基于优化BP神经网络的点云孔洞修补效果比较分析

针对逆向软件对三维点云模型孔洞修补精度不高,提出在软件修补的基础上利用神经网络算法进行调整的修补策略。以挖掘机斗齿内腔孔洞为例,利用思维进化算法(MEA)优化BP神经网络的初始权值和阈值,得到改进的神经网络模型MEA-BP,对修补数据进行调整,并与PSO-BP神经网络、GA-BP神经网络孔洞修补结果做出对比。结果表明:MEA-BP神经网络模型在预测精度有所提高的情况下,效率更高,修补效果更加优越,能够有效地、精确地对孔洞进行修复。

Comparative Analysis of Hole Repairing Effect of Point Cloud Based on Optimized BP Neural Network

In view of the low precision of reverse software for 3D point cloud model hole repair, a repair strategy based on neural network algorithm on the basis of software repair was proposed. Taking the hole of the inner cavity of the bucket tooth of a excavator as an example, the mind evolution algorithm (MEA) was used to optimize the initial weight and threshold of the BP neural network. The improved neural network model MEA-BP was obtained, and the repair data were adjusted, and the results were compared with those of the PSO-BP neural network and the GA-BP neural network hole repair.The results show that the MEA-BP neural network model has higher efficiency and better repair effect, and can repair holes effectively and accurately under the condition that the prediction accuracy is improved.