Pierre Féménias

Bio: Pierre Féménias is an academic researcher from European Space Agency. The author has contributed to research in topics: Altimeter & Microwave radiometer. The author has an hindex of 12, co-authored 47 publications receiving 885 citations.

First Three Years of the Microwave Radiometer Aboard ENVISAT: In-Flight Calibration, Processing and Validation of the Geophysical Products

Abstract: The Envisat microwave radiometer is designed to correct the satellite altimeter data for the excess path delay resulting from tropospheric humidity. Neural networks have been used to formulate the inversion algorithm to retrieve this quantity from the measured brightness temperatures. The learning database has been built with European Centre for Medium-Range Weather Forecasts (ECMWF) analyses and simulated brightness temperatures by a radiative transfer model. The in-flight calibration has been performed in a consistent way by adjusting measurements on simulated brightness temperatures. Finally, coincident radiosonde measurements are used to validate the Envisat wet-tropospheric correction, and this comparison shows the good performances of the method.

First three years of the microwave radiometer aboard envisat : in-flight calibration, processing and validation of the geophysical products

Abstract: The Envisat microwave radiometer is designed to correct the satellite altimeter data for the excess path delay resulting from tropospheric humidity. Neural networks have been used to formulate the inversion algorithm to retrieve this quantity from the measured brightness temperatures. The learning database has been built with European Centre for Medium-Range Weather Forecasts (ECMWF) analyses and simulated brightness temperatures by a radiative transfer model. The in-flight calibration has been performed in a consistent way by adjusting measurements on simulated brightness temperatures. Finally, coincident radiosonde measurements are used to validate the Envisat wet-tropospheric correction, and this comparison shows the good performances of the method.

Global trends in wind speed and wave height over the past 25 years

Abstract: Wind speeds over the world’s oceans have increased over the past two decades, as have wave heights. Studies of climate change typically consider measurements or predictions of temperature over extended periods of time. Climate, however, is much more than temperature. Over the oceans, changes in wind speed and the surface gravity waves generated by such winds play an important role. We used a 23-year database of calibrated and validated satellite altimeter measurements to investigate global changes in oceanic wind speed and wave height over this period. We find a general global trend of increasing values of wind speed and, to a lesser degree, wave height, over this period. The rate of increase is greater for extreme events as compared to the mean condition.

Unsupervised Deep Feature Extraction for Remote Sensing Image Classification

Abstract: This paper introduces the use of single layer and deep convolutional networks for remote sensing data analysis. Direct application to multi- and hyper-spectral imagery of supervised (shallow or deep) convolutional networks is very challenging given the high input data dimensionality and the relatively small amount of available labeled data. Therefore, we propose the use of greedy layer-wise unsupervised pre-training coupled with a highly efficient algorithm for unsupervised learning of sparse features. The algorithm is rooted on sparse representations and enforces both population and lifetime sparsity of the extracted features, simultaneously. We successfully illustrate the expressive power of the extracted representations in several scenarios: classification of aerial scenes, as well as land-use classification in very high resolution (VHR), or land-cover classification from multi- and hyper-spectral images. The proposed algorithm clearly outperforms standard Principal Component Analysis (PCA) and its kernel counterpart (kPCA), as well as current state-of-the-art algorithms of aerial classification, while being extremely computationally efficient at learning representations of data. Results show that single layer convolutional networks can extract powerful discriminative features only when the receptive field accounts for neighboring pixels, and are preferred when the classification requires high resolution and detailed results. However, deep architectures significantly outperform single layers variants, capturing increasing levels of abstraction and complexity throughout the feature hierarchy.