Flame Oscillation Frequency Based on Image Correlation

陈志斌,胡隆华,霍然,祝实
DOI: https://doi.org/10.3321/j.issn:1006-8740.2008.04.015
2008-01-01
Abstract:Oscillation frequency is a key parameter of flame.Based on Matlab image processing tool,this paper provides a new method of getting the flame oscillation frequency by applying fast Fourier transform to time variation of image correlation coefficients of a continuous image sequences.The gasoline flame oscillation frequency of a square burner of 0.7 m×0.7 m is obtained by the new method,which is 1.80 Hz,agreeing well with the theoretically predicted value by Pagni equation,which is 1.71 Hz.In addition,the image correlation coefficients are of great difference for different colors considered,which are R,G and B,in the same image sequence;however,their changing frequency is the same,demonstrating that colors have no effect on frequency value obtained by this method.
What problem does this paper attempt to address?