查詢結果分析
來源資料
相關文獻
- Applying Hashing Search and Fuzzy Fault-Tolerant Algorithms for the Fast Recognition of Multi-Font Printed Chinese Characters
- A Highly Reliable Character Recognition Mechanism
- 應用聯線手寫文字辨識之前處理信號方法
- 聯線手寫日本語文字辨識中之連寫字探討
- 談中文光學文字辨識應用
- Building a Web-based Chinese Information Retrieval System Using Optical Character Recognition Technology: Project Report for Heleni Linton Bursary 1998.
- 應用聯線手寫文字辨識之前處理信號方法
- 聯線手寫文字辨識引進副圖形之探討
- 以類神經網路為基礎之線上手寫中文字辨識系統
- 中文古籍全文資料庫建置比較研究
頁籤選單縮合
題名 | A Highly Reliable Character Recognition Mechanism=一種高可靠性文字辨識機制 |
---|---|
作者 | 黃雅軒; | 書刊名 | 影像與識別 |
卷期 | 4:5 1998.04[民87.04] |
頁次 | 頁34-42 |
分類號 | 312.84 |
關鍵詞 | 文字辨識; Reliable recognition; OCR; Competitive learning; Rejection; |
語文 | 英文(English) |
英文摘要 | This paper proposes a novel method which enables a character recognition system to obtain reliable recognition. In this method, wo thresholds, i.e. class region threshold R �R and dis-ambiguity threshold A �R are used by each character k when the classifier is designed based on the nearest neighbor urle, where R �R defines the pattern distribution region of character k, and A �R prevents the samples not belonging to character k from being ambiguosusly recognized to character k. A novel algorithm to derive appropriate thresholds A �R and R �R is developed so that a better recognition reliability can be obtained through interative learning. Experiments performed on the ITRI printed Chinese character database have achived highly reliable recognition performance (such as 0.999 reliability with a 95.14% recognition rate), which shows the feasibility and effectiveness of the proposed method. |
本系統之摘要資訊系依該期刊論文摘要之資訊為主。