National Repository of Grey Literature 2 records found  Search took 0.00 seconds. 
Possibilities of object-based classification for selected biotopes above tree-line in the Krkonoše Mts. National Park detection
Jakešová, Lucie ; Červená, Lucie (advisor) ; Potůčková, Markéta (referee)
Possibilities of object-based classification for selected biotopes above tree-line in the Krkonoše Mts. National Park detection Abstract The bachelor thesis is focused on the object-based classification of vegetation above the tree-line in the Krkonoše Mts. National Park using orthophoto with near infrared band and spatial resolution of 12.5 cm. Orthophoto was acquired in summer 2012. The classification legend was compiled by botanist of the national park. Software ENVI 5.1 was used for object-based classification using the field data. It provides two approaches to classification - Example-based and Rule-based. The overall accuracy of the best classification result was 75.97 % for 13 classes. Keywords: object-based classification, KRNAP, biotopes above tree-line, aerial optical scanner
Possibilities of object based image analysis for monitoring of meadow vegetation and management in the Krkonoše Mountains National Park
Dorič, Roman ; Kupková, Lucie (advisor) ; Potůčková, Markéta (referee)
Possibilities of object based image analysis for monitoring of meadow vegetation and management in the Krkonoše Mountains National Park Abstract The main aim of the thesis was to evaluate possibilities of Object Based Image Analysis (OBIA) of WorldView-2 satellite image data and aerial optical scanner for meadow vegetation and managment types classification in Krkonoše Mountains National Park. The classification was based on legend prepared by botanist of the national park. The second goal was to compare classification accuracy of Object Based Image Analysis and neural net classification method that was used by Pomahačová (2012) for the same area and the same WorldView-2 data. OBIA for meadow vegetation was conducted using SVM algorithm and "Decision Tree" algorithm. The classification accuracy was estimated using reference points from the field. The thesis puts the requirements (optimal parameters and conditions) for successfull object based classification of mountain meadow vegetation into a new perspective. Key words: Object based classification, meadows, WorldView-2, aerial optical scanner, SVM, KRNAP

Interested in being notified about new results for this query?
Subscribe to the RSS feed.