題名: | Handwritten Numeral Recognition Based on Reduced Features Extraction and Fuzzy Membership Functions |
作者: | Jou, Chichang Hsiao, Tai-Yuan Lee, Hung-Chang |
關鍵字: | Handwritten Numeral Recognition Pattern Recognition Features Extraction Fuzzy Membership Functions |
期刊名/會議名稱: | 2002 ICS會議 |
摘要: | We design and develop a handwritten numeral recognition system based on reduced features extraction and fuzzy membership functions with the intention to find a minimal set of feature primitives without sacrificing the recognition rate. We first perform preprocessing of smoothing and thinning to obtain a skeleton for each image. We then extract the following feature primitives for each skeleton: loop, horizontal, vertical, C-like curve, and D-like curve. Two fuzzy S-functions are used as membership functions to estimate their likelihood of being close to the top and to the bottom. A tree-like classifier based on the reduced feature primitives and fuzzy memberships is then applied to recognize the numerals. |
日期: | 2006-10-24 |
分類: | 2002年 ICS 國際計算機會議 |
文件中的檔案:
檔案 | 描述 | 大小 | 格式 | |
---|---|---|---|---|
ce07ics002002000338.PDF | 155.61 kB | Adobe PDF | 檢視/開啟 |
在 DSpace 系統中的文件,除了特別指名其著作權條款之外,均受到著作權保護,並且保留所有的權利。