Showing papers on "Document layout analysis published in 1997"

System for searching a corpus of document images by user specified document layout components

Abstract: A document search system provides a user with a programming interface for dynamically specifying features of documents recorded in a corpus of documents. The programming interface operates at a high-level that is suitable for interactive user specification of layout components and structures of documents. In operation, a bitmap image of a document is analyzed by the document search system to identify layout objects such as text blocks or graphics. Subsequently, the document search system computes a set of attributes for each of the identified layout objects. The set of attributes which are identified are used to describe the layout structure of a page image of a document in terms of the spatial relations that layout objects have to frames of reference that are defined by other layout objects. After computing attributes for each layout object, a user can operate the programming interface to define unique document features. Each document feature is a routine defined by a sequence of selections operations which consume a first set of layout objects and produce a second set of layout objects. The second set of layout objects constitutes the feature in a page image of a document. Using the programming interface, a user flexibly defines a genre of document using the user-specified document features.

Adaptive document binarization

Abstract: A new method is presented for adaptive document image binarization, where the page is considered as a collection of subcomponents such as text, background and picture. The problems caused by noise, illumination and many source type related degradations are addressed. The algorithm uses document characteristics to determine (surface) attributes, often used in document segmentation. Using characteristic analysis, two new algorithms are applied to determine a local threshold for each pixel. An algorithm based on soft decision control is used for thresholding the background and picture regions. An approach utilizing local mean and variance of gray values is applied to textual regions. Tests were performed with images including different types of document components and degradations. The results show that the method adapts and performs well in each case.

Method and apparatus for image based document processing

Abstract: The present invention provides a document processing apparatus, document processing method and a storage medium for storing thereof on purpose to offer document filing in which document can be registered with a little computation cost and with high speed, and retrieval can be performed with little oversight. In the document processing apparatus, a similar character classifying element classifies characters in a document image into similar character categories in advance and stores the classified categories together with their representative image features. When the document image is registered, a pseudo character recognizing element executes, without identifying each character in the text region, classification into character categories based on the image features less than those used in the ordinary character recognition and stores the category strings generated by identifying each character with the inputted image. In retrieval, a retrieval executing element converts each character in the retrieval keyword into nearest category, and retrieves a document including the converted category string as a part as a result of retrieval.

A fast algorithm for bottom-up document layout analysis

Abstract: This paper describes a new bottom-up method for document layout analysis. The algorithm was implemented in the CLIDE (Chemical Literature Data Extraction) system, but the method described here is suitable for a broader range of documents. It is based on Kruskal's algorithm and uses a special distance-metric between the components to construct the physical page structure. The method has all the major advantages of bottom-up systems: independence from different text spacing and independence from different block alignments. The algorithms computational complexity is reduced to linear by using heuristics and path-compression.

Extraction of key sections from texts using automatic indexing techniques

Abstract: A document condensation method and apparatus produce a document synopsis are provided in which automatic indexing techniques are used to analyze an input document to determine a list of words and phrases characteristic of the subject matter of the document. Sections of the document are compared to the list of characteristic words and phrases to determine which sections of the document are most like the overall document in view of subject matter. A predetermined number of sections determined to be most similar to the overall document in content are provided as a condensed version of the whole document.

System for sorting document images by shape comparisons among corresponding layout components

Abstract: A programming interface of document search system enables a user to dynamically specifying features of documents recorded in a corpus of documents. The programming interface provides category and format flexibility for defining different genre of documents. The document search system initially segments document images into one or more layout objects. Each layout object identifies a structural element in a document such as text blocks, graphics, or halftones. Subsequently, the document search system computes a set of attributes for each of the identified layout objects. The set of attributes are used to describe the layout structure of a page image of a document in terms of the spatial relations that layout objects have to frames of reference that are defined by other layout objects. Using the set of attributes a user defines features of a document with the programming interface. After receiving a feature or attribute and a set of document images selected by a user, the system forms a set of image segments by identifying those layout objects in the set of document images that make up the selected feature or attribute. The system then sorts the set of image segments into meaningful groupings of objects which have similarities and/or recurring patterns. In operation, the system sorts images in the image domain based on segments (or portions) of a document image which have been automatically extracted by the system. As a result, searching becomes more efficient because it is performed on limited portions of a document. Subsequently, document images in the set of document images are order and displayed to a user in accordance with the meaningful groupings.

