Parameterized characterization of elliptic sperm heads using Fourier representation and wavelet transform

In this paper, we have studied a new method of sperm morphological classification using Fourier representation and a dyadic wavelet transform (WT). Sperm head boundaries with elliptic shape were parameterized by discrete Fourier transform and reconstructed with dyadic data points. The enclosed area of boundaries as a classification feature was calculated and transformed by wavelet transform. At successive dyadic resolution levels, the root-mean-square distances from the four standard classes were calculated and the boundary was classified into its nearest. The results suggest that if we have sufficiently large databases, the classification by wavelet decomposition is reliable and can be clinically used as a substitution