Myanmar Portable Document Format Recognition with Formatting

Abstract

This paper contributes Myanmar Printed Character Recognition with format. This system consist recognition and formatting. It recognizes for Myanmar Portable Document Format (.pdf) such as font size, font style, alignment and table. It converts the existing document to Machine Editable Word Document (.doc). It contains paragraph and table classification. In table classification, table recognition and formatting can also be performed. The extraction of text format, paragraph format and table format can be done efficiently. The system is based on MICR (Myanmar Intelligent Character Recognition) which is a kind of ICR (Intelligent Character Recognition). MICR uses statistical and semantic information which includes width and height ratio, black stroke counts, number of loops, open directions and histogram value, etc. The final decision is made by the voting system. The system use image processing and Matlab programming.