Filtering enhanced tomographic PIV reconstruction based on deep neural networks. Issue 1 (1st March 2020)

• Record Type: Journal Article
• Authors: Liang, Jiaming; Cai, Shengze; Xu, Chao; Chu, Jian
• Is Part Of: IET cyber-systems and robotics. Volume 2:Issue 1(2020)
• Page Start: 43
• Page End: 52
• Publication Date: 2020-03-01
• Subjects: image reconstruction -- tomography -- biomimetics -- velocimeters -- velocity measurement -- neural nets -- cameras -- flow visualisation; enhanced tomographic PIV reconstruction -- deep neural networks -- tomographic particle image velocimetry -- three‐dimensional flow field -- spatial particle distribution -- multiple cameras -- different viewing angles -- popular reconstruction method -- multiplicative algebraic reconstruction technique -- high‐computational speed -- low particle seeding reconstruction -- dense particle distributions -- symmetric encoder–decoder -- reconstruction quality -- particle field -- MART approach -- regenerated image -- blurred particles -- irregular particles -- trained neural network -- ghost particles -- filtering method -- reconstruction accuracy -- estimated velocity fields; Robotics -- Periodicals; Cybernetics -- Periodicals; Cybernetics; Robotics; Periodicals; 629
• DOI: 10.1049/iet-csr.2019.0040
• Languages: English
• ISSNs: 2631-6315

Abstract

Abstract : Tomographic particle image velocimetry (Tomo‐PIV) has been successfully applied in measuring three‐dimensional (3D) flow field in recent years. Such technology highly relies on the reconstruction technique which provides the spatial particle distribution by using images from multiple cameras at different viewing angles. As the most popular reconstruction method, the multiplicative algebraic reconstruction technique (MART) has advantages in high computational speed and high accuracy for low particle seeding reconstruction. However, the accuracy is not satisfactory in the case of dense particle distributions to be reconstructed. To overcome this problem, a symmetric encode–decoder fully convolutional network is proposed in this paper to improve the reconstruction quality of MART. The input of the neural network is the particle field reconstructed by the MART approach, while the output is the regenerated image with the same resolution. Numerical evaluations indicate that those blurred or irregular particles can be significantly refined by the trained neural network. Most of the ghost particles can also be removed by this filtering method. The reconstruction accuracy can be improved by more than 10% without increasing the computational cost. Experimental evaluations indicate that the trained neural network can also provide similar satisfactory reconstruction and improved velocity fields.