Research on three-dimensional flow field measurement in a microchannel using a single-lens dual-camera imaging system
Z. Yang(1,2), T. Cai(1,2), W. Zhou(1,2)
1: School of Energy and Power Engineering, University of Shanghai for Science and Technology, Shanghai 200093, China.
2: Shanghai Key Laboratory of Multiphase Flow and Heat Transfer for Power Engineering, Shanghai, 200093, China.
DOI:
Accurate three-dimensional (3D) flow field measurement in microchannels is crucial to understanding the complex transmission phenomena in droplet-based microfluid systems. This research uses a single-lens dual-camera imaging system based on the principle of out-of-focus ranging (depth from defocus), which can realize the accurate three-dimensional positioning of fluorescent tracer particles and reconstruct the internal flow field of droplets. The system adopts a telecentric industrial lens and a beam splitter to effectively eliminated the issue of inaccurate positioning in defocused images by capturing two out-of-focus images of the same particle under different focal length offsets. In addition, the dual-camera image fusion model based on convolutional neural network (CNN) realizes the three-dimensional positioning of particles by learning the positional relationship between the image characteristics and depth of out-of-focus particles. The model is good at particle depth prediction, and the error range is controlled within ±10%. At the same time, the three-dimensional reconstruction of the droplet flow field in the rectangular microchannel is realized in combination with depth information and particle tracking speed measurement technology (PTV). The experimental results reveal the significant shear-induced circulation and internal recirculation mode near the droplet interface, and explain the droplet movement characteristics in the restricted microchannel.
