识别器

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.

实验数据表明，小波变网格特征向量具有稳定性高、区分性强的特点，基于此的智能字符识别机解决了单、多分类器协调和特征协调问题，在应用快速二值化方法加强处理实时性的同时有效地提高了车牌字符的识别率和识别可靠性。

The experimental results indicate that the LMBP neural network has higher classification rate, more rapidly converge velocity by comparing with other improved BP algorithm and the Radial Basis Function neural network. Through the experiments of recognizing the sonar image polluted with Gaussian noise, it is proved that the LMBP neural network has good stability and better anti-noise performance.

通过与标准BP算法和几种常用改进型BP神经网络以及径向基函数网络比较，验证了用LMBP神经网络作为声纳图像识别系统中的分类器，能够提高图像的识别率，加快网络的收敛速度，通过对受不同程度高斯白噪声污染的声纳图象的识别，验证了其性能稳定，具有较好的抗噪声性能。

At the recognition stage, we propose the combination of characteristic input and Multi-layer Perception Network, which both reduces information input and improves self-study and fault tolerance of artificial neural network, resulting in a more precise recognition.

在识别阶段，本文采取特征提取和多层感知器网络相结合的识别方法，既可以有效减少信息量的输入，又充分发挥了神经网络的学习和容错能力，从而达到较高的识别精度。

There are two ways to improve the robustness of the speaker recognition system. One is to select a proper type of neural networks, to optimize the parameters, so that it can maintain high performance of sample representation even if the features of voice signal was overlapped on sample space.

本文在对神经网络、话者识别，尤其是鲁棒话者识别的研究现状和发展进行了认真详细的分析的基础上，利用神经网络构建特定话者的特征映射器来改善话者识别系统的鲁棒性。

According to the different face representation mode, there are four identifying methods. They are the mode identifying method based on geometry feature vector, the feature vector identifying method based on algebra feature vector, the facial feature extraction method based on NN and design method of sorter.

根据表征方式的不同，人脸识别方法有：基于层次聚类的几何特征矢量模式识别法、基于代数特征矢量的特征矢量识别法、基于神经网络的人脸特征提取和分类器设计法。

By means of the novelly decoded manner and the well constructed fitting function, the global searching performance for the GA is developed so that optimized RBPNN approaches to the parsimonious structure;(4) A ROLS micro-genetic algorithm was built by integrating the micro-genetic algorithm with good convergence performance based on tiny class of group with the ROLS-SOA.

提出一种新的基于径向基概率神经网络的模式识别融合系统，即多分类器系统，避免了单一径向基概率神经网络分类器的分类局限性，减少了寻求单一分类单元结构最优化时的计算复杂度，通过使用D-S融合算法，大大提高了对复杂模式识别问题的分类精度。

The main points include the following several parts: Firstly, in chapter1 of the dissertation, it narrates the backgroud、researchful signification of face recognition and its development. Also including the presentation of face database、classifier and distance measurement. In chapter2, it introduces the development of Fisher discriminant analysis and correlative theory foundation of locality preserving projects. Based the LPP, then it introduces the method of Laplacianfaces in face recognition.

主要包括以下几个方面的内容：首先，在第一章绪论部分主要介绍了人脸识别技术的应用背景、研究的目和意义，以及人脸识别技术研究方法的发展现状和一些人脸数据库、分类器和距离度量的内容；第二章介绍了Fisher鉴别分析的发展现状以及本地保留影射相关的理论基础，着重讲解了基于该理论的拉普拉斯脸在人脸识别中的应用。

Recognition memory: In an OCR device, the ROM which holds the pattern characteristics of the particular range of founts which the machine is programmed to read.

识别贮记器：在光学字符识别装置内，贮存着一定数量的字款的唯读贮记器。这些字款已纳入程式，可以阅读。

And then, combing a proposed feature data preprocessing mechanism and the multiple classifiers fusion, an ultrasound uterus image recognition framework is proposed.4, A fast and automatic ultrasound uterus image recognition algorithm for the intra-uterine device is proposed and realized. The algorithm is composed of automatic ultrasound uterus image segmentation and the specific recognition framework, which are connected by object of interest.

随后结合特征降维进一步提出了一种特征数据预处理机制，并在此基础上结合多分类器融合提出了一种超声子宫图像识别框架。4、设计并实现了一种快速全自动的超声子宫图像节育环物体识别及定位算法，其包括全自动超声子宫图像分割和特定模式识别框架两大部分，通过感兴趣物体(Obiect of interest，OOI)进行连接。

Experimental results show that the method proposed in this paper has excellent recognition rate and speed.

通过实验对比，基于置信度的多分类器互补集成手写数字识别在识别率和识别速度上达到了满意的结果。

Cynanchum Lingtai apricot production in the average weight 65 grams, the brightly-colored fruit, juicy rich, sweet-sour taste, sweet from the nucleolus, when the late Qing Dynasty famous Shaanxi, Gansu provinces, the Qing imperial court Tongzhi tribute for years.

灵台生产的牛心杏平均单果重65克，果实色泽鲜艳，汁多味浓，甜酸适口，离核仁甜，清末时就驰名陕、甘两省，清同治年间曾为朝廷贡品。

Chenopodium album,Solanum nigrum, and Amaranthus retroflexus were very susceptible to the herbicides. Polygonum persicaria and Abutilon theophrasti were relatively less susceptible to the herbicides, and Lycopersicon esculentum was not susceptible to it. The relationship between reduction rates of weed biomass and PPM values of weed leaves 2,4, and 6 days after treatment was established.

供试的6种杂草对该混剂的敏感性存在显著差异：红心藜Chenopodium album、龙葵Solanum nigrum和反枝苋Amaranthus retroflexus对该混剂最敏感，ED90值分别为47.65、71.67和29.17g/hm2；春蓼Polygonum persicaria和苘麻Abutilon theophrasti敏感，ED90值分别为96.91、114.20g/hm2；而番茄不敏感。