Background:Increasing threat to Central Europe’s forests from the growing population of the European spruce bark beetle Ips typographus (L.) calls for developing highly effective methods of detection of the infestati...Background:Increasing threat to Central Europe’s forests from the growing population of the European spruce bark beetle Ips typographus (L.) calls for developing highly effective methods of detection of the infestation spots. The main goal of this study was to establish an automatic workflow for detection of dead trees and trees in poor condition of Picea abies using Middle Wave Infrared spectral range obtained from the aircraft.Methods:The studies were conducted in Wigry National Park (Poland) in 2020. A fusion of aircraft thermal data and laser scanning was used. Synchronous with thermal data acquisition ground reference data were obtained for P. abies in different health conditions. Determination of the range of canopy temperatures characteristic of the three condition states (‘healthy’,‘poor condition’,‘dead’) was performed using K-mean clustering. The accuracy of the method was evaluated on two validation sets:(1) individual tree canopies determined by photointerpretation, and (2) automatic segmentation of laser scanning data.Results:The results showed that the average temperature of ‘healthy’trees was 27.70℃, trees in ‘poor condition’28.57℃, and ‘dead’trees 30.17℃. High temperature differences between ‘healthy’and ‘dead’P. abies made it possible to distinguish these two condition classes with high accuracy. Lower accuracies were obtained for the class of ‘poor condition’, which was found to be confusing with both ‘healthy’and ‘dead’trees. According to results from the first validation set, a high overall accuracy of 0.60 was obtained. For the second validation set, the overall accuracy was reduced by 11%.Conclusions:This study indicates that canopy temperature recorded from the airborne level is a variable that differentiates ‘healthy’spruces from those in ‘poor condition’and the ‘dead’trees. The results confirmed that thermal and airborne laser scanning data fusion allows for creating a quick and simple workflow, which can successfully separate individual tree canopies and identify P. abies attacked by I. typographus. Further research is needed to identify trees in the early stages of invasion.展开更多
基金co-financed by the European Union from the European Social Fund under the "InterDOC-STARt" project (POWR.03.02.00-00-I033/16-00) and from the Operational Programme Infrastructure and Environment under the program 2.4.4d-assessment of the state of natural resources in national parks using modern remote sensing technologies"Acquisition of multi-source remote sensing data and their analysis for the area of Wigry National Park with a part of Wigry lake and the Czarna Hańcza river" project。
文摘Background:Increasing threat to Central Europe’s forests from the growing population of the European spruce bark beetle Ips typographus (L.) calls for developing highly effective methods of detection of the infestation spots. The main goal of this study was to establish an automatic workflow for detection of dead trees and trees in poor condition of Picea abies using Middle Wave Infrared spectral range obtained from the aircraft.Methods:The studies were conducted in Wigry National Park (Poland) in 2020. A fusion of aircraft thermal data and laser scanning was used. Synchronous with thermal data acquisition ground reference data were obtained for P. abies in different health conditions. Determination of the range of canopy temperatures characteristic of the three condition states (‘healthy’,‘poor condition’,‘dead’) was performed using K-mean clustering. The accuracy of the method was evaluated on two validation sets:(1) individual tree canopies determined by photointerpretation, and (2) automatic segmentation of laser scanning data.Results:The results showed that the average temperature of ‘healthy’trees was 27.70℃, trees in ‘poor condition’28.57℃, and ‘dead’trees 30.17℃. High temperature differences between ‘healthy’and ‘dead’P. abies made it possible to distinguish these two condition classes with high accuracy. Lower accuracies were obtained for the class of ‘poor condition’, which was found to be confusing with both ‘healthy’and ‘dead’trees. According to results from the first validation set, a high overall accuracy of 0.60 was obtained. For the second validation set, the overall accuracy was reduced by 11%.Conclusions:This study indicates that canopy temperature recorded from the airborne level is a variable that differentiates ‘healthy’spruces from those in ‘poor condition’and the ‘dead’trees. The results confirmed that thermal and airborne laser scanning data fusion allows for creating a quick and simple workflow, which can successfully separate individual tree canopies and identify P. abies attacked by I. typographus. Further research is needed to identify trees in the early stages of invasion.
