A Study On The Relevant Technology Post-Processing Of Mongolian Characters Recognition

Optical character recognition is a very important branch in pattern recognition research area, and also a comprehensive subject involving the image processing technique, artificial intelligence, theory of information, formal language, automata, fuzzy mathematics, linguistics and so on .At present, optical character recognition has achieved some success in dealing with prevailing character sets such as Chinese and Latin, which has already been applied.Since 1980s, some research on elementary engineering construction and theoretical research on Mongolian information processing have been done, and a number of applied systems have been developed. Yet contrast with abroad and domestics other character recognition, the Mongolian characters recognition research started later.The article researched post-processing of Mongolian Characters Recognition on the basis of sufficiently considers the Mongolian with self-characteristics and the present status about Mongolian characters recognition research from the angle of optical character recognition. The paper solves the ambiguity problem in the process of Code conversion using the Mongolian language prior knowledge, which is playing an active part in improving the system recognition accuracy.Here is the structure and the content of this article: Chapter one mainly introduces some basic research which closely connects with the paper, namely the target of the study, the some basic concept of Optical character recognition, the previous status and the difficulty in Mongolian characters recognition research, the main research content of the paper, etc. Chapter two briefly introduces a core recognition scheme of "Multi-Font Printed Mongolian (the mixture of Chinese and English) Character Recognition System" and the establishment of recognition system's base resources; Chapter three discusses the realizing of "Multi-Font Printed Mongolian (the mixture of Chinese and English) Character Recognition System" post-process, including the grapheme combination and the conversion between presentation code and character code. Chapter four explains the further work and prospect for relevant research; the conclusion is the summary of the whole article. The paper lists the appendix after the content, such as basic recognition sets, different classes standard sample, regular example, the sample of OCR dictionary.