Joint method for reconstructing three-dimensional temperature of flame using Lucy-Richardson and nearest neighbor filtering using light-field imaging

Abstract

The real-time measurement and reconstruction of three-dimensional temperature are essential when studying the combustion process inside a flame. In the proposed sectioned flame light field imaging model, flame sections are reconstructed based on the characteristics identified using the Lucy-Richardson or nearest neighbor filtering methods, or a joint method involving a combination of both, to improve on the accuracy of two-dimensional temperature reconstruction. Compared to the results of a previous work, the proposed method increases the resolution in the depth of field direction and the extraction of the flame temperature of all sections from the same height ensures the accuracy of the reconstructed temperature distribution across the cross-section of the flame. The relative error when compared to the actual flame is less than 10%. Thus, this approach allows the accurate estimation of the three-dimensional temperature distribution inside a flame, and enables the rapid and effective reconstruction of the three-dimensional temperature field.