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.

Supervised classification systems used for land cover mapping require accurate reference databases. These reference data come generally from different sources such as field measurements, thematic maps, or aerial photographs. Due to misregistration, update delay, or land cover complexity, they may contain class label noise, i.e., a wrong label assignment. This study aims at evaluating the impact of mislabeled training data on classification performances for land cover mapping. Particularly, it addresses the random and systematic label noise problem for the classification of high resolution satellite image time series. Experiments are carried out on synthetic and real datasets with two traditional classifiers: Support Vector Machines (SVM) and Random Forests (RF). A synthetic dataset has been designed for this study, simulating vegetation profiles over one year. The real dataset is composed of Landsat-8 and SPOT-4 images acquired during one year in the south of France. The results show that both classifiers are little influenced for low random noise levels up to 25%–30%, but their performances drop down for higher noise levels. Different classification configurations are tested by increasing the number of classes, using different input feature vectors, and changing the number of training instances. Algorithm complexities are also analyzed. The RF classifier achieves high robustness to random and systematic label noise for all the tested configurations; whereas the SVM classifier is more sensitive to the kernel choice and to the input feature vectors. Finally, this work reveals that the cross-validation procedure is impacted by the presence of class label noise.

The exploitation of new high revisit frequency satellite observations is an important opportunity for agricultural applications. The Sentinel-2 for Agriculture project S2Agri (http://www.esa-sen2agri.org/SitePages/Home.aspx) is designed to develop, demonstrate and facilitate the Sentinel-2 time series contribution to the satellite EO component of agriculture monitoring for many agricultural systems across the globe. In the framework of this project, this article studies the construction of a dynamic cropland mask. This mask consists of a binary “annual-cropland/no-annual-cropland” map produced several times during the season to serve as a mask for monitoring crop growing conditions over the growing season. The construction of the mask relies on two classical pattern recognition techniques: feature extraction and classification. One pixel- and two object-based strategies are proposed and compared. A set of 12 test sites are used to benchmark the methods and algorithms with regard to the diversity of the agro-ecological context, landscape patterns, agricultural practices and actual satellite observation conditions. The classification results yield promising accuracies of around 90% at the end of the agricultural season. Efforts will be made to transition this research into operational products once Sentinel-2 data become available.

Cropland mapping relies heavily on field data for algorithm calibration, making it, in many cases, applicable only at the field campaign scale. While the recently launched Sentinel-2 satellite will be able to deliver time series over large regions, it will not really be compatible with the current mapping approach or the available in situ data. This research introduces a generic methodology for mapping annual cropland along the season at high spatial resolution with the use of globally available baseline land cover and no need for field data. The methodology is based on cropland-specific temporal features, which are able to cope with the diversity of agricultural systems, prior information from which mislabeled pixels have been removed and a cost-effective classifier. Thanks to the JECAM network, eight sites across the world were selected for global cropland mapping benchmarking. Accurate cropland maps were produced at the end of the season, showing an overall accuracy of more than 85%. Early cropland maps were also obtained at three-month intervals after the beginning of the growing season, and these showed reasonable accuracy at the three-month stage (>70% overall accuracy) and progressive improvement along the season. The trimming-based method was found to be key for using spatially coarse baseline land cover information and, thus, avoiding costly field campaigns for prior information retrieval. The accuracy and timeliness of the proposed approach shows that it has substantial potential for operational agriculture monitoring programs.

The optimal exploitation of the information provided by hyperspectral images requires the development of advanced image-processing tools. This paper proposes the construction and the processing of a new region-based hierarchical hyperspectral image representation relying on the binary partition tree (BPT). This hierarchical region-based representation can be interpreted as a set of hierarchical regions stored in a tree structure. Hence, the BPT succeeds in presenting: 1) the decomposition of the image in terms of coherent regions, and 2) the inclusion relations of the regions in the scene. Based on region-merging techniques, the BPT construction is investigated by studying the hyperspectral region models and the associated similarity metrics. Once the BPT is constructed, the fixed tree structure allows implementing efficient and advanced application-dependent techniques on it. The application-dependent processing of BPT is generally implemented through a specific pruning of the tree. In this paper, a pruning strategy is proposed and discussed in a classification context. Experimental results on various hyperspectral data sets demonstrate the interest and the good performances of the BPT representation.

Developing better agricultural monitoring capabilities based on Earth Observation data is critical for strengthening food production information and market transparency. The Sentinel-2 mission has the optimal capacity for regional to global agriculture monitoring in terms of resolution (10–20 meter), revisit frequency (five days) and coverage (global). In this context, the European Space Agency launched in 2014 the “Sentinel­2 for Agriculture” project, which aims to prepare the exploitation of Sentinel-2 data for agriculture monitoring through the development of open source processing chains for relevant products. The project generated an unprecedented data set, made of “Sentinel-2 like” time series and in situ data acquired in 2013 over 12 globally distributed sites. Earth Observation time series were mostly built on the SPOT4 (Take 5) data set, which was specifically designed to simulate Sentinel-2. They also included Landsat 8 and RapidEye imagery as complementary data sources. Images were pre-processed to Level 2A and the quality of the resulting time series was assessed. In situ data about cropland, crop type and biophysical variables were shared by site managers, most of them belonging to the “Joint Experiment for Crop Assessment and Monitoring” network. This data set allowed testing and comparing across sites the methodologies that will be at the core of the future “Sentinel­2 for Agriculture” system.