Sorting image segments into clusters based on a distance measurement

Abstract: A programming interface of document search system enables a user to dynamically specifying features of documents recorded in a corpus of documents. The programming interface provides category and format flexibility for defining different genre of documents. The document search system initially segments document images into one or more layout objects. Each layout object identifies a structural element in a document such as text blocks, graphics, or halftones. Subsequently, the document search system computes a set of attributes for each of the identified layout objects. The set of attributes are used to describe the layout structure of a page image of a document in terms of the spatial relations that layout objects have to frames of reference that are defined by other layout objects. Using the set of attributes a user defines features of a document with the programming interface. After receiving a feature or attribute and a set of document images selected by a user, the system forms a set of image segments by identifying those layout objects in the set of document images that make up the selected feature or attribute. The system then sorts the set of image segments into meaningful groupings of objects which have similarities and/or recurring patterns. Subsequently, document images in the set of document images are ordered and displayed to a user in accordance with the meaningful groupings.

Apparatus and method for extracting management information from image

Abstract: A management information extraction apparatus learns the structure of the ruled lines of a document and the position of user-specified management information such as a title, etc. during a form learning process, and stores them in a layout dictionary. During the operation, the structure of the ruled lines extracted from an image of an input document is matched with that of the document in the layout dictionary. Then, position information in the layout dictionary is referred to, and the management information is extracted from the input document.

Desktop publishing software for automatically changing the layout of content-filled documents

Abstract: A publisher program for automatically changing the layout of content-filled desktop publishing documents. The publisher program allows the author of a document to use a menu-driven utility to define a first layout for the document. A program module known as a Wizard then renders a document in the first layout. The author makes changes to the document while the document is in the first layout, typically by adding content and making author-defined changes to the document rendered by the Wizard. The author may then return to the menu-driven utility to select a second layout for the document. The Wizard renders a document in the second layout and a program module known as a Page Manager automatically applies the author's changes to the document and renders a content-filled document in the second layout.

Transmission of subsets of layout objects at different resolutions

Abstract: In a document search and retrieval system, document images are segmented into layout objects. Each layout object identifies different structural elements in a document image. In addition, the system computes attributes and features for each segmented layout object. Before any document images are transmitted between a client and a server, users specify which document image attributes and features are most relevant to their browsing or searching tasks. Transmission (and/or display) of document images is then divided into two stages. During the first stage, those layout objects which are identified as having the specified features or attributes are transmitted at a first or high resolution; the remaining layout objects in an image are transmitted at a second or lower resolution (or in the form of bounding polygons). If the second stage is invoked, those remaining layout objects are re-transmitted at the first or high resolution. The second stage of transmission may be invoked when either a user request is received or when there is a system timeout.

Logical structure analysis of book document images using contents information

Abstract: Numerous studies have so far been carried out extensively for the analysis of document image structure, with particular emphasis placed on media conversion and layout analysis. For the conversion of a collection of books in a library into the form of hypertext documents, a logical structure extraction technology is indispensable, in addition to document layout analysis. The table of contents of a book generally involves very concise and faithful information to represent the logical structure of the entire book. That is to say, we can efficiently analyze the logical structure of a book by making full use of its contents pages. This paper proposes a new approach for document logical structure analysis to convert document images and contents information into an electronic document. First, the contents pages of a book are analyzed to acquire the overall document logical structure. Thereafter, we are able to use this information to acquire the logical structure of all the pages of the book by analyzing consecutive pages of a portion of the book. Test results demonstrate very high discrimination rates: up to 97.6% for the headline structure, 99.4% for the text structure, 97.8% for the page-number structure and almost 100% for the head-foot structure.

Performance evaluation of document layout analysis algorithms on the UW data set

Abstract: A performance evaluation protocol for the layout analysis is discussed in this paper. In the University of Washington English Document Image Database-III, there are 1600 English document images that come with manually edited ground truth of entity bounding boxes. These bounding boxes enclose text and non-text zones, text-lines, and words. We describe a performance metric for the comparison of the detected entities and the ground truth in terms of their bounding boxes. The Document Attribute Format Specification is used as the standard data representation. The protocol is intended to serve as a model for using the UW-III database to evaluate the document analysis algorithms. A set of layout analysis algorithms which detect different entities have been tested based on the data set and the performance metric. The evaluation results are presented in this paper.

