The study presents the preliminary results of two classification exercises assessing the capabilities of pre-operational (August 2015) Sentinel-2 (S2) data for mapping crop types and tree species. In the first case study, an S2 image was used to map six summer crop species in Lower Austria as well as winter crops/bare soil. Crop type maps are needed to account for crop-specific water use and for agricultural statistics. Crop type information is also useful to parametrize crop growth models for yield estimation, as well as for the retrieval of vegetation biophysical variables using radiative transfer models. The second case study aimed to map seven different deciduous and coniferous tree species in Germany. Detailed information about tree species distribution is important for forest management and to assess potential impacts of climate change. In our S2 data assessment, crop and tree species maps were produced at 10 m spatial resolution by combining the ten S2 spectral channels with 10 and 20 m pixel size. A supervised Random Forest classifier (RF) was deployed and trained with appropriate ground truth. In both case studies, S2 data confirmed its expected capabilities to produce reliable land cover maps. Cross-validated overall accuracies ranged between 65% (tree species) and 76% (crop types). The study confirmed the high value of the red-edge and shortwave infrared (SWIR) bands for vegetation mapping. Also, the blue band was important in both study sites. The S2-bands in the near infrared were amongst the least important channels. The object based image analysis (OBIA) and the classical pixel-based classification achieved comparable results, mainly for the cropland. As only single date acquisitions were available for this study, the full potential of S2 data could not be assessed. In the future, the two twin S2 satellites will offer global coverage every five days and therefore permit to concurrently exploit unprecedented spectral and temporal information with high spatial resolution.

First Experience with Sentinel-2 Data for Crop and Tree Species Classifications in Central Europe

https://hal.archives-ouvertes.fr/hal-03272371/document

Understanding the temporal behavior of crops using Sentinel-1 and Sentinel-2-like data for agricultural applications

Assessing the robustness of Random Forests to map land cover with high resolution satellite image time series over large areas

A detailed and accurate knowledge of land cover is crucial for many scientific and operational applications, and as such, it has been identified as an Essential Climate Variable. This accurate knowledge needs frequent updates. This paper presents a methodology for the fully automatic production of land cover maps at country scale using high resolution optical image time series which is based on supervised classification and uses existing databases as reference data for training and validation. The originality of the approach resides in the use of all available image data, a simple pre-processing step leading to a homogeneous set of acquisition dates over the whole area and the use of a supervised classifier which is robust to errors in the reference data. The produced maps have a kappa coefficient of 0.86 with 17 land cover classes. The processing is efficient, allowing a fast delivery of the maps after the acquisition of the image data, does not need expensive field surveys for model calibration and validation, nor human operators for decision making, and uses open and freely available imagery. The land cover maps are provided with a confidence map which gives information at the pixel level about the expected quality of the result.

Operational High Resolution Land Cover Map Production at the Country Scale Using Satellite Image Time Series

Latest remote sensing sensors are capable of acquiring high spatial and spectral Satellite Image Time Series (SITS) of the world. These image series are a key component of classification systems that aim at obtaining up-to-date and accurate land cover maps of the Earth’s surfaces. More specifically, current SITS combine high temporal, spectral and spatial resolutions, which makes it possible to closely monitor vegetation dynamics. Although traditional classification algorithms, such as Random Forest (RF), have been successfully applied to create land cover maps from SITS, these algorithms do not make the most of the temporal domain. This paper proposes a comprehensive study of Temporal Convolutional Neural Networks (TempCNNs), a deep learning approach which applies convolutions in the temporal dimension in order to automatically learn temporal (and spectral) features. The goal of this paper is to quantitatively and qualitatively evaluate the contribution of TempCNNs for SITS classification, as compared to RF and Recurrent Neural Networks (RNNs) —a standard deep learning approach that is particularly suited to temporal data. We carry out experiments on Formosat-2 scene with 46 images and one million labelled time series. The experimental results show that TempCNNs are more accurate than the current state of the art for SITS classification. We provide some general guidelines on the network architecture, common regularization mechanisms, and hyper-parameter values such as batch size; we also draw out some differences with standard results in computer vision (e.g., about pooling layers). Finally, we assess the visual quality of the land cover maps produced by TempCNNs.

Temporal Convolutional Neural Network for the Classification of Satellite Image Time Series

Crop area extent estimates and crop type maps provide crucial information for agricultural monitoring and management. Remote sensing imagery in general and, more specifically, high temporal and high spatial resolution data as the ones which will be available with upcoming systems, such as Sentinel-2, constitute a major asset for this kind of application. The goal of this paper is to assess to what extent state-of-the-art supervised classification methods can be applied to high resolution multi-temporal optical imagery to produce accurate crop type maps at the global scale. Five concurrent strategies for automatic crop type map production have been selected and benchmarked using SPOT4 (Take5) and Landsat 8 data over 12 test sites spread all over the globe (four in Europe, four in Africa, two in America and two in Asia). This variety of tests sites allows one to draw conclusions applicable to a wide variety of landscapes and crop systems. The results show that a random forest classifier operating on linearly temporally gap-filled images can achieve overall accuracies above 80% for most sites. Only two sites showed low performances: Madagascar due to the presence of fields smaller than the pixel size and Burkina Faso due to a mix of trees and crops in the fields. The approach is based on supervised machine learning techniques, which need in situ data collection for the training step, but the map production is fully automatic.

