GEOBIA at the Terapixel Scale: Toward Efficient Mapping of Small Woody Features from Heterogeneous VHR Scenes

François Merciol,Christophe Sannier,Loïc Faucqueur,Baudouin Desclée,Antoine Masse,Pierre-Yves Rémy,Bharath Damodaran,Fabrice Dazin,Sébastien Lefèvre

doi:10.3390/ijgi8010046

Abstract

Land cover mapping has benefited a lot from the introduction of the Geographic Object-Based Image Analysis (GEOBIA) paradigm, that allowed to move from a pixelwise analysis to a processing of elements with richer semantic content, namely objects or regions. However, this paradigm requires to define an appropriate scale, that can be challenging in a large-area study where a wide range of landscapes can be observed. We propose here to conduct the multiscale analysis based on hierarchical representations, from which features known as differential attribute profiles are derived over each single pixel. Efficient and scalable algorithms for construction and analysis of such representations, together with an optimized usage of the random forest classifier, provide us with a semi-supervised framework in which a user can drive mapping of elements such as Small Woody Features at a very large area. Indeed, the proposed open-source methodology has been successfully used to derive a part of the High Resolution Layers (HRL) product of the Copernicus Land Monitoring service, thus showing how the GEOBIA framework can be used in a big data scenario made of more than 38,000 Very High Resolution (VHR) satellite images representing more than 120 TB of data.

Highlights

While the Geographic Object-Based Image Analysis (GEOBIA) paradigm has led to significant improvements in the analysis and understanding of high resolution remote sensing images thanks to the processing of objects instead of pixels [1], it still requires to identify the objects before applying sets of rules for classifying the extracted objects
The proposed scalable solution fully relies on open source components (Orfeo ToolBox [8], Boost [9], GDAL [10], Shark [11], Triskele OTB remote module) and so can be used in any GEOBIA application
To make the GEOBIA paradigm compliant with very large-area analysis, we propose here to perform a pixelwise analysis of object-based features, in a semi-supervised classification framework instead of the standard application of GEOBIA rulesets over pre-extracted objects

Summary

Introduction

While the GEOBIA paradigm has led to significant improvements in the analysis and understanding of high resolution remote sensing images thanks to the processing of objects (i.e., regions) instead of pixels [1], it still requires to identify the objects (or segment the image into regions) before applying sets of rules for classifying the extracted objects This segmentation step is not straightforward and often relies on user expertise or empirical tuning to be adapted to each new scene to be processed, even if some automated approaches exists [2,3]. It cannot be used for Big GeoData where large-area analyses require methods that are both very efficient and robust (applicable on different contexts) to the wide variety of scenes to be observed. The proposed scalable solution fully relies on open source components (Orfeo ToolBox (https://www.orfeo-toolbox.org) [8], Boost (https://www.boost.org) [9], GDAL (https://www.gdal.org) [10], Shark (http://image.diku. dk/shark/) [11], Triskele (https://sourcesup.renater.fr/triskele) OTB remote module) and so can be used in any GEOBIA application

Methods

Results

Discussion

Conclusion