A schlieren motion estimation method for water flow velocimetry
-
摘要: 本论文基于光流优化算法,发展了一种适用于水流的纹影特性光流测速算法。在存在密度梯度的条件下,纹影图像的亮度反映了流场折射率的一阶导数,结合纹影亮度方程和流体连续性方程推导了适用于水的物理约束条件,采用二阶散度–旋度正则化作为空间平滑约束条件,基于两个约束条件构建了能量方程,通过变分法对能量方程进行最小化求解获得速度场。以热羽流为例,使用该算法对浮力羽流纹影图像进行了计算,并与互相关算法和传统光流算法的结果进行了比较。结果显示:本研究提出的算法能更好地体现流动特性,得到更高的空间分辨率。该方法基于纹影图像,无需在流场中添加示踪粒子,对流场无干扰,具有结构简单、使用方便等优点。Abstract: In the present study, a schlieren motion estimation algorithm is proposed for two-dimensional seedless water flow measurement. A physical constraint is derived combining the schlieren intensity and continuity equations. The space smoothness constraint adopts the second-order div-curl regularizer. Based on the two constraints, the global cost function is derived and minimized to resolve the velocity field using the variation method. As an example, a buoyant plume in the water tank is tested using a Z-type schlieren imaging system. The sample images are calculated using a correlation algorithm in PIVlab, an optimized optical flow algorithm, and the newly proposed schlieren motion estimation algorithm. The results show that the new algorithm can resolve more details of the flow field with higher spatial resolution, while the velocity gradient is consistent with the continuity characteristics. It also features with good robustness without obvious outlier mistakes. The schlieren setup is simple and cost effective in setup. The proposed algorithm has shown great potentials for velocity measurement in more complex configurations.
-
Key words:
- schlieren imaging /
- velocity estimation /
- optical flow /
- variational method
-
表 1 与水的密度相关的参数
Table 1. Coefficient related to the density of water
a2 a1 a0 b2 b1 b0 1.4578×10–9 –2.3505×10–6 5.3267×10–3 –4.8538×10–7 7.5956×10–4 3.7763 表 2 水温0~100 ℃时水的密度
Table 2. Density of water in the range 0-100 ℃
T/℃ 0 10 20 30 40 50 60 70 80 90 100 ρ/(g·cm–3) 0.999 84 0.999 70 0.998 21 0.995 65 0.992 22 0.988 03 0.98320 0.97778 0.97182 0.96535 0.95840 -
[1] 金光,焦晶晶,吴晅. 典型流场测速技术应用研究进展[J]. 矿山机械,2015,43(12):10-15. doi: 10.16816/j.cnki.ksjx.2015.12.003JIN G,JIAO J J,WU X. Research progress on application of typical flow field velocimetry technology[J]. Mining & Processing Equip-ment,2015,43(12):10-15. doi: 10.16816/j.cnki.ksjx.2015.12.003 [2] 冯旺聪,郑士琴. 粒子图像测速(PIV)技术的发展[J]. 仪器仪表用户,2003,10(6):1-3. doi: 10.3969/j.issn.1671-1041.2003.06.001FENG W C,ZHENG S Q. The development of particle image velocimetry[J]. Electronic Instrumentation Customer,2003,10(6):1-3. doi: 10.3969/j.issn.1671-1041.2003.06.001 [3] MERZKIRCH W,KAUFFMAN C W. Flow visualization[J]. Physics Today,1975,28(11):66-67. doi: 10.1063/1.3069208 [4] FUJISAWA N,FUJITA Y,YANAGISAWA K,et al. Simultaneous observation of cavitation collapse and shock wave formation in cavitating jet[J]. Experimental Thermal and Fluid Science,2018,94:159-167. doi: 10.1016/j.expthermflusci.2018.02.012 [5] PETR V, NIECZKOSKI E, LOZANO E. Shadowgraph optical technique for measuring the shock hugoniot from standard electric detonators[M]//Advancement of Optical Methods in Experimental Mechanics, Volume 3. Cham: Springer International Publishing, 2016: 279-293. doi: 10.1007/978-3-319-41600-7_36 [6] SOSA R, ARNAUD E, MÉMIN É, et al. Schlieren Image Velocimetry applied to EHD flow[C]//Proc of International Symposium on Electro-hydrodynamics. 