字識別率 的英文怎麼說

中文拼音 [zìzhìbiélǜ]
字識別率 英文
word recognition rate
  • 字 : Ⅰ名詞1 （文字） character; word 2 （字音） pronunciation 3 （字體） form of a written or printed ...
  • 識 : 識Ⅰ動詞[書面語] （記） remember; commit to memory Ⅱ名詞1. [書面語] （記號） mark; sign 2. （姓氏） a surname
  • 別 : 別動詞[方言] （改變） change （sb. 's opinion）
  • 率 : 率名詞（比值） rate; ratio; proportion
  • 識別 : 1 （辯別; 辯認） discriminate; distinguish; discern; tell the difference; spot 2 [計算機] identif...
  1. Research show that wavelet varied - grid feature vector is characterized by high - stable and high - distinguish. based on this vector the apery cognitron has solved the harmony of single - classifier and multi - classifier and the harmony of multi - feature. the data shows that the recognition rate and reliability has been effective improve

    實驗數據表明，小波變網格特徵向量具有穩定性高、區分性強的特點，基於此的智能字元識別機解決了單、多分類器協調和特徵協調問題，在應用快速二值化方法加強處理實時性的同時有效地提高了車牌字元的識別率和識別可靠性。
  2. Firstly, the pretreatment of character to be recognized is researched, stressed discussing the subdivisional process of character, and a kind of fast shape preserving morphological thinning algorithm is used. in the following, how to construct the structure model on the basis of characteristic point ) sub - stroke and their interrelation is discussed, and a kind of method that describe sub - stroke through the feature of the sequence of curvature is brought forward. finally, the paper adopts the recognising model of printing number that base on repository, and describes the constructing method of repository

    文中首先對待識別字元的預處理進行了研究，著重探討了字元的細化過程，採用了一種基於數學形態學的保形的快速細化演算法；接著探討了如何以特徵點和子筆段及其相互關系為基礎構造結構模型，提出了一種以曲率序列性質描述子筆段的方法：最後採用了以知識庫為基礎的印刷體數字識別模型，並詳細地描述了知識庫的構造方法。
  3. The thesis describes the detailed procedures of hand - writing data processing, including chinese character dot matrix building and display, vector data storage, chinese character dot matrix compression, transfer protocol and hand - writing chinese character re - building. also, stroke order based hand - writing chinese character recognition method is explored, and characteristic - recognition methods for chinese character with similar stroke order codes and the structure of hand - writing chinese character recognition dictionary are discussed

    第四，設計了服務器端的服務軟體，可以實現數據的解碼、重建、筆劃特徵提取、識別、數據管理功能；採用了基於筆段筆順編碼和結構特徵的識別方法並設計了識別字典的結構，可以有效的提高識別速度，達到較高的識別率。
  4. These series of processing methods offer high stability and good repetition. abundant experiments have proved that this system gives a feasible method in the recognition of the reading of the water meter. the recognition rate of this system has reached the request

    利用本文的演算法開發的數字圖像水表讀數自動識別系統對現場採集的大量水表圖像進行測試，結果表明該方法抗干擾能力強，識別率完全達到要求。
  5. In modern chinese, classified word formation is a big advantage that has greatly increased the rate of gram reusing and lexis identification and thus has promoted chinese to be a concise and highly efficient language

    摘要現代漢語根字屬性及其歸類構詞的基本規律，使得歸類構詞成為現代漢語的一大優勢，大大提高了漢字的復用率和漢語詞匯的識別率，使漢語成為一種「簡捷高效」的語言。
  6. In order to overcome the weakness of conventional segmentation algorithm in ocr, this paper presents a new segmentation method for gray document image

    摘要字元切皆目前已經成為限制ocr技術發展的瓶，對于圖像質量較差、中英文混排和背景色變化的文本圖像，傳統切分方法造成的切分錯誤使得文字識別率大大降低。
  7. The software uses delphi 6. 0 do experiments, from the effect of experiment. we could see, its correct recognition rate has reach up to 98 % ( 10 candidates ) this result is satisfying and encouraging

    本系統採用delphi6 . 0進行軟體開發，對寫字較為規范正規的手寫體，其識別率達到98以上（ 10候選） ，取得了令人滿意的結果。
  8. Those adaptions make the afrr increase to 97 %

    這些改進使得平均字體識別率達到了97 。
  9. But the afrr reaches 95. 62 % after the above wo rks

    經過以上的工作，平均字體識別率提高到95 . 62 。
  10. Many kinds of ocr technology has gotten very high recognition rate

    各種類型的漢字識別技術已經達到了相當可觀的識別率。
  11. This recognizer can get quite low wer by testing on sietill german connected digits speech database

