Pieter Kempeneers's research while affiliated with The Joint Commission and other places

Publications (105)

Preprint
Full-text available
Field boundaries are at the core of many agricultural applications and are a key enabler for operational monitoring of agricultural production to support food security. Recent scientific progress in deep learning methods has highlighted the capacity to extract field boundaries from satellite and aerial images with a clear improvement from object-ba...
Article
Full-text available
This paper is on the optimization of computing resources to process geospatial image data in a cloud computing infrastructure. Parallelization was tested by combining two different strategies: image tiling and multi-threading. The objective here was to get insight on the optimal use of available processing resources in order to minimize the process...
Article
Full-text available
Detailed parcel-level crop type mapping for the whole European Union (EU) is necessary for the evaluation of agricultural policies. The Copernicus program, and Sentinel-1 (S1) in particular, offers the opportunity to monitor agricultural land at a continental scale and in a timely manner. However, so far the potential of S1 has not been explored at...
Conference Paper
Full-text available
This work presents a novel parcel-based classification system that predicts the crop type of agricultural parcels from multispectral time series, leveraging on 1D convolutional neural network supervised learning capacity. The study area consists of the declared parcels in Catalonia, Spain, for the 2017-2018 cultivation cycle. Comparison results wit...
Preprint
Full-text available
Detailed parcel-level crop type mapping for the whole European Union (EU) is necessary for the evaluation of agricultural policies. The Copernicus program, and Sentinel-1 (S1) in particular, offers the opportunity to monitor agricultural land at a continental scale and in a timely manner. However, so far the potential of S1 has not been explored at...
Article
Full-text available
At present, accessing and processing Earth Observation (EO) data on different cloud platforms requires users to exercise distinct communication strategies as each backend platform is designed differently. The openEO API (Application Programming Interface) standardises EO-related contracts between local clients (R, Python, and JavaScript) and cloud...
Article
Full-text available
Large-scale land cover classification from satellite imagery is still a challenge due to the big volume of data to be processed, to persistent cloud-cover in cloud-prone areas as well as seasonal artefacts that affect spatial homogeneity. Sentinel-2 times series from Copernicus Earth Observation program offer a great potential for fine scale land c...
Article
Full-text available
A new Python package, pyjeo, that deals with the analysis of geospatial data has been created by the Joint Research Centre (JRC). Adopting the principles of open science, the JRC strives for transparency and reproducibility of results. In this view, it has been decided to release pyjeo as free and open software. This paper describes the design of p...
Conference Paper
Full-text available
The Copernicus Sentinel-2 mission offers new opportunities for mapping human settlements over large areas and for the update and improvement of the Global Human Settlement Layer. This paper presents the fully automated processing workflows tailored for large scale mapping of built-up areas from Sentinel-2 imagery. The first results provide insights...
Article
In recent years, several critical issues have been identified concerning the performance of recently established spruce forest plantations in Ireland, in particular coniferous reforestation sites. More specifically, the reforestation of peatland areas in the midlands of Ireland have been subject to encroachment by broadleaf species, such as birch,...
Article
Full-text available
Motivated by the increasing availability of open and free Earth observation data through the Copernicus Sentinel missions, this study investigates the capacity of advanced computational models to automatically generate thematic layers, which in turn contribute to and facilitate the creation of land cover products. In concrete terms, we assess the p...
Article
Full-text available
With the advent of Sentinel-2, it is now possible to generate large-scale chlorophyll content maps with unprecedented spatial and temporal resolution, suitable for monitoring ecological processes such as vegetative stress and/or decline. However methodological gaps exist for adapting this technology to heterogeneous natural vegetation and for trans...
Article
Full-text available
Processing large amounts of image data such as the Sentinel-2 archive is a computationally demanding task. However, for most applications, many of the images in the archive are redundant and do not contribute to the quality of the final result. An optimization scheme is presented here that selects a subset of the Sentinel-2 archive in order to redu...
Conference Paper
Full-text available
The JRC Earth Observation Data and Processing Platform (JEODPP) is a versatile petabyte-scale platform that serves the needs of a wide variety of projects. This is achieved by providing a cluster environment for batch processing, a webbased remote desktop access with a variety of software suites, and a web-based interactive visualisation and analys...
Presentation
Full-text available
With its open-source policy and accommodation of a wide variety of programming languages, the Jupyter web-application has recently positioned itself as the most popular environment for interactive scientific computing. In this paper, the use of Jupyter notebooks based on IPython for interactive visualisation and analysis of geospatial data is put f...
Conference Paper
Full-text available
With its open-source policy and accommodation of a wide variety of programming languages, the Jupyter web-application has recently positioned itself as the most popular environment for interactive scientific computing. In this paper, the use of Jupyter notebooks based on IPython for interactive visualisa-tion and analysis of geospatial data is put...
