Οptical microscopy experimental setup for the quantitative analysis of pollen particles via autofluorescence

Abstract

This diploma thesis presents the development of an optical microscopy system for the quantitative analysis of pollen particles through their intrinsic autofluorescence. The proposed approach exploits the natural autofluorescence of the exine to enable label-free imaging while preserving the native morphology of the sample, with darkfield microscopy providing complementary structural verification. A complete Python-based image-processing workflow was developed, combining CLAHE, median filtering, Otsu thresholding, and morphological operations to extract quantitative morphological and photometric features. The results demonstrate that consistent quantification of biological microstructures is feasible using a low-cost optical setup without chemical staining, provided that adequate signal quality and appropriate feature selection are ensured.