| 1. | Cobol , identification division
普通商用语言识别部分 |
| 2. | In this paper , a handwritten recognition system is set up . this system includes learning part and recognition part
本文建立了一个脱机手写数字识别系统，系统主要是由两个部分构成即：学习部分和识别部分。 |
| 3. | This paper studies the technologies of image recognition , including the image edge detection and image matching and their applications based on the project of the image monitor system of the unmanned substations
本论文结合具体的变电站图像监控系统的项目，对图像识别部分的图像边缘检测与图像匹配技术及其应用进行了研究。 |
| 4. | The human recognition section provides an efficient algorithm , which performs well in the system by employing the unique features of human figure and the priori knowledge of the scene
在人体识别部分，本文利用人体在运动时具有的独特形体特征，以及根据场景的先验知识，构造了准确高效的人体识别算法，在系统使用中获得了令人满意的效果。 |
| 5. | In the learning part , three bp neural networks are trained . there are four steps in the recognition part . they are preprocessing , getting character , deleting fault character and combined recognition
学习部分主要完成三个bp神经网络的学习训练；识别部分是由下列几个环节构成：预处理环节、特征抽取环节、伪特征消除环节和组合识别器识别环节。 |
| 6. | Based on the characteristic of architectural ichnography and criterion of architectural design fires prevention , the recognizing identification method and main program based on dxf format is proposed
在研究了建筑工程图表示特点和建筑设计防火规范的基础上，给出了基于建筑工程矢量图形文件（ dxf文件）的几何特征识别方法，并给出了识别部分的主要程序。 |
| 7. | Now , there is no real - time defect auto - detection in x - ray system in domestic factory . in this thesis , defect auto - detection and auto - recognition is the emphases of research and it keeps ahead in domestic research
本文主要以缺陷自动检测与识别部分为研究重点，设计并实现的x射线无损探伤缺陷自动检测系统，从而填补了国内x射线无损探伤系统中缺陷自动检测的空白。 |
| 8. | Secondly , we realize a keyword spotting baseline system . but the speed of recognition of this system can not satisfy the real - time requirement . then , we improve the system for real - time application from two aspects
其次,实现一个关键词检出基线系统,利用863语料库离线训练模型库,并使用令牌传递( tokenpassing )算法实现在线识别部分,但该系统的识别速度不能达到实时性要求。 |
| 9. | After discussing the properties of the parameters generated in the signal analysis procedure , the mfcc parameter is finally chosen as the features of the voice . through introduce the hmm model and the ann model ， the paper put forward a new model - hmmnn ， and accomplished the theoretical analyses
在训练与识别部分，通过介绍隐含马尔可夫模型和神经网络模型，采用将cdhmm和自组织神经网络结合而形成的新模型hmmnn ，并进行了理论分析。 |
| 10. | Statistic pattern recognition based on the combination of multiple classifiers is imposed in the char recognition step , which is helped to sub - classify the char by local structure character analysis . that is , first to recognize the car license plate chars roughly with combination of multiple classifiers , and then distinguishes the chars that are very similar in the statistic character by structure analysis . the effect is very good
字符识别部分引入了多分类器集成的统计识别模式，并辅之以局部结构特征分析法进行细分类，即首先用多分类器集成作总体的字符识别，对于部分统计特征非常相似的字符，再加以结构分析来区分，取得了良好的效果。 |
