ORCID Profile
0000-0002-4652-7778
Current Organisation
Univ. Bretagne Sud/IRISA
Does something not look right? The information on this page has been harvested from data sources that may not be up to date. We continue to work with information providers to improve coverage and quality. To report an issue, use the Feedback Form.
Publisher: MDPI AG
Date: 18-02-2017
DOI: 10.3390/RS9020173
Publisher: MDPI AG
Date: 03-11-2017
DOI: 10.3390/RS9111119
Publisher: MDPI AG
Date: 19-11-2021
DOI: 10.3390/RS13224668
Abstract: Crop maps are key inputs for crop inventory production and yield estimation and can inform the implementation of effective farm management practices. Producing these maps at detailed scales requires exhaustive field surveys that can be laborious, time-consuming, and expensive to replicate. With a growing archive of remote sensing data, there are enormous opportunities to exploit dense satellite image time series (SITS), temporal sequences of images over the same area. Generally, crop type mapping relies on single-sensor inputs and is solved with the help of traditional learning algorithms such as random forests or support vector machines. Nowadays, deep learning techniques have brought significant improvements by leveraging information in both spatial and temporal dimensions, which are relevant in crop studies. The concurrent availability of Sentinel-1 (synthetic aperture radar) and Sentinel-2 (optical) data offers a great opportunity to utilize them jointly however, optimizing their synergy has been understudied with deep learning techniques. In this work, we analyze and compare three fusion strategies (input, layer, and decision levels) to identify the best strategy that optimizes optical-radar classification performance. They are applied to a recent architecture, notably, the pixel-set encoder–temporal attention encoder (PSE-TAE) developed specifically for object-based classification of SITS and based on self-attention mechanisms. Experiments are carried out in Brittany, in the northwest of France, with Sentinel-1 and Sentinel-2 time series. Input and layer-level fusion competitively achieved the best overall F-score surpassing decision-level fusion by 2%. On a per-class basis, decision-level fusion increased the accuracy of dominant classes, whereas layer-level fusion improves up to 13% for minority classes. Against single-sensor baseline, multi-sensor fusion strategies identified crop types more accurately: for ex le, input-level outperformed Sentinel-2 and Sentinel-1 by 3% and 9% in F-score, respectively. We have also conducted experiments that showed the importance of fusion for early time series classification and under high cloud cover condition.
Publisher: IEEE
Date: 07-2017
Publisher: IEEE
Date: 07-2015
Publisher: Elsevier BV
Date: 12-2016
Publisher: IEEE
Date: 06-2017
Publisher: Springer Science and Business Media LLC
Date: 06-02-2019
Publisher: IEEE
Date: 07-2016
Publisher: IEEE
Date: 07-2016
Publisher: MDPI AG
Date: 04-03-2019
DOI: 10.3390/RS11050523
Abstract: 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.
Publisher: Springer Science and Business Media LLC
Date: 05-03-2020
Publisher: MDPI AG
Date: 27-06-2019
DOI: 10.3390/RS11131520
Abstract: This paper discusses the combined use of remotely sensed optical and radar data for the estimation and mapping of soil texture. The study is based on Sentinel-1 (S-1) and Sentinel-2 (S-2) data acquired between July and early December 2017, on a semi-arid area about 3000 km2 in central Tunisia. In addition to satellite acquisitions, texture measurement s les were taken in several agricultural fields, characterized by a large range of clay contents (between 13% and 60%). For the period between July and August, various optical indicators of clay content Short-Wave Infrared (SWIR) bands and soil indices) were tested over bare soils. Satellite moisture products, derived from combined S-1 and S-2 data, were also tested as an indicator of soil texture. Algorithms based on the support vector machine (SVM) and random forest (RF) methods are proposed for the classification and mapping of clay content and a three-fold cross-validation is used to evaluate both approaches. The classifications with the best performance are achieved using the soil moisture indicator derived from combined S-1 and S-2 data, with overall accuracy (OA) of 63% and 65% for the SVM and RF classifications, respectively.
Publisher: Springer Science and Business Media LLC
Date: 04-03-2021
Publisher: MDPI AG
Date: 08-2021
DOI: 10.3390/IJGI10080523
Abstract: Multitemporal environmental and urban studies are essential to guide policy making to ultimately improve human wellbeing in the Global South. Land-cover products derived from historical aerial orthomosaics acquired decades ago can provide important evidence to inform long-term studies. To reduce the manual labelling effort by human experts and to scale to large, meaningful regions, we investigate in this study how domain adaptation techniques and deep learning can help to efficiently map land cover in Central Africa. We propose and evaluate a methodology that is based on unsupervised adaptation to reduce the cost of generating reference data for several cities and across different dates. We present the first application of domain adaptation based on fully convolutional networks for semantic segmentation of a dataset of historical panchromatic orthomosaics for land-cover generation for two focus cities Goma-Gisenyi and Bukavu. Our experimental evaluation shows that the domain adaptation methods can reach an overall accuracy between 60% and 70% for different regions. If we add a small amount of labelled data from the target domain, too, further performance gains can be achieved.
No related grants have been discovered for Charlotte Pelletier.