A spectral and morphologic method for white blood cell classification

The identification of white blood cells is important as it provides an assay for diagnosis of various diseases. To overcome the complexity and inaccuracy of traditional methods based on light microscopy, we proposed a spectral and morphologic method based on hyperspectral blood images. We applied mathematical morphology-based methods to extract spatial information and supervised method is employed for spectral analysis. Experimental results show that white blood cells could be segmented and classified into five types with an overall accuracy of more than 90%. Moreover, the experiments including spectral features reached higher accuracy than the spatial-only cases, with a maximum improvement of nearly 20%. By combing both spatial and spectral features, the proposed method provides higher classification accuracy than traditional methods.