Master of Science (MS)
Electrical and Computer Engineering
First Committee Member
Number of Pages
In this thesis, a hierarchical tree representation is introduced to represent the logical structure of a form document. But different forms might have the same logical structure, so the representation will be ambiguous. In this thesis, an improvement is proposed to solve the ambiguity problem by using the physical information of the blocks. To fulfill the application of hierarchical tree representation and extract the physical information of blocks, a pixel tracing approach is used to extract form layout structures from form documents. Compared with Hough transform, the pixel tracing algorithm requires less computation. This algorithm has been tested on 50 different table forms. It effectively extracts all the line information required for the hierarchical tree representation, represents the form by a hierarchical tree, and distinguishes the different forms. The algorithm applies to table form documents.
Documents; Form; Identification; Layouts; Recognition
Electrical engineering; Computer science
University of Nevada, Las Vegas
If you are the rightful copyright holder of this dissertation or thesis and wish to have the full text removed from Digital Scholarship@UNLV, please submit a request to email@example.com and include clear identification of the work, preferably with URL.
Luo, Kai, "Recognition and identification of form document layouts" (2003). UNLV Retrospective Theses & Dissertations. 1602.