Soot temperature measurement within 3D flame by light-field imaging based on wave optics theory

Light-field imaging technique provides a new approach to non-intrusive temperature measurement, which can achieve 3D measurement via single detector equipment. We consider wave characteristics of radiation rays in the light-field imaging model in this work. We propose an improved light-field imaging model of an emitting and absorbing medium by combining wave optics theory and radiation transfer process in semitransparent media. The combination of Fresnel diffraction theory and phase conversion principles of the lens provides a theoretical basis for a description of radiation ray propagation in space and across the lens. The point spread function of the light-field camera can be calculated and used to attain a light-field image in accordance with the convolution imaging model. The light-field image of the flame is regarded as input to reconstruct the soot temperature field of a 3D flame. Inverse problem is formulated as a linear optimization problem and solved by Landweber algorithm. Simulated reconstruction results show that axisymmetric and non-axisymmetric temperature distribution can be reconstructed successfully, thereby verifying the applicability of the presented method. The effect of the measurement noise and the Tikhonov regularization on the reconstructed quality of the temperature are investigated as well. The results show that the temperature can be reconstructed with acceptable accuracy under different noise levels. Tikhonov regularization method is proven available in this problem.