Extraction of indicative summary sentences from imaged documents

Abstract: A system for selecting sentences from an imaged document for presentation as part of a document summary is presented. The extracts are identified without the use of optical character recognition. The sentences are selected based on a set of discrete features characterizing the words within a sentence and the location of the sentence within the imaged document. Each sentence is scored based on the values of the discrete features using a statistically based classifier. The imaged document is processed to identify the word locations, the reading order of words, and the location of sentence and paragraph boundaries in the text. The words are grouped into equivalence classes to mimic the terms in a text document. A sample extract for a technical document is shown, and evaluation against a set of abstracts created by a professional abstracting company is given. These results are compared with text-based abstracts.

Device and method for document processing and storage medium storing document processing program

Abstract: PROBLEM TO BE SOLVED: To perform the document processing in response to the user's preference by shifting the document vector acquired via a document vector acquisition means to a preference vector. SOLUTION: A document (object document) which is the object of acquiring a preference document vector is acquired and stored in a RAM 113. A CPU 111 extracts the processing important phrases of a matrix from the object document and then decides the importance based on the frequency of appearance, the evaluation function, etc., included in the object document that is contained in a processing important document. Then a reference document vector is shifted and a preference document vector including the user's preference is obtained based on the frequency of appearance of the processing important phases included in the user's processing document. The similarity of another document is calculated to the preference document vector as an index to the preference such as the interest, the degree of attention, the purpose, etc., of the user. Thereby, it is possible to sort and retrieve the documents in response to the user's preference by sorting and retrieving the documents based on the similarity.

The function of documents

Abstract: The purpose of a document is to facilitate the transfer of information from its author to its readers. It is the author's job to design the document so that the information it contains can be interpreted accurately and efficiently. To do this, the author can make use of a set of stylistic tools. In this paper, we introduce the concept of document functionality, which attempts to describe the roles of documents and their components in the process of transferring information. A functional description of a document provides insight into the type of the document, into its intended uses, and into strategies for automatic document interpretation and retrieval. To demonstrate these ideas, we define a taxonomy of functional document components and show how functional descriptions can be used to reverse-engineer the intentions of the author, to navigate in document space, and to provide important contextual information to aid in interpretation.

Layout for Visualizing Large Software Structures in 3D

Abstract: There is increasing evidence that 3D visualization of complex structures has advantages over 2D visualization While nested directed graphs are an important method of representing information in 2D or 3D, they must be effectively organized in order to be understood Most work on graph layout has assumed that fully automatic layout is desirable Through our work with graphs representing large software structures, we have found that, due to the importance of the semantic content, it is necessary to combine automatic layout with manual layout This paper describes a system called GraphVisualizer3D, which was designed to help people understand large nested graph structures by displaying them in 3D This system is currently being applied to the problem of understanding large bodies of software In this paper we present the layout strategy developed for GraphVisualizer3D, which combines manual layout techniques and automatic algorithms in a synergistic manner In order to facilitate manual layout, a grid system is provided that can be nested to any arbitrary depth The automatic layout is accomplished by layering followed by a node migration algorithm, whereby nodes migrate to their final position under the influence of a variety of different forces Options are provided to allow users to switch back and forth between manual layout and automatic layout GV3D has been tested with large examples containing more than 35,000 nodes and 40,000 relationships

Use of document structure analysis to retrieve information from documents in digital libraries

Abstract: This paper describes an approach to retrieving information from document images stored in a digital library by means of knowledge-based layout analysis and logical structure derivation techniques. Queries on document image content are categorized in terms of the type of information that is desired, and are parsed to determine the type of document from which information is desired, the syntactic level of the information desired, and the level of analysis required to extract the information. Using these clauses in the query, a set of salient documents are retrieved, layout analysis and logical structure derivation are performed on the retrieved documents, and the documents are then analyzed in detail to extract the relevant logical components. A 'document browser' application, being developed based on this approach, allows a user to interactively specify queries on the documents in the digital library using a graphical user interface, provides feedback about the candidate documents at each stage of the retrieval process, and allows refinements of the query based on the intermediate results of the search. Results of a query are displayed either as an image or as formatted text.

