Yasuto Ishitani

Document skew detection based on local region complexity

A new method is proposed for detecting skew in document images which contain a mixture of text areas, photographs, figures, charts, and tables. Two basic ideas are introduced in the method. One idea is that a new parameter is used for skew detection to discern the orientation of text lines in document images. This parameter is based on the document image complexity and is obtained from the number of transitions from white to black pixels or vice versa. The other idea is that skew is detected in local regions in which only text lines are expected. Such local regions are extracted from a document image automatically and the obtained skew angle is defined as the overall document skew. Document skew has been measured in experiments with an error of 0.12 degrees on the average for all test documents.<<ETX>>

Document transformation system from papers to XML data based on pivot XML document method

This paper proposes a new method for document transformation using OCR to generate various XML documents from printed documents. The proposed method adopts a hierarchical transformation strategy based on a pivot XML document. Firstly, document elements such as title, authors, abstract, headings, paragraphs, lists, captions, tables and figures are extracted from document images. Secondly, the hierarchical structure of document elements is extracted and is described using a DOM tree. Thirdly, this document structure is converted into a pivot XML document described as an XHTML document by an XML parser. Finally, this pivot XML document is transformed into the target XML document by the XML parser with XSLT scripts or specific programs. Experimental results show the method is effective in transforming printed documents to various XML documents.

Logical structure analysis of document images based on emergent computation

A new method for logical structure analysis of document images is proposed in this paper as the basis for a document reader which can extract logical information from various printed documents. The proposed system consists of five basic modules: typography analysis, object recognition, object segmentation, object grouping and object modification. Emergent computation, which is a key concept of artificial life, is adopted for the cooperative interaction among the modules in the system in order to achieve an effective and flexible behavior of the whole system. It has two principal advantages over other methods: adaptive system configuration for various and complex logical structures, and robust document analysis that is tolerant of erroneous feature detection.

Model-based information extraction method tolerant of OCR errors for document images

A new method for information extraction from document images is proposed in this paper as the basis for a document reader which can extract required keywords and their logical relationship from various printed documents. Such documents obtained from OCR results may have not only unknown words and compound words, but also incorrect words due to OCR errors. To cope with OCR errors, the proposed method adopts robust keyword matching which searches for a string pattern from two dimensional OCR results consisting of a set of possible character candidates. This keyword matching uses a keyword dictionary that includes incorrect words with typical OCR errors and segments of words to deal with the above difficulties. After keyword matching, a global document matching is carried out between keyword matching results in an input document and document models which consist of keyword models and their logical relationship. This global matching determines the most suitable model for the input document and solves word segmentation problems accurately even if the document has unknown words, compound words, or incorrect words. Experimental results obtained for 100 documents show that the method is robust and effective for various document structures.

Model matching based on association graph for form image understanding

A new method of image understanding for forms based on model matching is proposed in this paper as the basis of OCR which can read a variety of forms. The outline of this method is described as follows. Ruled lines are extracted from the input image of a form. These lines are used for understanding the form, taking into account their feature attributes and the relationships between them. Each line in the input image of a form as expected to correspond to a line in one of the model forms, which are described as structured features. This correspondence is represented by a node in an association graph where an arc represents compatible correspondences established on the basis of feature relationships. The best match is found as the largest maximal clique in the association graph. Experimental results show the method is robust and effective for poor quality document images and also for various styles of forms.

Flexible and Robust Model Matching based on Association Graph for Form Image Understanding

Abstract:A new method of image understanding for forms based on model matching is proposed in this paper as the basis of an OCR which can read a variety of forms. The outline of this method is described as follows. First, ruled lines are extracted from the input image of a form. After that, several lines are grouped as one to be recognised as data corresponding to a sub-form. These lines and sub-forms are both used for understanding the form, taking into account their feature attributes and the relationships between them. Each feature in the input image of a form is expected to correspond to a feature in one of the model forms, which are described as structured features. This correspondence is represented by a node in an association graph, where an arc represents compatible correspondences established on the basis of feature relationships. The best match is found as the largest maximal clique in the association graph. Experimental results show the method is robust and effective for document images of poor quality, and also for various styles of forms.

Document layout analysis based on emergent computation

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.

Logical Structure Analysis of Document Images Based on Emergent Computation

A new method for logical structure analysis of document images is proposed in this paper as the basis for a document reader which can extract logical information from various printed documents. The proposed system consists of five basic modules: text line classification, object recognition, object segmentation, object grouping, and object modification. Emergent computation, which is a key concept of artificial life, is adopted for the cooperative interaction among modules in the system in order to achieve effective and flexible behavior of the whole system. It has three principal advantages over other methods: adaptive system configuration for various and complex logical structures, robust document analysis tolerant of erroneous feature detection, and feedback of high-level logical information to the low-level physical process for accurate analysis. Experimental results obtained for 150 documents show that the method is adaptable, robust, and effective for various document structures.

Table structure analysis based on cell classification and cell modification for XML document transformation

A new method of table structure analysis based on cell classification and cell modification is proposed in this paper as the basis of an OCR which can convert a variety of printed tables into XML documents in accordance with a specified XML schema. The outline of this method is described as follows. Firstly, cell features defined by ruled lines, which correspond to data fields, are extracted from the input image of a table. After that, each cell is classified to identify the irregular table whose ruled lines are not gridded and is modified to form regular cell arrangement. Next, the hierarchical table structure consisting of a regular row structure of cells is extracted from the modified regular table and is described using a DOM tree. In this case, logical objects within a cell are extracted and are converted into a sub-tree in the DOM tree. Finally, this DOM tree is transformed into a target XML document by an XML parser with information extraction process. Experimental results show the method is effective in transforming various printed tables to various XML documents.

Analysis of required elements for a next-generation document reader on the basis of user requirements

This report describes the required elements for a next-generation document reader. These results are derived from the analysis of the user requirements. The next-generation document reader will have a high degree of adaptability to a wide range of applications and specifications. Key requirements for the document reader are accurate character recognition, accurate layout analysis, portability, and adaptability. The prototype is developed to test the next-generation document readers functions.