查詢結果分析
來源資料
相關文獻
- 中文正、簡印刷字體辨識及互譯系統
- 中文正、簡印刷字體辨識及互譯系統
- 感應卡技術發展及市場概況--RFID應用範圍廣 接受度愈見高漲
- 最安全的人員確認方式--生物辨識系統
- Applying Hashing Search and Fuzzy Fault-Tolerant Algorithms for the Fast Recognition of Multi-Font Printed Chinese Characters
- A Highly Reliable Character Recognition Mechanism
- Why a Statistics-Based Face Recognition System Should Base Its Recognition on the Pure Face Portion: A Probabilistic Decision-Based Proof
- 單一視角下三度空間工件的辨識
- 煉焦工場電腦化控制系統的運用
- 手寫數字辨識模式之建立--結合遺傳演算法與類神經網路
頁籤選單縮合
| 題 名 | 中文正、簡印刷字體辨識及互譯系統 |
|---|---|
| 作 者 | 賀中鼎; 陳鏡明; 李蘇民; | 書刊名 | 電信研究 |
| 卷 期 | 21:4 1991.12[民80.12] |
| 頁 次 | 頁421-430 |
| 分類號 | 312.84 |
| 關鍵詞 | 中文正簡印刷字體; 辨識; 互譯; |
| 語 文 | 中文(Chinese) |
| 中文摘要 | 中文正、簡印刷字體辨識及互譯系統是結合中文印刷用辨識的技術、海峽兩岸正簡字互譯之需求及成熟之中文電腦環境三者而成。辨識率高、辨識速度快及具高親和力之使用者界面是其特色。 本系統由五個模組組成,分別是1.前處理模組、2.圖文分離模組、3.特徵抽取模組、4.快捷圖形比對模組、及5.後處理模組。其中包含了圖文分離、特徵抽取、比對與辨識、特徵學習、及正、簡互譯等關鍵性之技術,分別在文內介紹。目前實際試用結果:以教育部公佈之5401個常用字為基準,其辨識及互譯正確率達93%以上,辨識速度約每分鐘600個字,其性能已達實用水準。本系統已獲中央標準局專利。 |
| 英文摘要 | Printed Chinese Regular/Simple Font Translation and Recognition System is designed for the demand of regular/simple font translation. It mainly uses the technique of Optical Chinese Character Recognition (OCCR) and develops under a well-defined Chinese computer environment. The features of the system are high recognition rate, fast recognition speed and very friendly user interface. It is composed of five modules: (1). preprocess module, (2). text/graphic segmentation module, (3). feature extraction module, (4). high speed image matching module and (5) .postprocess module. Several key techniques of this system, including text/graphic segmentation, feature extraction, matching and recognition, feature learning and regular/simple font translation, will be introduced in the paper. Up to now, the practical test is run under 5401 characters which are announced by the Ministry of Education as a frequently used character set. The recognition and translation correct rate is above 93%, and the speed of rendition is 600 characters/min. |
本系統中英文摘要資訊取自各篇刊載內容。