A Comparative Analysis of Phytovolume Estimation Methods Based on UAV-Photogrammetry and Multispectral Imagery in a Mediterranean Forest

Abstract

Management and control operations are crucial for preventing forest fires, especially in Mediterranean forest areas with dry climatic periods. One of them is prescribed fires, in which the biomass fuel present in the controlled plot area must be accurately estimated. The most used methods for estimating biomass are time-consuming and demand too much manpower. Unmanned aerial vehicles (UAVs) carrying multispectral sensors can be used to carry out accurate indirect measurements of terrain and vegetation morphology and their radiometric characteristics. Based on the UAV-photogrammetric project products, four estimators of phytovolume were compared in a Mediterranean forest area, all obtained using the difference between a digital surface model (DSM) and a digital terrain model (DTM). The DSM was derived from a UAV-photogrammetric project based on the structure from a motion algorithm. Four different methods for obtaining a DTM were used based on an unclassified dense point cloud produced through a UAV-photogrammetric project (FFU), an unsupervised classified dense point cloud (FFC), a multispectral vegetation index (FMI), and a cloth simulation filter (FCS). Qualitative and quantitative comparisons determined the ability of the phytovolume estimators for vegetation detection and occupied volume. The results show that there are no significant differences in surface vegetation detection between all the pairwise possible comparisons of the four estimators at a 95% confidence level, but FMI presented the best kappa value (0.678) in an error matrix analysis with reference data obtained from photointerpretation and supervised classification. Concerning the accuracy of phytovolume estimation, only FFU and FFC presented differences higher than two standard deviations in a pairwise comparison, and FMI presented the best RMSE (12.3 m) when the estimators were compared to 768 observed data points grouped in four 500 m2 sample plots. The FMI was the best phytovolume estimator of the four compared for low vegetation height in a Mediterranean forest. The use of FMI based on UAV data provides accurate phytovolume estimations that can be applied on several environment management activities, including wildfire prevention. Multitemporal phytovolume estimations based on FMI could help to model the forest resources evolution in a very realistic way.

The Potential of Multispectral Imagery and 3D Point Clouds from Unoccupied Aerial Systems (UAS) for Monitoring Forest Structure and the Impacts of Wildfire in Mediterranean-Climate Forests

“…This study employed the Cloth Simulation Filter (CSF) from Zhang et al [44] in conjunction with a subsequent NDVI filter to classify ground points from which a DTM could be generated through interpolation. In Mediterranean shrublands, Carvajal-Ramirez et al [53] found the CSF performance was weaker relative to an NDVI classifier, and the authors called for the need to assess the impact of CSF parameter setting on ground-finding accuracy. In this study, CSF parameter optimization alone only marginally improved RMSE over the default parameters.…”

“…However, the addition of our NDVI threshold filter resulted in a large (>60%) improvement in RMSE. Drawing on the concept of the NDVI classifier from Carvajal-Ramirez et al [53], the combination of CSF with an NDVI filter leverages the strengths of both methods. The CSF identifies an initial set of ground points from which the NDVI filter removes misclassified vegetation.…”

Vegetation Cover Estimation in Semi-Arid Shrublands after Prescribed Burning: Field-Ground and Drone Image Comparison

“…The flights followed a simple grid. To address the potential issues of scale heterogeneity due to lowaltitude flights and steep terrain [35], and to obtain the most accurate information possible, the flight routes were planned using UgCS PRO software [36]. The flights were performed at an altitude of 30 m in the central hours of the day, when the sun's rays strike nearly perpendicular to the terrain, minimizing shadows that distort the spectral information.…”

Large-Scale Reality Modeling of a University Campus Using Combined UAV and Terrestrial Photogrammetry for Historical Preservation and Practical Use

“…Examples of unsupervised machine learning with UAV photogrammetry include analyzing photos of plants (more on object recognition after explaining NBV), agricultural models, and forestry management [63][64][65]. NBV appears in recent publications such as Bolourian and Hammad [66], Ashour et al [67], and Almadhoun et al [68].…”