This thesis describes a new line segmentation method that is optimized for medieval manuscripts. Using a thinned version of the binarized document image, the segmentation algorithm extracts two types of salient features from the handwritten patterns: nodes, whose distribution allows for the detection of line axes; segments, which are labeled according to the nodes they connect. This method obtains very good results on manuscripts that are usually considered hard to segment because of the numerous overlapping and touching lines. By contrast with many existing segmentation algorithms, this method does not rely on user-entered parameters and is not overly sensitive to the quality of the preprocessing treatments. Although more work is required to make it resistant to fluctuating lines, this line separation technique can already handle a large set of medieval documents and provides a useful input to a character segmentation program. / Line segmentation techniques in off-line handwriting recognition -- Line segmentation with the profile method -- Feature-based line segmentation -- Tests and conclusions. / Department of Computer Science

The problem of optical pattern recognition is a broad one. It ranges from identifying shapes in aerial photographs to recognizing letters in hand or machine printed words. This thesis examines many of the issues relating to pattern recognition and, specifically, those pertaining to the optical recognition of characters. It discusses several approaches to various parts of the problem as an illustration of the variety of methods of attack. Some of the particular strengths and weaknesses of those approaches are discussed as well. Finally, a new method of approaching OCR is introduced, developed, and studied. At the conclusion, the study is summarized, the results are examined, and suggestions are made for continued research. / Department of Computer Science

Thesis (M.S.)--University of Nevada, Reno, 2006. / "May, 2006." Includes bibliographical references (leaves 94-99). Online version available on the World Wide Web.

Thesis (M.S.)--Rochester Institute of Technology, 1989. / "References": leaves 50-51.

Thesis (M.S.)--Rochester Institute of Technology, 1992. / Typescript. Includes bibliographical references.

Thesis (M.S.)--Ohio University, June, 1992. / Title from PDF t.p.

Thesis (Ph.D.)--Rochester Institute of Technology, 2008. / Typescript. Includes bibliographical references (p. 175-179) and index.

Thesis (Ph. D.)--University of Hong Kong, 2005.

Thesis (M.Comp. Sc.)--Dept. of Computer Science, Concordia University, 1995. / Includes bibliographical references (leaves 119-128) and index. Available also on the Internet.

Thesis (Ph. D.)--University of Missouri-Columbia, 2000. / Typescript. Vita. Includes bibliographical references (leaves 96-109). Also available on the Internet.