Fuzzy Rule-Based Handwritten Chinese Character Recognition System
碩士 === 大同大學 === 電機工程研究所 === 85 === The goal of this thesis is to develop a handwritten Chinese character recognition system based on the fuzzy logic theory. Because of the fact that fuzzy theory is found to be naturally effective for any human-like cognition systems and can deal with noisy and...
| Main Authors: | , |
|---|---|
| Other Authors: | |
| Format: | Others |
| Language: | zh-TW |
| Published: |
1997
|
| Online Access: | http://ndltd.ncl.edu.tw/handle/15293821733202088575 |
| Summary: | 碩士 === 大同大學 === 電機工程研究所 === 85 ===
The goal of this thesis is to develop a handwritten Chinese character recognition system based on the fuzzy logic theory. Because of the fact that fuzzy theory is found to be naturally effective for any human-like cognition systems and can deal with noisy and imprecise information effectively, it can be applied to pattern or handwritten character systems with vagueness and uncertainty. The architecture of the developed system is simpler than traditional methods and is robust against vagueness. At first, we introduce how to describe the characteristics of characters using fuzzy variables and how to extract the useful fuzzy features. Then we build the fuzzy rules and finally the handwritten Chinese character recognition system is constructed completely. Simulations demonstrate that the proposed system is effective, robust and has high recogniton rate.
|
|---|