期刊文献+
共找到3篇文章
< 1 >
每页显示 20 50 100
基于离散小波变换的深度树时空模型
1
作者 曲浩 吴楠 吕志强 《计算机科学与应用》 2023年第12期2417-2431,共15页
交通流量预测是智能交通领域的研究热点,有利于提高交通资源分配的合理性和出行政策制定的有效性。新型冠状病毒的爆发,严重影响了交通出行的正常秩序。许多国家为了减少疫情的传播速度均颁布了限制居民出行的政策,导致交通流量数据出... 交通流量预测是智能交通领域的研究热点,有利于提高交通资源分配的合理性和出行政策制定的有效性。新型冠状病毒的爆发,严重影响了交通出行的正常秩序。许多国家为了减少疫情的传播速度均颁布了限制居民出行的政策,导致交通流量数据出现了较高的离散性和不规则性。为了克服数据离散性对交通流量预测的影响,本文采用离散小波变换(DWT)将交通流量数据拆分为离散量、变化趋势和离散基线。为了提升模型对高离散性交通数据预测的准确率,本文设计了两种不同的模型来分别预测变化趋势和离散量。由于出行区域的限定,疫情期间的交通状况变化呈现出小规模聚集性。图卷积神经网络的节点邻接计算方法适用于节点随机均匀分布的图结构,对于节点小规模聚集分布的图结构的计算效果较差。本文提出了一种树卷积网络(TreeCN)来分析交通网络的空间相关性,并采用时序卷积网络来分析交通数据的时间相关性。为了解决交通流量数据的高离散性问题,本文提出了一个离散预测模块(DPM),用于将离散小波变换分离出的离散量转换为高维离散特征。最后,使用离散小波变换对预测的交通数据进行分割,然后将新分割的交通趋势和离散基线与离散预测模块预测的离散模型进行逆离散小波变换,得到最终的交通流预测结果。在对比实验中,将这项工作与现有的高级基线进行了比较,本文模型要优于现有基线模型。 展开更多
关键词 时序卷积网络 树卷积网络 离散小波变换 交通流量预测 离散数据计算
下载PDF
Data Reduction and Triangulation Approach to Scattered Points
2
作者 蒋丹 王兰成 《Journal of Donghua University(English Edition)》 EI CAS 2004年第5期79-82,共4页
For the generation of the model in reverse engineering, a laser scanner is currently used a lot due to the fast measuring speed and high precision. Direct triangulation of data points captured from a physical object h... For the generation of the model in reverse engineering, a laser scanner is currently used a lot due to the fast measuring speed and high precision. Direct triangulation of data points captured from a physical object has a great advantage in that it can reduce the time and error in modeling process. It is important to reduce the number of data points for triangulating points with maintaining precision. To triangulate data points within a tolerance ε a new approach is developed in this paper. Different level of triangulations can be generated directly from data points using the proposed strategy that reduces and triangulates data points based on triangulation of 3D parametric surfaces. An experimental example is presented to demonstrate the effectiveness and efficiency of the proposed algorithm. 展开更多
关键词 TRIANGULATION B-SPLINE NURBS CAD
下载PDF
Geometric Feature Extraction and Model Reconstruction Based on Scattered Data
3
作者 胡鑫 习俊通 金烨 《Journal of Donghua University(English Edition)》 EI CAS 2004年第4期86-89,共4页
A method of 3D model reconstruction based on scattered point data in reverse engineering is presented here. The topological relationship of scattered points was established firstly, then the data set was triangulated ... A method of 3D model reconstruction based on scattered point data in reverse engineering is presented here. The topological relationship of scattered points was established firstly, then the data set was triangulated to reconstruct the mesh surface model. The curvatures of cloud data were calculated based on the mesh surface, and the point data were segmented by edge-based method; Every patch of data was fitted by quadric surface of freeform surface, and the type of quadric surface was decided by parameters automatically, at last the whole CAD model was created. An example of mouse model was employed to confirm the effect of the algorithm. 展开更多
关键词 Reverse engineering Scattered data Feature extraction Model reconstruction.
下载PDF
上一页 1 下一页 到第
使用帮助 返回顶部