A model with some parameters is used to forecasting the beam camber in the construction . based on the differences between the design value and the survey of the bridge deck elevation in constructing the pre - stressed and so on working condition , the model parameters are identified by ann , and the beam camber of different sections are given 建立施工预拱度的数学模型，根据施工时的预应力张拉后等工况桥面实测标高与设计值的差异，用bp神经网络识别预拱度模型的参数，确定有关各截面的预拱值。
For example , firstly , the bam neural network is used to recognize the vehicle style , the trains of thought is concise , the learning is simple , rapid , the method is feasible secondly , color edge detector colorprewitt is used to locate the vehicle license plate with high location ratio 一、提出利用bam神经网络识别车型，比用前馈网络识别，学习速度快，收敛速度快，同时克服简单几何特征识别车型时容易引起混淆的问题。二、利用彩色图像边缘检测算子对牌照区域进行定位，定位准确，提高总体识别率。
The paper has formed a new type of method on logging facies analysis by abstract some parameters from log curve which reflect facies characteristics , and neural network identification study , conjunction with database technology . the study has the characteristics that new and original in choice of theme , and practical in work . 论文通过从测井曲线中提取反映沉积相特征的曲线形态参数，以及神经网络识别方法研究，结合数据库技术，形成了一套具有一定特色的测井沉积相分析方法，本研究具有选题新、实用性强等特点。
The deformable template methods ( dtm ) can resolve some problems that statistical and ann recognition methods can not accomplish . the prominent advantage of dtms is that they can make full use of the prior knowledge on the shape of the characters , and can deal with deformed characters without training work 可变形模板方法可以解决一些统计和神经网络识别方法不能解决的一些问题，其突出的优点在于能充分利用人对字符形状的先验知识，不需要用大量的字符样本进行训练，它对解决小字符集识重庆大学博士学位论文别问题具有一定的什值。
