L'extraction de catalogues de sources fiables à partir des images est cruciale pour un large éventail de recherches en astronomie.Cependant, l'efficacité des méthodes de détection de source actuelles est sérieusement limitée dans les champs encombrés, ou lorsque les images sont contaminées par des défauts optiques, électroniques et environnementaux.Les performances en termes de fiabilité et de complétude sont aujourd'hui souvent insuffisantes au regard des exigences scientifiques des grands relevés d'imagerie.Dans cette thèse, nous développons de nouvelles méthodes pour produire des catalogues sources plus robustes et fiables.Nous tirons parti des progrès récents en apprentissage supervisé profond pour concevoir des modèles génériques et fi...
This is an Open Access article, published by EDP Sciences, under the terms of the Creative Commons A...
In this work, we propose two convolutional neural network classifiers for detecting contaminants in ...
International audienceThe observation of the transient sky through a multitude of astrophysical mess...
Extracting reliable source catalogs from images is crucial for a broad range of astronomical researc...
This thesis presents new machine learning techniques for producing high energy astronomy survey cata...
Current synoptic sky surveys monitor large areas of the sky to find variable and transient astronomi...
We present a technique for optical transient detection using artificial neural networks, particularl...
International audienceIn this work, we propose two convolutional neural network classifiers for dete...
International audienceIn this work, we propose two convolutional neural network classifiers for dete...
We present MaxiMask, a contaminant detector for ground-based astronomical images based on convolutio...
In this work, we propose to use convolutional neural networks to detect contaminants in astronomical...
Context. The future deployment of the Square Kilometer Array (SKA) will lead to a massive influx of ...
As we enter the era of large-scale imaging surveys with the upcoming telescopes such as the Large Sy...
The Euclid telescope, due for launch in 2021, will perform an imaging and slitless spectroscopy surv...
This is an Open Access article, published by EDP Sciences, under the terms of the Creative Commons A...
In this work, we propose two convolutional neural network classifiers for detecting contaminants in ...
International audienceThe observation of the transient sky through a multitude of astrophysical mess...
Extracting reliable source catalogs from images is crucial for a broad range of astronomical researc...
This thesis presents new machine learning techniques for producing high energy astronomy survey cata...
Current synoptic sky surveys monitor large areas of the sky to find variable and transient astronomi...
We present a technique for optical transient detection using artificial neural networks, particularl...
International audienceIn this work, we propose two convolutional neural network classifiers for dete...
International audienceIn this work, we propose two convolutional neural network classifiers for dete...
We present MaxiMask, a contaminant detector for ground-based astronomical images based on convolutio...
In this work, we propose to use convolutional neural networks to detect contaminants in astronomical...
Context. The future deployment of the Square Kilometer Array (SKA) will lead to a massive influx of ...
As we enter the era of large-scale imaging surveys with the upcoming telescopes such as the Large Sy...
The Euclid telescope, due for launch in 2021, will perform an imaging and slitless spectroscopy surv...
This is an Open Access article, published by EDP Sciences, under the terms of the Creative Commons A...
In this work, we propose two convolutional neural network classifiers for detecting contaminants in ...
International audienceThe observation of the transient sky through a multitude of astrophysical mess...