    利用該語音識別器對sietill德語連續數字語料庫進行訓練和識別，可以達到很高的識別率。
  12. According to markov language model, use this probabilistic relationship between the chinese characters into chinese character post - processing

    根據markov語言模型，將同現概率這種文本上下文相關信息應用到漢字識別后處理中。
  13. Thirdly, decrease the difficulty of the character recognition through studying the method of character preprocessing ; at last, some effective methods of feature extraction, such as the one based on image transform with the invariant features which keep stable while the picture move, rotate and zoom in or out, and the other focused on the shape of character and features of direction

    三、深入探討了字元識別的預處理方法，減少字元識別時的運算量和難度。四、討論了三種特徵向量的提取方法，一種是基於平移、旋轉和尺度不變性的圖像變換法，一種是計算中心矩的方法，再就是強調字元形狀和結構的方向特徵量提取法，實驗中通過最小距離法進行字元的分類識別，匹配時間少，識別率有所提高。
  14. Secondly, comparing with the algorithm that was given by english scholars a. k. nandi and e. e. azzouz in 1995, this dissertation presents a modified automatic recognition algorithm of six kind of digital modulation types such as 2ask, 4ask, 2psk, 4psk, 2fsk and 4fsk, which only uses four key features, and has a very great exaltation in the rate of correct recognition and the performance of signal to noise, we puts forward the constitution of the best threshold which is applicable to the signal that changes within the scope of the signal to noise rate ( snr ) from 5db to 30db, and the overall success rate is not lower than 92 % when snr is over 7db

    其次，針對1995年英國學者a . k . nandi和e . e . azzouz提出的針對2ask 、 4ask 、 2psk 、 4psk 、 2fsk和4fsk這六種常用的數字調制信號識別的演算法，提出了一種數字信號調制方式識別演算法（ dmra ）的改進演算法。該演算法僅利用四個特徵參數，就可以識別這六種調制方式，而且在調制識別率和信噪比性能上都有了很大的提高，能夠在信噪比不低於7db時，對實際信號的識別正確率高於92 % 。並且提出了最佳門限的設置方法，設置的門限適用於信噪比在5db ~ 30db范圍內變化的信號。
  15. Some new ideas are proposed in this thesis based on svm and ica : firstly, a modified svm method based on posteriori probability theory is given, which makes the classification super plane corrected from the original one. a better classification result is obtained without finding the best quadric optimization algorithm and large scale training datasets are reduced to small scale training datasets at the same time. secondly, ica is applied to the preprocessing period of the recognition character images for purpose of feature extraction and dimension reduction

    本文在系統研究svm和ica的基礎上提出了以下新的觀點：其一是採用了引入后驗概率的修正svm方法，它在原分類超平面的基礎上不斷修正分類超平面，提高分類正確率，從而避免了尋找最優二次規劃的麻煩，同時將大規模訓練樣本集化為小規模訓練樣本集；其二是應用獨立分量分析ica對需要進行識別的字元圖像預處理，提取字元特徵，降低輸入數據的維數，從而可以為下一步的svm識別過程提供好的數據集，用以提高識別率和識別速度。
  16. In the environment of matlab6. 5, selecting different representative characters to do the experiment gains preferable effect, depending on the text, the identification ratio can reach 98. 60 %

    在matlab6 . 5環境下，選用不同的特徵字及組合，進行了鑒別實驗，取得了較好的效果，在文本依存的情況下，識別率達到了98 . 60 。
  17. 5. at the stage of recognition of license plate character, in order to improve the accuracy of recognition, a new way is invented for it

    5在車牌字元識別階段，為提高車牌字元識別率，設i十了一種新的車牌字簡識別方法。
  18. On the basis of the research above, the improper recognition reason of similar chinese characters is analyzed and a new similar chinese characters recognition algorithm is proposed

    而後，為了提高相似字的識別率，分析了手寫體漢字識別中相似字誤識的成因，提出了一種新的相似字識別方法? ?基於特徵點特徵的部分空間演算法。
  19. The requests of differentiating similar chinese characters under each condition are attained by analyzing. and feature points are extracted efficiently. what ’ s more, the algorithm induced the concept of end points pair and broken points

    通過對這些要求的分析，演算法在進行特徵點特徵提取時，著重地討論了特徵點相關特徵的提取，並引入端點對和斷點對等概念，賦予了特徵點更多的信息，以達到提高相似字識別率的目的。
  20. By this time, the rate of single handwritten character recognition, especially figure recognition, has reached 95 %. but to the sequential characters, or mathematic character recognition there is not a perfect product

    到現在，單字元手寫體識別，尤其是數字識別率接近95 ，但對于連續的字元書寫，尤其是有數學符號的情況下，由於聯機手寫的隨意性比較大，寫作不規范，還沒有一個成熟的產品。
分享友人
分享到 Line

分享到臉書