Article
Full-text available
The increasing amount of free and open geo-spatial data of interest to major societal questions calls for the development of innovative data-intensive computing platforms for the efficient and effective extraction of information from these data. This paper proposes a versatile petabyte-scale platform based on commodity hardware and equipped with op...
Article
Full-text available
Appropriate management of (semi-)natural areas requires detailed knowledge of the ecosystems present and their status. Remote sensing can provide a systematic, synoptic view at regular time intervals, and is therefore often suggested as a powerful tool to assist with the mapping and monitoring of protected habitats and vegetation. In this study, we...
Article
The project for on-board autonomy-vegetation (PROBA-V) satellite can produce global daily images at 300-m spatial resolution. Three sensors are mounted on the same platform. Two off-nadir-viewing sensors acquire imagery at 300-m spatial resolution, whereas a nadir-viewing sensor acquires imagery at 100-m spatial resolution. The swath of the nadir-v...
Chapter
This case study will demonstrate the creation of a vector based training dataset for a study area based in Houston, Texas. This task fits within the context of the IEEE GRSS Data Fusion Contest. The data created can be used as a training dataset for supervised image classification.
Chapter
This chapter covers one of the third party tool boxes selected in this book.
Chapter
This chapter will focus on geospatial vector data and presents examples of data processing using the OGR Simple Features Library that is part of the GDAL library. We provide clear examples of how the OGR interface can be used to retrieve metadata and summary information from vector data sets, but more importantly we demonstrate how OGR is used to e...
Chapter
This chapter covers the concept of pseudocolor images. We explain what it is and how you can generate such images via virtual files. In Sect. 14. 3. 6 of Chap. 14, we show how to create a program that adds a color table to an existing raster dataset.
Chapter
This chapter deals with manipulating raster data. Most important is gdal_translate, a utility to convert image and data formats, rescale and subset bands and spatial extents.
Chapter
Virtual rasters (VRT ) do not contain the actual pixel values of the raster cells. Instead, these virtual files in XML format describe the raster data by: name and location (path) of the real raster file, number of bands, lines and columns, projection information, etc.
Chapter
This concluding chapter summarizes the material covered in the book and comments on how the book has achieved its stated objectives. We also give some personal reflections on how we see geospatial analysis developing over the coming years in light of new Earth observation sensors and new open source technologies.
Chapter
In this chapter we will deal with three dimensional (3D) point cloud data.
Chapter
This book is designed as a reference and a tutorial to a number of open source geospatial command line utilities. It intends to demonstrate the power and flexibility of command line utilities to the novice user as well as serving as a reference to the more experienced user. We have designed the book in a way that enables the novice user to get star...
Chapter
This chapter deals with GDAL utilities that link raster to point data. With gdal_grid, point data can be converted to raster grids. The conversion of raster datasets to point data can be performed with gdallocationinfo and gdal2xyz.py. Other utilities discussed in this chapter are: gdal_fillnodata.py, gdal_proximity.py and gdaldem. We recall that t...
Chapter
GDAL API In spite of the variety of GDAL utilities that are available for processing geospatial raster data, it is likely that at some point you will need some functionality that is either not available or else does not perfectly match the problem at hand.
Chapter
Land cover classification is one of the principal applications of remote sensing. The goal is to distill satellite images into a spatially explicit information in the form of a categorical land cover map.
Chapter
This chapter introduces the GDAL utilities and explains some general principles on their usage via the command line. The GDAL utilities to process raster datasets are written in C/C++ and Python. In general, Python scripts have the extension .py (e.g. gdal_merge.py) whereas C/C++ executable programs come without an extension (e.g. gdalinfo).
Chapter
Image projections play a crucial role in geospatial data. As explained in Sect. 3. 1, map projections do convert positions on Earth, typically defined by an angular (latitude, longitude) pair of coordinates and a geodetic height to a (plane) projected coordinate system.
Chapter
The Orfeo Toolbox (OTB) consists of a C++ library as well as a set of command line utilities called applications relying on this library that have been specifically designed for processing and analysing large remote sensing datasets. It provides a range of image corrections algorithms, filtering and classification tools as well as data extraction u...
Chapter
In this chapter, we explain some basic terms on raster data. We will deal with coordinate systems, single and multi-band images and different encoding schemes to store raster data in files.
Chapter
As introduced in Sect. 1. 3, geospatial data can be represented by raster grids and vectors. Depending on the data content and operations to be performed on the data, one representation is preferred over the other. However, due to subsequent operations raster data can become more interesting when converted as vectors.
Article
Full-text available
Spectral consistency with SPOT-VEGETATION is an important mission objective for PROBA-V, in particular for its 1 km products. This must allow service providers such as the Copernicus Global Land Service to extend the 16-year long timeseries of SPOTVEGETATION global 1km data with similar PROBA-V products. To evaluate the extent of spectral consisten...
