Abstract
This study presents a new methodology for assessment of the spatial and qualitative aspects of forest disturbance at high spatial and spectral resolution based on coupling the use of multispectral imaging sensor with the UAV photogrammetry. We have used the multirotor imaging platform MikroKopter OctoXL coupled with the miniaturized multispectral sensor Tetracam MicroMCA 6, delivering six-band multispectral imagery with high spatial resolution.
The image enhancement phase included computed selected vegetation indices as support to original spectral bands for image classification to detect forest disturbance status on the stand.We have identified three indices, suitable for a distinct identification of the different forest disturbance categories in the conditions of a highly heterogeneous mixture of forest status categories - the Normalized Difference Vegetation Index, Simple Ratio 800/650 Pigment specific simple ratio B1 and Red-edge Index. The advanced classification techniques (Support Vector Machines, Random Forest) were used for classification of the spatial distribution of the detected qualitative categories of the forest quality status.
The results proved that the miniaturized multispectral sensors coupled with UAV platforms enable much more detailed insight into the mechanism and dynamics of the forest disturbance, compared to the conventional remote sensing data products with high potential for applications in nature conservation and forest management.