Comparison of Support Vector Machines and Random Forests for Corine Land Cover Mapping

Anca Dabija,Anna Tardà,Edwin Raczko,Jordi Corbera,Marcin Kluczek,Ahmed H Al-Sulttani,Bogdan Zagajewski,Lydia Pineda,Marlena Kycko

doi:10.3390/rs13040777

Abstract

Land cover information is essential in European Union spatial management, particularly that of invasive species, natural habitats, urbanization, and deforestation; therefore, the need for accurate and objective data and tools is critical. For this purpose, the European Union’s flagship program, the Corine Land Cover (CLC), was created. Intensive works are currently being carried out to prepare a new version of CLC+ by 2024. The geographical, climatic, and economic diversity of the European Union raises the challenge to verify various test areas’ methods and algorithms. Based on the Corine program’s precise guidelines, Sentinel-2 and Landsat 8 satellite images were tested to assess classification accuracy and regional and spatial development in three varied areas of Catalonia, Poland, and Romania. The method is dependent on two machine learning algorithms, Random Forest (RF) and Support Vector Machine (SVM). The bias of classifications was reduced using an iterative of randomized training, test, and verification pixels. The ease of the implementation of the used algorithms makes reproducing the results possible and comparable. The results show that an SVM with a radial kernel is the best classifier, followed by RF. The high accuracy classes that can be updated and classes that should be redefined are specified. The methodology’s potential can be used by developers of CLC+ products as a guideline for algorithms, sensors, and the possibilities and difficulties of classifying different CLC classes.

Highlights

Populated metropolitan with surrounding suburban areas characterized by very heterogeneous land cover, and fragments of large-scale farming and protected areas on five research areas were selected, each covering an area of 50 × 50 km, located in Romania. Braila is a meBraila (Romania), Catalonia (Tarragona), and in Poland
This enabled diverse land cover forms to test the impact of geographic variability as well as local variations resulting from the Corine Land Cover national teams
We focused on conducting analyzes on individual Sentinel-2 imaging granules, selecting three countries characterized by a different history, which influenced spatial management (starting from large-scale agricultural crops (South Braila), a large urban agglomeration (Warsaw) with direct operating facilities, including the suburban zone, which on the one hand provides protected areas and climatic resorts along Vistula river, as well as typical suburban zones, including small farms providing food for suburban towns), and coastal and river port areas

Summary

Introduction

Human activities have changed the environment for thousands of years. The pressure on ecosystems, natural habitats, and biodiversity loss are among the most intense impacts of climate change over territories. These impacts translate into changes in land cover; an update and quality thematic mapping become a key indicator for better analysis and decision making to define potential climate actions. Land cover, combining components of the environment and human activity, is one of the most important biophysical features of the Earth’s surface [1]. Current land cover maps are an essential element of environment management and monitoring [2].

Methods

Results

Discussion

Conclusion