Sentinel-2 Image Scene Classification: A Comparison between Sen2Cor and a Machine Learning Approach

Kashyap Raiyani,Luís Rato,Teresa Gonçalves,Pedro Salgueiro,José R Marques Da Silva

doi:10.3390/rs13020300

Abstract

Given the continuous increase in the global population, the food manufacturers are advocated to either intensify the use of cropland or expand the farmland, making land cover and land usage dynamics mapping vital in the area of remote sensing. In this regard, identifying and classifying a high-resolution satellite imagery scene is a prime challenge. Several approaches have been proposed either by using static rule-based thresholds (with limitation of diversity) or neural network (with data-dependent limitations). This paper adopts the inductive approach to learning from surface reflectances. A manually labeled Sentinel-2 dataset was used to build a Machine Learning (ML) model for scene classification, distinguishing six classes (Water, Shadow, Cirrus, Cloud, Snow, and Other). This models was accessed and further compared to the European Space Agency (ESA) Sen2Cor package. The proposed ML model presents a Micro-F1 value of 0.84, a considerable improvement when compared to the Sen2Cor corresponding performance of 0.59. Focusing on the problem of optical satellite image scene classification, the main research contributions of this paper are: (a) an extended manually labeled Sentinel-2 database adding surface reflectance values to an existing dataset; (b) an ensemble-based and a Neural-Network-based ML models; (c) an evaluation of model sensitivity, biasness, and diverse ability in classifying multiple classes over different geographic Sentinel-2 imagery, and finally, (d) the benchmarking of the ML approach against the Sen2Cor package.

Highlights

In remote sensing, classifying parts of the high-resolution optical satellite images into morphological categories is known as scene classification [1].Recently, the challenge of optical satellite image scene classification has been the focal point of many researchers
This study evaluates the performance of the developed machine learning models in classifying water, shadow, cirrus, cloud, snow, and other scenes over Sentinel-2 imagery
We evaluated ensemble methods (Random Forest & Extra Tree) and a deep learning based method (Convolutional Neural Network) using the built extended dataset for satellite imagery scene classification

Summary

Introduction

In remote sensing, classifying parts of the high-resolution optical satellite images into morphological categories (e.g., land, water, cloud, etc.) is known as scene classification [1]. The challenge of optical satellite image scene classification has been the focal point of many researchers. Earth observation can be defined as gathering physical, chemical, and biological information of the planet using Earth surveying techniques, which encompasses the collection of data [7]. In such Earth surveying techniques, optical satellites play a major role, and one such satellite is Sentinel-2 [8]. Sentinel-2 is part of the Earth observation mission from the Copernicus Programme, and systematically acquires optical imagery at a high spatial

Objectives

Methods

Results

Conclusion