摘要
提出了一种基于预识别及联合代数重建技术(SART)的单光场相机三维流场重建方法。首先结合光场图像,采用光线追迹的预识别技术预先筛选出控制体中包含示踪粒子的离散体素,实现对权矩阵结构的优化,然后利用插值法准确计算出测量体体素与图像像素之间的权函数。最终采用SART实现对示踪粒子三维空间位置的重构。单颗粒的重建结果表明:重建颗粒的中心位置准确,且其沿深度方向的拉伸程度较经典算法(MART)的结果明显减弱。此外,该算法在提高了重建效率的同时也降低了对权矩阵存储空间的需求。
In this paper,a pre-recognition-based SART algorithm is proposed to reconstruct the tracer particle distribution in the light field PIV.With the light field images,the pre-recognition technique based on the ray tracing method is utilized to eliminate the non-intensity discrete voxels in the measurement volume.A bilinear interpolation method is then used to calculate the weight coefficients of the voxels.The simultaneous algebraic reconstruction technique(SART)is finally used to reconstruct the particle volumes.The reconstruction result of the single particle indicates that the proposed reconstruction algorithm is capable of reconstructing the tracer particle with higher location accuracy and shorter depth elongation compared with the conventional MART algorithm.Additionally,the proposed algorithm also improves the reconstruction efficiency and releases the computational load.
作者
朱效宇
张彪
李健
许传龙
ZHU Xiao-Yu;ZHANG Biao;LI Jian;XU Chuan-Long(Key Laboratory of Energy Thermal Conversion and Control of Ministry of Education,School of Energy and Environment,Southeast University,Nanjing 210096,China)
出处
《工程热物理学报》
EI
CAS
CSCD
北大核心
2020年第6期1445-1451,共7页
Journal of Engineering Thermophysics
基金
国家自然科学基金资助项目(No.51676044)
江苏省重点研发计划(社会发展)(No.BE2018705)。
