Classification of human sperm heads using elliptic features and LDA

For diagnosis of infertility in men semen analysis is conducted in which sperm morphology i.e. the size and shape of the sperm, is one of the factors that are evaluated. Since manual assessment of sperm morphology is time consuming and subjective, automatic classification methods are being developed. Automatic classification of sperm heads is a complicated task due to the “within class” differences and “between class” similarities. To automatically classify the sperms, appropriate features should be extracted from their microscopic images. In this research, a set of previously proposed features is extracted and examined in an automatic framework in order to evaluate their discriminating capacity in classifying sperms into four classes of shapes (Normal, Tapered, Pyriform and Amorphous). Also, a new set of features called elliptic features is proposed and added to the original features to improve the classification results. Both sets of features are used with Linear Discriminant Analysis (LDA) classifier. It is shown that adding these new features, significantly improves the discrimination between those classes of sperm shapes.