Automatic acquisition of layout knowledge for understanding business cards

Abstract: Document knowledge plays a very important role in many document image understanding methods. In these methods, the document knowledge is utilized to classify/extract individual item data interpretatively from paper-based sheets as a kind of document model: these days, this knowledge is specified into the document image understanding system in advance. In this paper, we propose an experimental method to acquire the layout knowledge automatically from sample document images. In particular, we focus on the acquisition of the layout of business cards. Our idea is to generate the layout knowledge of business cards from a predefined logical structure, which is used as a kind of meta-knowledge to interpretatively generate the layout knowledge of given business cards.

Digital compound machine

Abstract: PROBLEM TO BE SOLVED: To obtain a digital compound machine which can store and take document image data out by accurately specifying a document image file. SOLUTION: This digital compound machine which can specify document image data for a specific purpose by using a document ID or the document ID mark corresponding to the document ID is equipped with a property display means which displays a table on a property storage means 6 where property information of the document image data corresponding to the document ID is registered and the contents registered in the property table when the document image data is specified. Not is only a cover document with a document ID mark specified for a document image file, but property information can be confirmed with the property table at need, so the document image file can accurately be specified. COPYRIGHT: (C)1999,JPO

Page segmentation using document model

Abstract: Transforming a paper document to its electronic version in a form suitable for efficient storage, retrieval and interpretation continues to be a challenging problem. An efficient document model is necessary to solve this problem. Document modeling involves techniques of thresholding, skew detection, geometric layout analysis and logical layout analysis. The derived model can then be used in document storage and retrieval. We use the traditional bottom-up approach based on the connected component extraction to efficiently implement page segmentation and region identification. A new document model which preserves top-down generation information is proposed based on which a document is logically represented for interactive editing, storage, retrieval, transfer and logical analysis.

Supporting information extraction from printed documents by Lexico-Semantic pattern matching

Abstract: Document analysis and understanding (DAU) systems aim not only at the recognition of text and document structures but also at the extraction of relevant information out of a scanned document. Depending on the class of a document, information to be extracted may be defined in advance in syntactic structures as well as in semantic structures. In this paper we present a system for detecting such information and transforming it into a semantic representation. The basic component is a pattern matcher which incorporates geometric positions to detect phrases in the document. By defining a Levenshtein distance, the component reacts more generously in order to be error tolerant against OCR failures.

Document layout analysis based on emergent computation

Abstract: A new method of document layout analysis is proposed for a document reader, to be used for reading a wide variety of documents. Emergent computation, which is a key concept of artificial life, is adopted to analyze various complex document structures. The proposed method uses a multilayer architecture consisting of four subsystems: region extraction, region analysis, region recognition, and region modification. Emergent computation is used for the interactions between subsystems to produce effective and flexible behavior of the entire system. The global layout structure of a document is extracted from these interactions. Experimental results obtained for 150 documents show the method is adaptable to various layout structures in documents.

Local skew angle estimation from background space in text regions

Abstract: Almost all document analysis approaches need to perform a global analysis of the page orientation as a separate process at an early stage. It would be preferable to estimate the orientation locally after page segmentation and classification, when more knowledge about the different regions is available. A novel local skew estimation method is presented that takes advantage of the information available after flexible and efficient page segmentation and classification methods have been applied to the document image. The proposed method accurately estimates the orientation of individual text regions by efficiently analysing the arrangement of background space contained in them. No assumption is made about the existence of a uniform or dominant orientation in the document. The whole process is very efficient, as only the regions of text are considered and the points used for the angle estimation are already available as by products of previous document analysis stages.

Automated document content characterization for a multimedia document retrieval system

Abstract: We propose a new approach to automate document image layout extraction for an object-oriented database feature populationusing rapid low level feature analysis, preclassification and predictive coding. The layout information comprised of region lo-cation and classification data is transformed into 'feature object(s)' . The information is then fed into an intelligent documentimage retrieval system (IDIR) to be utilized in document retrieval schemes. The IDIR system consists of user interface, object-oriented database and a variety of document image analysis algorithms. In this paper the object-oriented storage model and thedatabase system are presented in formal and functional domains. Moreover, the graphical user interface and a visual documentimage browser are described. The document analysis techniques used at document characterization are also presented. In thiscontext the documents consist of text, picture and other media (possibly embedded) data. Documents are stored in the databaseas document, page and region objects. Our test system has been implemented and tested using a document database of 10 000documents.Keywords: Document layout analysis, predictive coding, document database, retrieval, document content characterization, ob-ject-oriented database.