Classification of Tree Species in a Diverse African Agroforestry Landscape Using Imaging Spectroscopy and Laser Scanning

Abstract

Airborne imaging spectroscopy (IS) and laser scanning (ALS) have been explored widely for tree species classification during the past decades. However, African agroforestry areas, where a few exotic tree species are dominant and many native species occur less frequently, have not yet been studied. Obtaining maps of tree species would provide useful information for the characterization of agroforestry systems and detecting invasive species. Our objective was to study tree species classification in a diverse tropical landscape using IS and ALS data at the tree crown level, with primary interest in the exotic tree species. We performed multiple analyses based on different IS and ALS feature sets, identified important features using feature selection, and evaluated the impact of combining the two data sources. Given that a high number of tree species with limited sample size (499 samples for 31 species) was expected to limit the classification accuracy, we tested different approaches to group the species based on the frequency of their occurrence and Jeffries-Matusita (JM) distance. Surface reflectance at wavelengths between 400-450 nm and 750-800 nm, and height to crown width ratio, were identified as important features. Nonetheless, a selection of minimum noise fraction (MNF) transformed reflectance bands showed superior performance. Support vector machine classifier performed slightly better than the random forest classifier, but the improvement was not statistically significant for the best performing feature set. The highest F1-scores were achieved when each of the species was classified separately against a mixed group of all other species, which makes this approach suitable for invasive species detection. Our results are valuable for organizations working on biodiversity conservation and improving agroforestry practices, as we showed how the non-native Eucalyptus spp., Acacia mearnsii and Grevillea robusta (mean F1-scores 76%, 79% and 89%, respectively) trees can be mapped with good accuracy. We also found a group of six fruit bearing trees using JM distance, which was classified with mean F1-score of 65%. This was a useful finding, as these species could not be classified with acceptable accuracy individually, while they all share common economic and ecological importance.

Detecting European Aspen (Populus tremula L.) in Boreal Forests Using Airborne Hyperspectral and Airborne Laser Scanning Data

“…We executed tree species classification both with and without feature selection by applying two machine learning methods from R-package caret [55]: the support vector machine (SVM) [56] and the random forest (RF) [57]. They both are non-parametric classification/regression models that are widely used in image classification applying remote sensing data [43,[58][59][60]. We implemented SVM using the linear kernel method that separates different classes by maximizing the minimum margin to the closest support vector using a linear hyperplane.…”

“…A variety of different machine learning algorithms have been applied to hyperspectral imagery for tree species classification. In addition to the classification method used, the classification accuracies achieved have a wide variance depending on the spatial and spectral resolution of the data, forest conditions and types, species distribution, and stand structure [36,[39][40][41][42][43]. In boreal forest, the use of hyperspectral applications, with the most commonly used machine learning classifiers, support vector machine (SVM) and random forest (RF), has showed high potential for tree species classification [17].…”

Estimation of Above Ground Biomass in a Tropical Mountain Forest in Southern Ecuador Using Airborne LiDAR Data

“…An accurate estimation of AGB requires a tree level classification [23,38,39], which can be facilitated by modern instruments, like the Light Detection and Ranging (LiDAR) sensor [40,41]. From the three-dimensional LiDAR data, the horizontal and vertical forest structure can be derived as well as the dominant trees identified [42,43]. LiDAR data permits the generation of high resolution Digital Terrain Models (DTM) and high resolution Digital Surface Models (DSM), from which Canopy Height Models (CHM) can be obtained.…”

Integrating National Ecological Observatory Network (NEON) Airborne Remote Sensing and In-Situ Data for Optimal Tree Species Classification

“…Three commonly used classification approaches are non-parametric machine learning techniques including Support Vector Machines (SVM), Random Forest (RF), and neural networks [6,[26][27][28]. SVM and RF tend to perform similarly in terms of classification accuracy and training time [29][30][31]. Neural networks are increasingly used in ecological remote sensing studies for their ability to identify trends and patterns from data, model complex relationships, accept a wide variety of input predictor data, and produce high accuracies, at the expense of requiring large amounts of training data [13,[32][33][34].…”