基于球面DOG小波框架的骨料形状重构研究

AGGREGATE PARTICLE SHAPE RECONSTRUCTION BASED ON SPHERICAL DOG WAVELET FRAME

  • 摘要: 基于高斯差分小波(Difference of Gaussians, DOG)向球面上拓展产生的球面DOG小波函数,提出了基于球面DOG小波框架重构骨料形状的新方法。为探讨该方法的优势,采用该方法和传统球谐重构法,对采用三维扫描仪获取机制砂的表面点云数据进行重构,分析了球面DOG小波框架方法的参数设置,以及准确性。结果显示,随着球面DOG小波框架qmax的增大,重构骨料的形貌逐渐接近真实形态。且当qmax达到4时,支撑集足够小的球面DOG小波可有效避免高阶次球谐级数重构多棱角骨料时由于振铃效应导致的重构精度下降问题。

     

    Abstract: A new method for reconstructing aggregate shape based on the spherical Difference of Gaussians (DOG) wavelet frame was proposed, which is generated by expanding the DOG onto the sphere. To explore the advantages of this method, several manufactured sand particles were scanned by the three-dimensional laser scanner and respectively reconstructed by traditional spherical harmonic reconstruction method and spherical DOG wavelet frame reconstruction method. The results show that the shape of the reconstructed aggregates gradually approaches the real form as the maximum order of the spherical DOG wavelet frame (qmax) increases. When qmax reaches 4, the spherical DOG wavelet with a sufficiently small support set can effectively avoid the ringing effects, which might be observed when using a high degree spherical harmonic series to reconstruct the multi-angled aggregates.

     

/

返回文章
返回