Comparison of Machine Learning Algorithms for Wildland-Urban Interface Fuelbreak Planning Integrating ALS and UAV-Borne LiDAR Data and Multispectral Images

Abstract

Controlling vegetation fuels around human settlements is a crucial strategy for reducing fire severity in forests, buildings and infrastructure, as well as protecting human lives. Each country has its own regulations in this respect, but they all have in common that by reducing fuel load, we in turn reduce the intensity and severity of the fire. The use of Unmanned Aerial Vehicles (UAV)-acquired data combined with other passive and active remote sensing data has the greatest performance to planning Wildland-Urban Interface (WUI) fuelbreak through machine learning algorithms. Nine remote sensing data sources (active and passive) and four supervised classification algorithms (Random Forest, Linear and Radial Support Vector Machine and Artificial Neural Networks) were tested to classify five fuel-area types. We used very high-density Light Detection and Ranging (LiDAR) data acquired by UAV (154 returns·m−2 and ortho-mosaic of 5-cm pixel), multispectral data from the satellites Pleiades-1B and Sentinel-2, and low-density LiDAR data acquired by Airborne Laser Scanning (ALS) (0.5 returns·m−2, ortho-mosaic of 25 cm pixels). Through the Variable Selection Using Random Forest (VSURF) procedure, a pre-selection of final variables was carried out to train the model. The four algorithms were compared, and it was concluded that the differences among them in overall accuracy (OA) on training datasets were negligible. Although the highest accuracy in the training step was obtained in SVML (OA=94.46%) and in testing in ANN (OA=91.91%), Random Forest was considered to be the most reliable algorithm, since it produced more consistent predictions due to the smaller differences between training and testing performance. Using a combination of Sentinel-2 and the two LiDAR data (UAV and ALS), Random Forest obtained an OA of 90.66% in training and of 91.80% in testing datasets. The differences in accuracy between the data sources used are much greater than between algorithms. LiDAR growth metrics calculated using point clouds in different dates and multispectral information from different seasons of the year are the most important variables in the classification. Our results support the essential role of UAVs in fuelbreak planning and management and thus, in the prevention of forest fires.

Identifying Population Hollowing Out Regions and Their Dynamic Characteristics across Central China

“…To our knowledge, population hollowing was mainly affected by physical geographical factors, social and economic development, and disasters or unexpected events [79]. Firstly, from the location perspective, the six provinces of central China are adjacent to regions with relatively better economic status, such as Chongqing, and the eastern and southern coastal areas [80]. Our outcomes demonstrated that the population hollowing phenomena were mainly accumulated in central areas except for Shanxi and Henan provinces (Figure 5).…”

Methodological Proposal for Automated Detection of the Wildland–Urban Interface: Application to the Metropolitan Regions of Madrid and Barcelona

“…The use of LiDAR data acquired by unmanned aerial vehicles (UAVs), combined with other passive and active remote sensing data, has the greatest future for fuel mapping of the wildland-urban interface (WUI), using machine-learning algorithms [33]. It is also necessary to highlight the importance of short-range LiDAR for field data collection, together with the application of qualitative and quantitative mapping methods to visualize land use in a dynamic context since it is possible to record dynamic phenomena in space, thanks to images obtained cyclically by UAVs [34].…”

Mapping Tree Canopy in Urban Environments Using Point Clouds from Airborne Laser Scanning and Street Level Imagery

“…To reduce the training processing time, a variable selection was performed using the VSURF procedure [72]. The four most used algorithms in ML for this type of training were evaluated [73]; ANN, SVML, SVMR, and RF, executing the NNET, SVMLINEAR, SVMRADIAL, and RF methods using the CARET package in R software [74]. Finally, a cross-validation was performed using three replicates to control for overfitting.…”