Detekcia parezou vyťažených stromov pomocou bezpilotných leteckých prostriedkov

Abstract

The bachelor's thesis focuses on evaluation of using unmanned aerial vehicles (UAV) in forestry for detection of tree harvesting. The thesis work is divided into two parts. The first of them contains a theoretical part, which generally describes Remote Sensing of the Earth, the use of drones and image classification in remote sensing. The practical part contains mainly a description of entire works from image collection, image classification to surface measurement. Main result of the thesis are prepared maps outputs, evaluation of tree stupms detection success and of measuring their diameter and area from UAV outputs. The success rate in detection was high, exceeding 90% in all cases. While measuring areas of tree stumps and the images in the software, the difference was +/- 15%.