2006. [7] SETTLES G S. Schlieren and Shadowgraph Techniques: visualizing phenomena in transparent media[M]. Berlin, Heidelberg: Springer, 2001. [8] KEGERISE M A, SETTLES G S. Schlieren image-correlation velocimetry and its application to free-convection flows[EB/OL]. https://www.mne.psu.edu/PSGDL/Pubs/2000-Kegerise-ISFV.pdf. [9] PRAZDNY K. Egomotion and relative depth map from optical flow[J]. Biological Cybernetics,1980,36(2):87-102. doi: 10.1007/BF00361077 [10] HORN B K P,SCHUNCK B G. Determining optical flow[J]. Artificial Intelligence,1981,17(1-3):185-203. doi: 10.1016/0004-3702(81)90024-2 [11] MARTÍNEZ-GONZÁLEZ A,MORENO-HERNÁNDEZ D,GUERRERO-VIRAMONTES J A. Measurement of temperature and velocity fields in a convective fluid flow in air using schlieren images[J]. Applied Optics,2013,52(22):5562-5569. doi: 10.1364/AO.52.005562 [12] SUTER D. Motion estimation and vector splines[C]//Proceedings of IEEE Conference on Computer Vision and Pattern Recognition. 1994: 939-942. doi: 10.1109/CVPR.1994.323929 [13] LIU T S,SHEN L X. Fluid flow and optical flow[J]. Journal of Fluid Mechanics,2008,614:253-291. doi: 10.1017/s0022112008003273 [14] ARNAUD E, MÉMIN É, SOSA R, et al. A fluid motion estimator for schlieren image velocimetry[C]//Computer Vision - ECCV 2006, 2006: 198-210. doi: 10.1007/11744023_16 [15] WANG Q,WU Y,CHENG H T,et al. A schlieren motion estimation method for seedless velocimetry measurement[J]. Experimental Ther-mal and Fluid Science,2019,109:109880. doi: 10.1016/j.expthermflusci.2019.109880 [16] 梅笑寒,王倩,王江涛,等. 纹影特性测速算法修正及权重系数取值研究[J]. 中国科技论文,2020,15(6):694-700. doi: 10.3969/j.issn.2095-2783.2020.06.015MEI X H,WANG Q,WANG J T,et al. Correction and weight parameters study of speed measurement algorithm for schlieren characteristics[J]. China Sciencepaper,2020,15(6):694-700. doi: 10.3969/j.issn.2095-2783.2020.06.015 [17] GOLDSTEIN R S. Fluid mechanics measurement[EB/OL]. Washington: Hemisphere Publishing Corporation, 1983. [18] WEISS L,TAZIBT A,TIDU A,et al. Water density and polarizability deduced from the refractive index determined by interferometric measurements up to 250 MPa[J]. Journal of Chemical Physics,2012,136(12):124201. doi: 10.1063/1.3698481 [19] LIDE D R. CRC handbook of chemistry and physics: a ready-reference book of chemical and physical data[EB/OL]. CRC Press. https://ptabdata.blob.core.windows.net/files/2017/IPR2017-01493/v62_Ex.%201062%20-%20CRC%20Handbook%20of%20Chemistry%20and%20Physics.pdf. [20] 成昊天. 基于纹影成像的湍流运动算法研究[D]. 上海: 上海交通大学, 2019: 54-60.CHENG H T. Study on turbulent motion algorithm based on schlieren images[D]. Shanghai: Shanghai Jiao Tong University, 2019: 54-60. [21] GONO T,SYUTO T,YAMAGATA T,et al. Time-resolved scanning stereo PIV measurement of three-dimensional velocity field of highly buoyant jet[J]. Journal of Visualization,2012,15(3):231-240. doi: 10.1007/s12650-012-0129-y [22] THIELICKE W,STAMHUIS E J. PIVlab – towards user-friendly, affordable and accurate digital particle image velocimetry in MATLAB[J]. Journal of Open Research Software,2014,2(1):30. doi: 10.5334/jors.bl [23] SHAVIT U,LOWE R J,STEINBUCK J V. Intensity Capping: a simple method to improve cross-correlation PIV results[J]. Experi-ments in Fluids,2007,42(2):225-240. doi: 10.1007/s00348-006-0233-7 [24] SCARANO F,RIETHMULLER M L. Iterative multigrid approach in PIV image processing with discrete window offset[J]. Experiments in Fluids,1999,26(6):513-523. doi: 10.1007/s003480050318 [25] SUN D Q, ROTH S, BLACK M J. Secrets of optical flow estimation and their principles[C]//Proc of the 2010 IEEE Computer Society Conference on Computer Vision and Pattern Recognition. 2010: 2432-2439. doi: 10.1109/CVPR.2010.5539939