三维激光扫描曲面重构算法研究

Research on 3D Laser Scan Surface Reconstruction Algorithms

围绕三维激光扫描曲面重构做了以下工作:(1)分析了三维激光扫描仪的工作原理;(2)研究了利用多层感知器神经网络用于曲面重构,提出了基于多层感知器网络的曲面重构算法(M LPSR)。为了有效评估算法的性能,通过CYRAX 2500实测数据并利用M ATLAB环境进行仿真实验,结果表明:这种算法具有较高的重构精度和较快的重构速度,尤其在曲面受到破坏或不完全时特别有效,是值得推广的曲面重构算法。

The principle of acquiring 3D laser information methods based on computer vision is introduced. Due to the virtues of the neural network,the multilayer perceptions trained with the back-propagation algorithm and the radial-basis function neural network are employed to reconstruct 3D laser surface. In order to evaluate the algorithms, many computer simulation experiments are carried out. In these experiments, series of points from CYRAX2500 and MATLAB are used to train these neural networks and reconstruct the surface. The results show that the precision and speed of the MLPSR （Multilayer perceptions surface reconstruction） algorithms are higher than the traditional algorithm. Specially, when partially worn or damaged,the MLPSR algorithms are the reconstructed surfaces are complex or incomplete or more effective.