The project aims to utilize methods of digital image processing and pattern recognition in automatic classification of phytopathogenic fungi (within a selected taxonomy group) from microscopy images. First, proper biological specimens will be acquired, photographed and documented in cooperation with the Department of Phytopathology. Next, a set of methods for processing the microscopy images will be proposed and implemented. These include preprocessing, i.e. rectification of degradations introduced by microscopy imaging, image registration, fusion or segmentation. The methods must avoid introducing artificial information that could lead to false conclusions regarding the biology of the species. Afterwards, a number of representative specimens (the "training set") will be selected and manually classified. A set of salient features measured in the images will be selected or composed for the purpose of classification. Finally, a classifier optimal for automatic classification of phytopathogenic fungi within the selected taxonomy group will be proposed and constructed. The final toolbox will not only make taxonomization easier but also provide a set of methods for automatic processing of microscopy images for the purposes of further processing or presentation. We suppose it will be possible to modify the results for the purposes of processing other specimens from similar taxonomy groups.