Centre de documentation
scientifique

Object detection using component-graphs and ConvNets with application to astronomical images / Thanh Xuan Nguyen (2021)  

Public Titre : Object detection using component-graphs and ConvNets with application to astronomical images Type de document : Thèse/HDR Auteurs : Thanh Xuan Nguyen, Auteur ; Laurent Najman, Directeur de thèse ; Hugues Talbot, Directeur de thèse Editeur : Champs-sur-Marne [France] : Université Gustave Eiffel Année de publication : 2021 Importance : 175 p. Format : 21 x 30 cm Note générale : bibliographie Thèse présentée en vue de l'obtention du Doctorat de l'Université Gustave Eiffel, Discipline Informatique Langues : Anglais (eng) Descripteur : [Vedettes matières IGN] Traitement d'image optique [Termes IGN] algorithme de filtrage [Termes IGN] classification par réseau neuronal convolutif [Termes IGN] détection d'objet [Termes IGN] image multibande [Termes IGN] lissage de données [Termes IGN] morphologie mathématique [Termes IGN] théorie des graphes Index. décimale : THESE Thèses et HDR Résumé : (auteur) This work investigates object detection algorithms with application to astronomical images. We specifically target to detect faint astronomical sources which value near the image background level. Our main directions include Mathematical Morphology (MM) and Convolutional Neural Network (ConvNet). The contributions of this study are presented in two parts:The first part proposes a novel morphological-based approach based on component-graphs and statistical hypothesis tests. The component-graphs can efficiently handle multi-band images while the statistical hypothesis tests can identify components that are significantly different from the background level. Beyond the classical component-trees and their multivariate extensions, the component-graph holds the complete structural information of multi-band images as directed acyclic graphs (DAGs). Such DAGs are more general and more powerful at the cost of non-trivial object filtering algorithms. Then, we introduce two algorithms to filter duplicated and partial components in the component-graphs. Experiments demonstrate that our proposed approach significantly improves object detection on both multi-band simulated and real astronomical images.The second part turns our attention to ConvNet direction.We introduce a real dataset of annotated astronomical objects.Based on this dataset, we propose two models: a ConvNet-based model and a hybrid model. The ConvNet-based model tailors astronomical contexts with three novel components, including a normalization layer, an object differentiation module, and a smoothness regularizer. Besides, the hybrid model uses both Morphology and ConvNet. In the hybrid method, morphological modules select region proposals while ConvNet extracts relevant information from the selected proposals. Ablation studies show that the two proposed models outperform the state of the art on both synthetic and real datasets. Note de contenu : Introduction 1- Object Detection in Astronomy I- Mathematical morphology 2- Morphological Connected Operators 3- Object Detection with Component-graphs II- ConvNet and morphology 4- ConvNet Object Detection Literature 5- ConvNet and Morphology conclusions and perspectives Numéro de notice : 15766 Affiliation des auteurs : non IGN Thématique : IMAGERIE Nature : Thèse française Note de thèse : Thèse de Doctorat : Informatique : Université Gustave Eiffel : 2021 Organisme de stage : Laboratoire d'Informatique Gaspard-Monge DOI : sans En ligne : https://hal.science/tel-03622555v1 Format de la ressource électronique : URL Permalink : https://documentation.ensg.eu/index.php?lvl=notice_display&id=100960