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

被引:14
作者
Rodriguez-Puerta, Francisco [1 ,2 ,3 ]
Alonso Ponce, Rafael [2 ,3 ]
Perez-Rodriguez, Fernando [2 ]
Agueda, Beatriz [1 ,2 ,3 ]
Martin-Garcia, Saray [2 ,4 ]
Martinez-Rodrigo, Raquel [2 ,3 ]
Lizarralde, Inigo [2 ,3 ]
机构
[1] Univ Valladolid, EiFAB, Campus Duques Soria S-N, Soria 42004, Spain
[2] Fora Forest Technol Sll, Campus Duques Soria S-N, Soria 42004, Spain
[3] Univ Valladolid INIA, Sustainable Forest Management Res Inst, Campus Duques Soria S-N, Soria 42004, Spain
[4] Univ Santiago Compostela, Dept Enxenaria Agroforestal, Biodivers LaboraTe IBADER, Lugo 27001, Spain
基金
欧盟地平线“2020”;
关键词
artificial intelligence; UAV-LiDAR; satellite imagery; large-scale LiDAR; RANDOM FOREST; AIRBORNE LIDAR; CLASSIFICATION; CLASSIFIERS; VEGETATION; SELECTION; BIOMASS; FUSION;
D O I
10.3390/drones4020021
中图分类号
TP7 [遥感技术];
学科分类号
081102 ; 0816 ; 081602 ; 083002 ; 1404 ;
摘要
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.
引用
收藏
页码:1 / 18
页数:18
相关论文
共 55 条
[1]   Estimating standing biomass in papyrus (Cyperus papyrus L.) swamp: exploratory of in situ hyperspectral indices and random forest regression [J].
Adam, Elhadi ;
Mutanga, Onisimo ;
Abdel-Rahman, Elfatih M. ;
Ismail, Riyad .
INTERNATIONAL JOURNAL OF REMOTE SENSING, 2014, 35 (02) :693-714
[2]   Review of Machine Learning Approaches for Biomass and Soil Moisture Retrievals from Remote Sensing Data [J].
Ali, Iftikhar ;
Greifeneder, Felix ;
Stamenkovic, Jelena ;
Neumann, Maxim ;
Notarnicola, Claudia .
REMOTE SENSING, 2015, 7 (12) :16398-16421
[3]   Potential of Sentinel-2A Data to Model Surface and Canopy Fuel Characteristics in Relation to Crown Fire Hazard [J].
Arellano-Perez, Stefano ;
Castedo-Dorado, Fernando ;
Antonio Lopez-Sanchez, Carlos ;
Gonzalez-Ferreiro, Eduardo ;
Yang, Zhiqiang ;
Alberto Diaz-Varela, Ramon ;
Gabriel Alvarez-Gonzalez, Juan ;
Antonio Vega, Jose ;
Daria Ruiz-Gonzalez, Ana .
REMOTE SENSING, 2018, 10 (10)
[4]  
Ascoli D., 2018, Encyclopedia of Wildfires and Wildland-Urban Interface (WUI) Fires, P1, DOI DOI 10.1007/978-3-319-51727-8_70-1
[5]  
Baatz M., 2000, ANGEW GEOGRAPHISCHE, V12, P12, DOI DOI 10.3390/RS5010183
[6]   Multi-resolution, object-oriented fuzzy analysis of remote sensing data for GIS-ready information [J].
Benz, UC ;
Hofmann, P ;
Willhauck, G ;
Lingenfelder, I ;
Heynen, M .
ISPRS JOURNAL OF PHOTOGRAMMETRY AND REMOTE SENSING, 2004, 58 (3-4) :239-258
[7]   Modelling Forest Ecosystems: a crossroad between scales, techniques and applications [J].
Blanco, Juan A. ;
Ameztegui, Aitor ;
Rodriguez, Francisco .
ECOLOGICAL MODELLING, 2020, 425
[8]   Geographic Object-Based Image Analysis - Towards a new paradigm [J].
Blaschke, Thomas ;
Hay, Geoffrey J. ;
Kelly, Maggi ;
Lang, Stefan ;
Hofmann, Peter ;
Addink, Elisabeth ;
Feitosa, Raul Queiroz ;
van der Meer, Freek ;
van der Werff, Harald ;
van Coillie, Frieke ;
Tiede, Dirk .
ISPRS JOURNAL OF PHOTOGRAMMETRY AND REMOTE SENSING, 2014, 87 :180-191
[9]   Integrating LIDAR data and multispectral imagery for enhanced classification of rangeland vegetation: A meta analysis [J].
Bork, Edward W. ;
Su, Jason G. .
REMOTE SENSING OF ENVIRONMENT, 2007, 111 (01) :11-24
[10]  
Breiman L., 2001, Mach. Learn., V45, P5