作　　者： ; ;

机构地区： 中国计量学院机电工程学院

出　　处： 《现代制造工程》 2013年第1期31-37,共7页

摘　　要： 基于自组织特征映射(Self-Organizing Feature Map,SOFM)神经网络构建的矩形网格模型可以实现密集散乱点数据自组织压缩,生成双有序点列,但该模型存在矩形网格的逼近误差和边缘误差。为减小矩形网格的逼近误差和边缘误差,改进了矩形网格模型的训练模式。首先采用整个测量点集对矩形网格模型中的所有神经元进行整体训练;然后对矩形网格中的网格神经元的位置权重,沿网格顶点法矢方向进行修正;最后采用测量点集中的边界点集,对矩形网格模型中的网格边界神经元进行训练。算例表明,应用该训练模式,可以有效减小矩形网格的边缘误差,矩形网格逼近散乱数据点集的逼近精度得到大幅提高并覆盖散乱数据点集整体分布范围。 An approach based on the Self-Organizing Feature Map （SOFM） neural network has been developed to reconstruct rec- tangular mesh for the dense 3 D scattered data. However the approach suffers from approximation and boundary problems. A three- steps training method is proposed in order to reduce the approximation error and boundary error. First all the neurons of the mesh model are trained directly over the 3D scattered points. Next the neuron location weights of the mesh model are adjusted along the normal vectors of the mesh vertices. Last only the boundary neurons of the mesh model undergo training by the boundary points of the measured points. As a result of applying the proposed training method, the boundary error is greatly reduced and the mesh is drawn toward the sampled object with higher precision comparing with the original SOFM training algorithm. The feasibility of the developed training method was demonstrated on three examples.

关 键 词： 逆向工程 矩形网格 神经网络 逼近误差 边缘误差 散乱点数据

领　　域： [自动化与计算机技术] [自动化与计算机技术]