eCite Digital Repository

A validated ensemble method for multinomial land-cover classification

Citation

Diengdoh, VL and Ondei, S and Hunt, M and Brook, BW, A validated ensemble method for multinomial land-cover classification, Ecological Informatics, 56 Article 101065. ISSN 1574-9541 (2020) [Refereed Article]


Preview
PDF
Pending copyright assessment - Request a copy
7Mb

Preview
PDF
Pending copyright assessment - Request a copy
876Kb

DOI: doi:10.1016/j.ecoinf.2020.101065

Abstract

Land-cover data provides valuable information for landscape management and can be generated using machine learning algorithms. Ensemble models or model averaging can overcome difficulties in selecting an adequate algorithm and improve model predictions, but its use is limited among ecologists. The objective of this study is to highlight the benefits and limitations of weighted and unweighted majority voting ensemble models for land-cover classification and to enable easy and wider implementation of the method by providing an R-script (for use in the R software). Using a case study of three mixed-use landscapes from southern Australia (Tasmania), land cover was classified into six classes using Landsat 8 imagery and ancillary data, and support vector machine, random forest, k-nearest neighbour and na´ve Bayesian as base algorithms. The predicted classifications of the base algorithms were then averaged using both an unweighted and weighted (using the true skill statistic) majority voting ensemble algorithm. Cross-validation results showed the base algorithms achieved similar accuracy making algorithm selection difficult. The base algorithms achieved high and similar predictive accuracy when the classified land-cover and training data belong to the same geographic region but lower and different predictive accuracy when the classified land-cover and training data belong to different geographic regions. The weighted and unweighted ensemble achieved similar overall accuracy, equivalent to the best performing base algorithm. We conclude that the majority voting ensemble can be adopted to overcome difficulties in model selection during land-cover classification.

Item Details

Item Type:Refereed Article
Keywords:land-cover classification, support vector machine, random forest, K-nearest neighbour, Naive Bayesian, majority voting ensemble classification
Research Division:Information and Computing Sciences
Research Group:Artificial Intelligence and Image Processing
Research Field:Image Processing
Objective Division:Environment
Objective Group:Ecosystem Assessment and Management
Objective Field:Ecosystem Assessment and Management at Regional or Larger Scales
UTAS Author:Diengdoh, VL (Mr Vishesh Diengdoh)
UTAS Author:Ondei, S (Dr Stefania Ondei)
UTAS Author:Hunt, M (Professor Mark Hunt)
UTAS Author:Brook, BW (Professor Barry Brook)
ID Code:137386
Year Published:2020
Deposited By:Plant Science
Deposited On:2020-02-11
Last Modified:2020-02-17
Downloads:0

Repository Staff Only: item control page