A system for the recognition of biofouling on the surface of the monuments of cultural heritage

This paper proposes a system for the recognition of biological contaminants existing on the surface of the cultural heritage monuments. Knowing the type of biological contaminants is necessary to protect the monuments from their harmful impact. The proposed recognition system uses as input a series of images of cultural heritage monuments. The series contains images obtained in the visible and near infrared spectral bands. During preprocessing a series, all images are converted to a selected shooting point, and the background is removed. Feature vector is built from combinations of formal vegetation indices. To recognize the type of biological contaminants, a pretrained classifier based on SVM method with RBF kernel is used.