Description in English
The aim of the project will be to develop a system for the detection, recognition and tracking of persons, rescuers, vehicles and fire using drone image processing. The use of the system is for assistance of Integrated Rescue Systém (IRS) or for CBRN (Chemical, Biological, Radiological and Nuclear) operations and defense systems. A processed image from drone will bring IRS or defense systems bring information and additional data for exploration, mission management, and correct decision-making in an incident or emergency. The subject of the research and development will be modern digital image processing techniques, which will enable automation of search, detection and classification of defined patterns of persons, rescuers, vehicles and firefights or fire (objects) that may be found in the scene of an emergency. In the captured perimeter, the system also allows to define zones, track and evaluate the trajectories of classified and non-classified objects. Thanks to thermal camera data processing, this system also detects and recognizes objects even under reduced visibility (i.e. in the dark, in the fog, in the smoke). The processed image from drone is primarily targeted at large-scale emergencies (large fires, landslides, storms, floods, or rescue searches for missing persons) where people and rescuers cannot see each other (e.g. large distances, terrain) or reduced visibility (night or smoke). Drones equipped with cameras or thermal cameras with modern image processing techniques will increase the safety of rescuers and allow them to explore the search area or fire much faster and from a remote safe place. During an extraordinary event, this system also allows real-time monitoring of event locations from a height, search for outbreaks, motion detection, or perimeter determination even under reduced visibility (using temperature relief).

Key words in English
drone- image processing- vision system- trajectory- perimeter- camera- public security forces