Mangrove forest classification and aboveground biomass estimation using an atom search algorithm and adaptive neuro-fuzzy inference system.

Minh Hai Pham,Van-Manh Pham,Thi Hoai Do,Quang-Thanh Bui,Francisco Martínez-Álvarez

doi:10.1371/journal.pone.0233110

Minh Hai Pham, Van-Manh Pham + Show 3 more

Open Access

https://doi.org/10.1371/journal.pone.0233110

Copy DOI

Abstract

BackgroundAdvances in earth observation and machine learning techniques have created new options for forest monitoring, primarily because of the various possibilities that they provide for classifying forest cover and estimating aboveground biomass (AGB).MethodsThis study aimed to introduce a novel model that incorporates the atom search algorithm (ASO) and adaptive neuro-fuzzy inference system (ANFIS) into mangrove forest classification and AGB estimation. The Ca Mau coastal area was selected as a case study since it has been considered the most preserved mangrove forest area in Vietnam and is being investigated for the impacts of land-use change on forest quality. The model was trained and validated with a set of Sentinel-1A imagery with VH and VV polarizations, and multispectral information from the SPOT image. In addition, feature selection was also carried out to choose the optimal combination of predictor variables. The model performance was benchmarked against conventional methods, such as support vector regression, multilayer perceptron, random subspace, and random forest, by using statistical indicators, namely, root mean square error (RMSE), mean absolute error (MAE), and coefficient of determination (R2).ResultsThe results showed that all three indicators of the proposed model were statistically better than those from the benchmarked methods. Specifically, the hybrid model ended up at RMSE = 70.882, MAE = 55.458, R2 = 0.577 for AGB estimation.ConclusionFrom the experiments, such hybrid integration can be recommended for use as an alternative solution for biomass estimation. In a broader context, the fast growth of metaheuristic search algorithms has created new scientifically sound solutions for better analysis of forest cover.

Highlights

Biomass, which includes above- and belowground biomass, is a critical component of carbon budget accounting and carbon monitoring, especially under the context of climate change [1,2]
The model performance was benchmarked against conventional methods, such as support vector regression, multilayer perceptron, random subspace, and random forest, by using statistical indicators, namely, root mean square error (RMSE), mean absolute error (MAE), and coefficient of determination (R2)
The ASO significantly improved the performance of the adaptive neuro-fuzzy inference system (ANFIS) regression through comparison with benchmarked functions by using common statistical indicators with different combinations of features

Summary

Introduction

Biomass, which includes above- and belowground biomass, is a critical component of carbon budget accounting and carbon monitoring, especially under the context of climate change [1,2]. The destructive methods require tree cutting and further indoor weighing procedures [4] These methods are limited to smaller areas and are usually employed to measure the biomass of sample plots that can be used as ground truth samples. A clustering method was used to define the optimal clusters for this training dataset These data were fed into Layer 2 with the following membership function: mCjiðxiÞ 1⁄4 1 þ j x1iÀ cij j2bij ð1Þ aij where i = 1:m, in which m is the number of input variables, and j = 1: k, where k is the number of clusters as well as the number of rules in this study.

Objectives

Methods

Conclusion