Assessment of an Operational System for Crop Type Map Production Using High Temporal and Spatial Resolution Satellite Optical Imagery

Land surface temperature (LST) is an important variable involved in the Earth's surface energy and water budgets and a key component in many aspects of environmental research. The Landsat program, jointly carried out by NASA and the USGS, has been recording thermal infrared data for the past 40 years. Nevertheless, LST data products for Landsat remain unavailable. The atmospheric correction (AC) method commonly used for mono-window Landsat thermal data requires detailed information concerning the vertical structure (temperature, pressure) and the composition (water vapor, ozone) of the atmosphere. For a given coordinate, this information is generally obtained through either radio-sounding or atmospheric model simulations and is passed to the radiative transfer model (RTM) to estimate the local atmospheric correction parameters. Although this approach yields accurate LST data, results are relevant only near this given coordinate. To meet the scientific community's demand for high-resolution LST maps, we developed a new software tool dedicated to processing Landsat thermal data. The proposed tool improves on the commonly-used AC algorithm by incorporating spatial variations occurring in the Earth's atmosphere composition. The ERA-Interim dataset (ECMWFmeteorological organization) was used to retrieve vertical atmospheric conditions, which are available at a global scale with a resolution of 0.125 degrees and a temporal resolution of 6 h. A temporal and spatial linear interpolation of meteorological variables was performed to match the acquisition dates and coordinates of the Landsat images. The atmospheric correction parameters were then estimated on the basis of this reconstructed atmospheric grid using the commercial RTMsoftware MODTRAN. The needed surface emissivity was derived from the common vegetation index NDVI, obtained from the red and near-infrared (NIR) bands of the same Landsat image. This permitted an estimation of LST for the entire image without degradation in resolution. The software tool, named LANDARTs, which stands for Landsat automatic retrieval of surface temperatures, is fully automatic and coded in the programming language Python. In the present paper, LANDARTs was used for the local and spatial validation of surface temperature obtained from a Landsat dataset covering two climatically contrasting zones: southwestern France and central Tunisia. Long-term datasets of in situ surface temperature measurements for both locations were compared to corresponding Landsat LST data. This temporal comparison yielded RMSE values ranging from 1.84 • C–2.55 • C. Landsat surface temperature data obtained with LANDARTs were then spatially compared using the ASTER data products of kinetic surface temperatures (AST08) for both geographical zones. This comparison yielded a satisfactory RMSE of about 2.55 • C. Finally, a sensitivity analysis for the effect of spatial validation on the LST correction process showed a variability of up to 2 • C for an entire Landsat image, confirming that the proposed spatial approach improved the accuracy of Landsat LST estimations.

https://hal.archives-ouvertes.fr/hal-01357429/document

A Software Tool for Atmospheric Correction and Surface Temperature Estimation of Landsat Infrared Thermal Data

Optical sensor time series images allow one to produce land cover maps at a large scale. The supervised classification algorithms have been shown to be the best to produce maps automatically with good accuracy. The main drawback of these methods is the need for reference data, the collection of which can introduce important production delays. Therefore, the maps are often available too late for some applications. Domain adaptation methods seem to be efficient for using past data for land cover map production. According to this idea, the main goal of this study is to propose several simple past data fusion schemes to override the current land cover map production delays. A single classifier approach and three voting rules are considered to produce maps without reference data of the corresponding period. These four approaches reach an overall accuracy of around 80% with a 17-class nomenclature using Formosat-2 image time series. A study of the impact of the number of past periods used is also done. It shows that the overall accuracy increases with the number of periods used. The proposed methods require at least two or three previous years to be used.

https://www.mdpi.com/2072-4292/9/11/1151/pdf

Fusion Approaches for Land Cover Map Production Using High Resolution Image Time Series without Reference Data of the Corresponding Period

Land-cover map production using remote-sensing imagery is governed by data availability. In our case, data sources are two-fold: on one hand, optical data provided regularly by satellites such as Sentinel-2, and on the other hand, reference data which allow calibrating mapping methods or validating the results. The lengthy delays due to reference data collection and cleansing are one of the main issues for applications. In this work, the use of Optimal Transport (OT) is proposed. OT is a Domain Adaptation method that uses past data, both images and reference data, to produce the land-cover map of the current period without updated reference data. Seven years of Formosat-2 image time series and the corresponding reference data are used to evaluate two OT algorithms: conventional EMD transport and regularized transport based on the Sinkhorn distance. The contribution of OT to a classification fusion strategy is also evaluated. The results show that with a 17-class nomenclature the problem is too complex for the Sinkhorn algorithm, which provides maps with an Overall Accuracy (OA) of 30%. In contrast, with the EMD algorithm, an OA close to 70% is obtained. One limitation of OT is the number of classes that can be considered at the same time. Simplification schemes are proposed to reduce the number of classes to be transported. Cases of improvement are shown when the problem is simplified, with an improvement in OA varying from 5% and 20%, producing maps with an OA near 79%. As several years are available, the OT approaches are compared to standard fusion schemes, like majority voting. The gain in voting strategies with OT use is lower than the gain obtained with standard majority voting (around 5%).

Assessment of Optimal Transport for Operational Land-Cover Mapping Using High-Resolution Satellite Images Time Series without Reference Data of the Mapping Period

Benjamin Tardy

Papers

Operational High Resolution Land Cover Map Production at the Country Scale Using Satellite Image Time Series

Assessment of an Operational System for Crop Type Map Production Using High Temporal and Spatial Resolution Satellite Optical Imagery

A Software Tool for Atmospheric Correction and Surface Temperature Estimation of Landsat Infrared Thermal Data

Fusion Approaches for Land Cover Map Production Using High Resolution Image Time Series without Reference Data of the Corresponding Period

Assessment of Optimal Transport for Operational Land-Cover Mapping Using High-Resolution Satellite Images Time Series without Reference Data of the Mapping Period