heuristic scores can be shown to be a true upper bound on actual template-image scores, the heuristic scores are accurate enough to use in place of actual scoring during text line decoding. The heuristic scores essentially reduce the expensive two-dimensional computation of the actual template-image scores required by prior decoding methods to a simpler but accurate one-dimensional computation.

Document image decoding using text line column-based heuristic scoring

A text recognition system represents the decoded message of a document image as a path through an image network. A method for integrating a language model into the network selectively expands the network to accommodate the language model only for certain ones of the paths in the network, effectively managing the memory storage requirements and computational complexities of integrating the language model efficiently into the network. The language model generates probability distributions indicating the probability of a certain character occurring in a string, given one or more previous characters in the string. Selectively expanding the image network is achieved by initially using upper bounds on the language model probabilities on the branches of an unexpanded image network. A best path search operation is then performed to determine an estimated best path through the image network using these upper bound scores. After decoding, only the nodes on the estimated best path are expanded with new nodes and with branches incoming to the new nodes that accommodate new language model scores reflecting actual character histories in place of the upper bound scores. Decoding and selectively expanding the image network are repeated until the final output transcription of the text image has been produced.

Document image decoding using an integrated stochastic language model

A system encodes a separate assist channel that carries only a small amount of additional information in a hardcopy document to compensate for failure of an OCR system to accurately reconstruct a scanned electronic version of the hardcopy document. The assist channel protects symbols printed on the hardcopy document by computing guard values that are derived from a hash of the character code of each character on the hardcopy document. Template match searches that are performed during decoding are limited to those interpretations that produce matching guard values.

Assist channel coding with convolution coding

In a text recognition system, the computational efficiency of a text line image decoding operation is improved by utilizing the characteristic of a graph known as the cut set. The branches of the data structure that represents the image are initially labeled with estimated scores. When estimated scores are used, the decoding operation must perform iteratively on a text line before producing the best path through the data structure. After each iteration, nodes in the best path are re-scored with actual scores. The decoding operation incorporates an operating mode called skip mode. When the number of consecutive image positions for which the change value of cumulative path scores between current and prior iterations is substantially constant and exceeds a threshold, this signals the presence of a cut set, and the score change value is added to a previously computed path score until a re-scored node is encountered, thereby eliminating the expensive computation of new cumulative path scores at those image positions.

Growing community of inventors

San Carlos, CA, United States of America

Ashok Chhabedia Popat