The object of the project is the development of an automated method for the identification of nano- and microscale particles based on electron microscopic investigations. The classification shall be performed by an algorithm based on high-resolution electron microscopic (EM) images (morphology) and energy dispersive X-ray spectra (EDS) (element composition) for single particles. This requires training and validation of the method for known particle types such as diesel soot, fibres, inorganic, mineral and metallurgical particles.
In particular, an improvement of the classification of carbon-dominated particles such as diesel soot, particles from biomass combustion, particles of biological origin and organic secondary particles is to be achieved through advanced image processing. The aim is to separate a particle mixture as exhaustively as possible. For this purpose, experimental collection methods will be developed and tested, which will enable the most representative sampling of airborne particles possible. The method will be embedded in a suitable software environment and published scientifically together with the algorithm in order to enable developing the method further for regulatory purposes.
Unit 4.5 "Particulate Hazardous Substances, Advanced Materials"