Book
Full-text available
This book focuses on the use of open source software for geospatial analysis. It demonstrates the effectiveness of the command line interface for handling both vector, raster and 3D geospatial data. Appropriate open-source tools for data processing are clearly explained and discusses how they can be used to solve everyday tasks. A series of fully w...
Article
Full-text available
A data assimilation method to produce complete temporal sequences of synthetic medium-resolution images is presented. The method implements a Kalman filter recursive algorithm that integrates medium and moderate resolution imagery. To demonstrate the approach, time series of 30-m spatial resolution NDVI images at 16-day time steps were generated us...
Chapter
The objective of this case study is to present a work flow for processing LiDAR data using a fully open source toolbox. We have selected SPDLib because it is: (1) fully open source with no limitations on its use or data size; (2) mature; and (3) provides considerable functionality. SPDLib also supports full waveform LiDAR datasets but this is not c...
Chapter
Image overviews, tiling and pyramids are techniques to view large images more efficiently.
Conference Paper
Due to technological constraints, hyperspectral earth observation imagery are often a mosaic of overlapping flight lines collected in different passes over the area of interest. This causes variations in aqcuisition conditions such that the reflected spectrum can vary significantly between these flight lines. Partly, this problem is solved by atmos...
Article
Full-text available
This study focused on the need of accurate digital surface models rather than existing digital terrain models for the geometric correction of high spatial resolution images over forests. Based on both theoretical and experimental results, it was shown here that even for close to nadir observations (view angles less than 7°), the geometric error inc...
Article
Full-text available
A pan-European forest cover map (FMAP2006) was produced using a novel automated classification approach using remotely sensed data from fine resolution satellite instruments. In contrast to previous classification accuracy assessments of such continental scale land cover products, the current study aimed for a reliable assessment at different geogr...
Article
Full-text available
A two stage burned scar detection approach is applied to produce a burned scar map for Mediterranean Europe using IRS-AWiFS imagery acquired at the end of the 2009 fire season. The first stage identified burned scar seeds based on a learning algorithm (Artificial Neural Network) coupled with a bootstrap aggregation process. The second stage impleme...
Article
Full-text available
A data fusion method for land cover (LC) classification is proposed that combines remote sensing data at a fine and a coarse spatial resolution. It is a two-step approach, based on the assumption that some of the LC classes can be merged into a more generalized LC class. Step one creates a generalized LC map, using only the information available at...
Article
This study presents a novel cloud masking approach for high resolution remote sensing images in the context of land cover mapping. As an advantage to traditional methods, the approach does not rely on thermal bands and it is applicable to images from most high resolution earth observation remote sensing sensors. The methodology couples pixel-based...
Article
Given the hyperspectral-oriented waveband configuration of multiangular CHRIS/Proba imagery, the scope of its application could widen if the present 18-m resolution would be improved. The multiangular images of CHRIS could be used as input for superresolution (SR) image reconstruction. A critical procedure in SR is an accurate registration of the l...
Article
Existing remote sensing-based Burnt scar mapping methodologies are mainly based on medium spatial resolution sensors. These methodologies cannot accurately detect burnt areas smaller than 200 ha. However, smaller burnt areas can represent a significant number of the total burnt areas. They play an important ecological roll, and also have relevant s...
Article
The Mediterranean region is highly susceptible to wildfires. On average, about 60,000 fires take place in this region every year, burning on average half a million hectares of forests and natural vegetation. Wildfires cause environmental degradation and affect the lives of thousands of people in the region. In order to minimize the consequences of...
Conference Paper
The Mediterranean basin region is highly susceptible to wildfire, with approximately 60,000 individual fires and half a million ha of natural vegetation burnt per year. Of particular concern in this region is the impact of repeated wildfires on the ability of natural lands to return to a pre-fire state, and of the possibility of desertification of...
Article
Full-text available
Kempeneers, P.; Deronde, B.; Provoost, S., and Houthuys, R., 2009. Synergy of airborne digital camera and lidar data to map coastal dune vegetation. Journal of Coastal Research, SI(53), xx-xx. Driven by the successful applications of lidar in forestry and the availability of lidar technology, new research is being carried out in other ecosystems. W...
Presentation
In compliance with the Habitats Directive of 1992, all European member states are obliged to report about the conservation status of a range of Natura 2000 protected habitat types to the European Union every 6 years. Up till now, the assessment of the status of these habitats is still being performed based solely on intensive, time-consuming and ex...
Article
Full-text available
This paper presents the results of estimation of leaf chlorophyll concentration through model inversion, from hyperspectral imagery of artificially treated orchard crops. The objectives were to examine model inversion robustness under changing viewing conditions, and the potential of multi-angle hyperspectral data to improve accuracy of chlorophyll...