区域公路网络结构的几何模型及计算机实现、最短路的计算及其辨识在交通量预测中起着重要的作用。
The geometric model and computer programming of local highway network plays an important role in traffic estimate, as well as the calculation of shortcut and its reorganization.
由于坐标测量机几何误差变化规律复杂，采用一般的BP神经网络模型算法，速度慢且难以收敛。
Owing to the complicated variable rule of CMMs geometry error, it's difficult to convergence for using common BP neural network model arithmetic with a slow velocity.
提出了一种多特征融合的地形匹配算法，充分利用地形的各种不同的统计特征和几何特征，构造了一种地形匹配网络模型。
A new terrain matching neural network algorithm mode is constructed by means of multi-features fusion, which includes different statistical and geometrical features.
而对于不同的驾驶室模型，运用该方法，可以训练出其用于反求的神经网络结构，该方法可以很好的对整个驾驶室的座椅几何参数进行反演。
For different cab, using this method, we can train several inverse models to design and optimize the geometric parameters to make the reduction of noise in it.
结果表明：对于矩形平面厅堂，选择少数与厅堂声级相关性高的几何、物理参量作为神经网络模型的输入变量，可以准确地预测厅堂声级。
It shows that the good agreements between measured and calculated results can be obtained if the basic parameters used as inputs to the first layer of the neutral network are reasonable.
分析了坐标测量机几何误差的几种常用模型，提出了基于神经网络的单项几何误差模型。
Several coordinate measuring machine geometry error models of several kinds in common use are analyzed in this paper.
目前，基于路段中心线的二维道路网络模型已经普遍存在，但其在几何特征及拓扑关系表达等方面都难以满足复杂交通系统的需求。
The information of centerlines can not cover the changes of lanes, and the two-dimensional model is also inadequate in the expression of three-dimensional transportation networks.
提出了基于神经网络实现多特征融合的地形匹配算法，充分利用地形的各种不同的统计特征和几何特征，构造了一种地形匹配网络模型。
A new terrain matching neural network algorithm mode is constructed by means of multi-feature fusion, which includes different statistical and geometrical features.
建立了三维变动几何约束网络的运动学模型的一般表示式，从而构成了公差大小优化的等式约束。
The general expressions of kinematic model of 3-dimensional VGCN are presented, which is the equation constraint in optimization of tolerance values.
第一，用巨大的连续几何块创建网络的方法，第二是我们使用的光照模型。
First, the way the mesh is created as a giant continuous piece of geometry.
该算法基于点簇对约束网络图进行归约，求得归约序列然后重构几何模型，具有求解速度快、可靠性高、应用范围广等优点。
The reduction is carried out by using geometric constraint graph based on the point clusters. Then the sequence of reduction is computed and the geometric model is rebuilded.
网络模型能够自动抑制含较大误差控制点对模型纠正精度的影响，在实际应用中可以提高几何纠正效率。
Collinearity Equation Model. Besides, the neural network can eliminate the influence of GCPs with gross error, and hence can better improve the efficiency.
网络模型能够自动抑制含较大误差控制点对模型纠正精度的影响，在实际应用中可以提高几何纠正效率。
Collinearity Equation Model. Besides, the neural network can eliminate the influence of GCPs with gross error, and hence can better improve the efficiency.
应用推荐