Advanced Kernel Methods For Remote Sensing Image Classification

Résumé Suite aux recentes avancées technologiques, les archives d'images digitales ont connu une croissance qualitative et quantitative sans précédent. Malgré les énormes possibilités qu'elles offrent, ces avancées posent de nouvelles questions quant au traitement des masses de données saisies. Cette question est à la base de cette Thèse: les problèmes de traitement d'information digitale à très haute résolution spatiale et/ou spectrale y sont considérés en recourant à des approches d'apprentissage statistique, les méthodes à noyau. Cette Thèse étudie des problèmes de classification d'images, c'est à dire de catégorisation de pixels en un nombre réduit de classes refletant les propriétés spectrales et contextuelles des objets qu'elles représentent. L'accent est mis sur l'efficience des algorithmes, ainsi que sur leur simplicité, de manière à augmenter leur potentiel d'implementation pour les utilisateurs. De plus, le défi de cette Thèse est de rester proche des problèmes concrets des utilisateurs d'images satellite sans pour autant perdre de vue l'intéret des méthodes proposées pour le milieu du machine learning dont elles sont issues. En ce sens, ce travail joue la carte de la transdisciplinarité en maintenant un lien fort entre les deux sciences dans tous les développements proposés. Quatre modèles sont proposés: le premier répond au problème de la haute dimensionalité et de la redondance des données par un modèle optimisant les performances en classification en s'adaptant aux particularités de l'image. Ceci est rendu possible par un système de ranking des variables (les bandes) qui est optimisé en même temps que le modèle de base: ce faisant, seules les variables importantes pour résoudre le problème sont utilisées par le classifieur. Le manque d'information étiquétée et l'incertitude quant à sa pertinence pour le problème sont à la source des deux modèles suivants, basés respectivement sur l'apprentissage actif et les méthodes semi-supervisées: le premier permet d'améliorer la qualité d'un ensemble d'entraînement par interaction directe entre l'utilisateur et la machine, alors que le deuxième utilise les pixels non étiquetés pour améliorer la description des données disponibles et la robustesse du modèle. Enfin, le dernier modèle proposé considère la question plus théorique de la structure entre les outputs: l'intègration de cette source d'information, jusqu'à présent jamais considérée en télédétection, ouvre des nouveaux défis de recherche. Advanced kernel methods for remote sensing image classification Devis Tuia Institut de Géomatique et d'Analyse du Risque September 2009 Abstract The technical developments in recent years have brought the quantity and quality of digital information to an unprecedented level, as enormous archives of satellite images are available to the users. However, even if these advances open more and more possibilities in the use of digital imagery, they also rise several problems of storage and treatment. The latter is considered in this Thesis: the processing of very high spatial and spectral resolution images is treated with approaches based on data-driven algorithms relying on kernel methods. In particular, the problem of image classification, i.e. the categorization of the image's pixels into a reduced number of classes reflecting spectral and contextual properties, is studied through the different models presented. The accent is put on algorithmic efficiency and the simplicity of the approaches proposed, to avoid too complex models that would not be used by users. The major challenge of the Thesis is to remain close to concrete remote sensing problems, without losing the methodological interest from the machine learning viewpoint: in this sense, this work aims at building a bridge between the machine learning and remote sensing communities and all the models proposed have been developed keeping in mind the need for such a synergy. Four models are proposed: first, an adaptive model learning the relevant image features has been proposed to solve the problem of high dimensionality and collinearity of the image features. This model provides automatically an accurate classifier and a ranking of the relevance of the single features. The scarcity and unreliability of labeled. information were the common root of the second and third models proposed: when confronted to such problems, the user can either construct the labeled set iteratively by direct interaction with the machine or use the unlabeled data to increase robustness and quality of the description of data. Both solutions have been explored resulting into two methodological contributions, based respectively on active learning and semisupervised learning. Finally, the more theoretical issue of structured outputs has been considered in the last model, which, by integrating outputs similarity into a model, opens new challenges and opportunities for remote sensing image processing.

Pratique du scanner en Suisse: fréquence et évolution temporelle [CT utilisation in Switzerland: frequency and secular trends]

PURPOSE: This study analyzes CT examinations in Switzerland. MATERIALS AND METHODS: Using different sources (administrative data on the equipment, a 1998 nationwide inquiry into practices, and data provided by the Swiss University Hospitals of Basel, Zurich, and Lausanne), we determined the frequency of CT examinations (hospitals and private radiologists) in 1998 according to different descriptive variables and studied the progression in CT use over time. RESULTS: CT scanners increased by 7% between 1998 and 2004. The average annual number of CT examinations in 1998 was 46.3/1000 population, 3.4% of all radiological examinations in Switzerland in 1997-1998. The most frequent examination was CT of the skull (24%), while private radiology institutes perform more CTs of the spine. More CT examinations were performed for men than for women (sex ratio M/F=1.17). The average annual increase in CT in Swiss hospitals varied from 8% for Basel to 18% for Lausanne. Finally, the proportion of pediatric examinations was 5%; their numbers appear to be stabilizing. CONCLUSION: There is a significant increase in CT examinations. It is hoped that our study will heighten awareness among doctors of CT examinations in order to optimize their use.

Prévention du métanome en Suisse: où en sommes-nous [Melanoma prevention in Switzerland: where do we stand?]

Over the last 50 years, skin cancer rates (particularly melanoma) have markedly increased in Caucasian populations. Switzerland, with some 1,600 cases of, and 220 deaths from, malignant melanoma per year has among the highest rates in Europe. This public health issue, affecting relatively young people, has led to primary and secondary nationwide prevention campaigns being carried out for nearly 20 years. Observed changes in sun protection knowledge and attitudes have yet to impact on incidence trend. Early detection has resulted in a large increase in rates of thin melanoma with little change in rates of thick melanoma. Mortality has levelled off and a recent decrease, especially in women, cannot be ruled out. The efficiency of prevention campaigns should soon become more blatant if current efforts persist.