L'analyse et la reconstruction automatique de scène urbaine 3D est un problème fondamental dans le domaine de la vision par ordinateur et du traitement numérique de la géométrie. Cette thèse présente des méthodologies pour résoudre le problème complexe de la reconstruction d'éléments urbains en 3D à partir de données aériennes Lidar ou bien de maillages générés par imagerie Multi-View Stereo (MVS). Nos approches génèrent une représentation précise et compacte sous la forme d'un maillage 3D comportant une sémantique de l'espace urbain. Deux étapes sont nécessaires ; une identification des différents éléments de la scène urbaine, et une modélisation des éléments sous la forme d'un maillage 3D. Le Chapitre 2 présente deux méthodes de classifications des éléments urbains en classes d'intérêts permettant d'obtenir une compréhension approfondie de la scène urbaine, et d'élaborer différentes stratégies de reconstruction suivant le type d'éléments urbains. Cette idée, consistant à insérer à la fois une information sémantique et géométrique dans les scènes urbaines, est présentée en détails et validée à travers des expériences. Le Chapitre 3 présente une approche pour détecter la 'Végétation' incluses dans des données Lidar reposant sur les processus ponctuels marqués, combinée avec une nouvelle méthode d'optimisation. Le Chapitre 4 décrit à la fois une approche de maillage 3D pour les 'Bâtiments' à partir de données Lidar et de données MVS. Des expériences sur des structures urbaines larges et complexes montrent les bonnes performances de nos systèmes. / Analysis and 3D reconstruction of urban scenes from physical measurements is a fundamental problem in computer vision and geometry processing. Within the last decades, an important demand arises for automatic methods generating urban scenes representations. This thesis investigates the design of pipelines for solving the complex problem of reconstructing 3D urban elements from either aerial Lidar data or Multi-View Stereo (MVS) meshes. Our approaches generate accurate and compact mesh representations enriched with urban-related semantic labeling.In urban scene reconstruction, two important steps are necessary: an identification of the different elements of the scenes, and a representation of these elements with 3D meshes. Chapter 2 presents two classification methods which yield to a segmentation of the scene into semantic classes of interests. The beneath is twofold. First, this brings awareness of the scene for better understanding. Second, deferent reconstruction strategies are adopted for each type of urban elements. Our idea of inserting both semantical and structural information within urban scenes is discussed and validated through experiments. In Chapter 3, a top-down approach to detect 'Vegetation' elements from Lidar data is proposed using Marked Point Processes and a novel optimization method. In Chapter 4, bottom-up approaches are presented reconstructing 'Building' elements from Lidar data and from MVS meshes. Experiments on complex urban structures illustrate the robustness and scalability of our systems.

<p dir="ltr">Digital soil mapping (DSM) combines field and laboratory data with environmental factors to predict soil properties. The accuracy of these predictions depends on factors such as model selection, data quality and quantity, and landscape characteristics. In our study, we investigated the impact of sample density and the use of various environmental covariates (ECs) including slope, topographic position index, topographic wetness index, multiresolution valley bottom flatness, and multiresolution ridge top flatness, as well as the spatial resolution of these ECs on the predictive accuracy of four predictive models; Cubist (CB), Random Forest (RF), Regression Kriging (RK), and Ordinary Kriging (OK). Our analysis was conducted at three sites in Indiana: the Purdue Agronomy Center for Research and Education (ACRE), Davis Purdue Agriculture Center (DPAC), and Southeast Purdue Agricultural Center (SEPAC). Each site had its unique soil data sampling designs, management practices, and topographic conditions. The primary focus of this study was to predict the spatial distribution of soil properties, including soil organic matter (SOM), cation exchange capacity (CEC), and clay content, at different depths (0-10cm, 0-15cm, and 10-30cm) by utilizing five environmental covariates and four spatial resolutions for the ECs (1-1.5 m, 5 m, 10 m, and 30 m).</p><p dir="ltr">Various evaluation metrics, including R², root mean square error (RMSE), mean square error (MSE), concordance coefficient (pc), and bias, were used to assess prediction accuracy. Notably, the accuracy of predictions was found to be significantly influenced by the site, sample density, model type, soil property, and their interactions. Sites exhibited the largest source of variation, followed by sampling density and model type for predicted SOM, CEC, and clay spatial distribution across the landscape.</p><p dir="ltr">The study revealed that the RF model consistently outperformed other models, while OK performed poorly across all sites and properties as it only relies on interpolating between the points without incorporating the landscape characteristics (ECs) in the algorithm. Increasing sample density improved predictions up to a certain threshold (e.g., 66 samples at ACRE for both SOM and CEC; 58 samples for SOM and 68 samples for CEC at SEPAC), beyond which the improvements were marginal. Additionally, the study highlighted the importance of spatial resolution, with finer resolutions resulting in better prediction accuracy, especially for SOM and clay content. Overall, comparing data from the two depths (0-10cm vs 10-30cm) for soil properties predications, deeper soil layer data (10-30cm) provided more accurate predictions for SOM and clay while shallower depth data (0-10cm) provided more accurate predictions for CEC. Finally, higher spatial resolution of ECs such as 1-1.5 m and 5 m contributed to more accurate soil properties predictions compared to the coarser data of 10 m and 30 m resolutions.</p><p dir="ltr">In summary, this research underscores the significance of informed decisions regarding sample density, model selection, and spatial resolution in digital soil mapping. It emphasizes that the choice of predictive model is critical, with RF consistently delivering superior performance. These findings have important implications for land management and sustainable land use practices, particularly in heterogeneous landscapes and areas with varying management intensities.</p>