A preprocessing algorithm for hyperspectral images of vessels

Hyperspectral imaging technology has been applied in histology and cytology for a long time. Image quality and spectra are usually affected by light source, optics and sensors in hyperspectral imaging systems. This paper proposes a continuum removal algorithm and radiometric correction preprocessing method to eliminate these influences. The algorithm removes the continuum of image after the spectra correction is applied to the original one. Experimental results show that the algorithm evidently eliminates these influences, and the quality of the image is promoted. The result of the algorithm, which is applied to the hyperspectral image of vessel cell, can be used for further segmentation and classification.