京都大学 / 0048 / 新制・課程博士 / 博士(工学) / 甲第19291号 / 工博第4088号 / 新制||工||1630(附属図書館) / 32293 / 京都大学大学院工学研究科都市社会工学専攻 / (主査)教授 小林 潔司, 教授 大津 宏康, 准教授 松島 格也 / 学位規則第4条第1項該当 / Doctor of Philosophy (Engineering) / Kyoto University / DFAM

京都大学 / 0048 / 新制・課程博士 / 博士(情報学) / 甲第21214号 / 情博第667号 / 新制||情||115(附属図書館) / 京都大学大学院情報学研究科システム科学専攻 / (主査)教授 杉江 俊治, 教授 太田 快人, 教授 大塚 敏之 / 学位規則第4条第1項該当 / Doctor of Informatics / Kyoto University / DGAM

No description available.

No description available.

No description available.

This study investigated the performance of the three estimators of surplus production model including process-observation-error-estimator with normal distribution (POE_N), observation-error-estimator with normal distribution (OE_N), and process-error-estimator with normal distribution (PE_N). The estimators with fat-tailed distributions including Student's t distribution and Cauchy distribution were also proposed and their performances were compared with the estimators with normal distribution. This study used Bayesian method, revised Metropolis Hastings within Gibbs sampling algorithm (MHGS) that was previously used to solve POE_N (Millar and Meyer, 2000), developed the MHGS for the other estimators, and developed the methodologies which enabled all the estimators to deal with data containing multiple indices based on catch-per-unit-effort (CPUE). Simulation study was conducted based on parameter estimation from two example fisheries: the Atlantic weakfish (Cynoscion regalis) and the black sea bass (Centropristis striata) southern stock. Our results indicated that POE_N is the estimator with best performance among all six estimators with regard to both accuracy and precision for most of the cases. POE_N is also the robust estimator to outliers, atypical values, and autocorrelated errors. OE_N is the second best estimator. PE_N is often imprecise. Estimators with fat-tailed distribution usually result in some estimates more biased than estimators with normal distribution. The performance of POE_N and OE_N can be improved by fitting multiple indices. Our study suggested that POE_N be used for population dynamic models in future stock assessment. Multiple indices from valid surveys should be incorporated into stock assessment models. OE_N can be considered when multiple indices are available. / Master of Science

This thesis introduces a multi-Bayesian framework for detection and classification of features in environments abundant with error-inducing noise. The approach takes advantage of Bayesian correction and classification in three distinct stages. The corrective scheme described here extracts useful but highly stochastic features from a data source, whether vision-based or otherwise, to aid in higher-level classification. Unlike many conventional methods, these features’ uncertainties are characterized so that test data can be correctively cast into the feature space with probability distribution functions that can be integrated over class decision boundaries created by a quadratic Bayesian classifier. The proposed approach is specifically formulated for road crack detection and characterization, which is one of the potential applications. For test images assessed with this technique, ground truth was estimated accurately and consistently with effective Bayesian correction, showing a 33% improvement in recall rate over standard classification. Application to road cracks demonstrated successful detection and classification in a practical domain. The proposed approach is extremely effective in characterizing highly probabilistic features in noisy environments when several correlated observations are available either from multiple sensors or from data sequentially obtained by a single sensor. / Master of Science / Humans have an outstanding ability to understand things about the world around them. We learn from our youngest years how to make sense of things and perceive our environment even when it is not easy. To do this, we inherently think in terms of probabilities, updating our belief as we gain new information. The methods introduced here allow an autonomous system to think similarly, by applying a fairly common probabilistic technique to the task of perception and classification. In particular, road cracks are observed and classified using these methods, in order to develop an autonomous road condition monitoring system. The results of this research are promising; cracks are identified and correctly categorized with 92% accuracy, and the additional “intelligence” of the system leads to a 33% improvement in road crack assessment. These methods could be applied in a variety of contexts as the leading edge of robotics research seeks to develop more robust and human-like ways of perceiving the world.

This dissertation proposes a new framework to Non-Field-of-view (NFOV) sound source localization and tracking in indoor environments. The approach takes advantage of sound signal information to localize target position through auditory sensors combination with other sensors within grid-based recursive estimation structure for tracking using nonlinear and non-Gaussian observations. Three approaches to NFOV target localization are investigated. These techniques estimate target positions within the Recursive Bayesian estimation (RBE) framework. The first proposed technique uses a numerical fingerprinting solution based on acoustic cues of a fixed microphone array in a complex indoor environment. The Interaural level differences (ILDs) of microphone pair from a given environment are constructed as an a priori database, and used for calculating the observation likelihood during estimation. The approach was validated in a parametrically controlled testing environment, and followed by real environment validations. The second approach takes advantage of acoustic sensors in combination with an optical sensor to assist target estimation in NFOV conditions. This hybrid of the two sensors constructs observation likelihood through sensor fusion. The third proposed model-based technique localizes the target by taking advantage of wave propagation physics: the properties of sound diffraction and reflection. This approach allows target localization without an a priori knowledge database which is required for the first two proposed techniques. To demonstrate the localization performance of the proposed approach, a series of parameterized numerical and experimental studies were conducted. The validity of the formulation and applicability to the actual environment were confirmed. / Ph. D.

La thèse porte sur l'extraction d'informations et l'amélioration des données RSO de un mètre de résolution visant à fournir des meilleurs descripteurs de contenu pour la compréhension des scènes et la reconnaissance de cibles, pour des produits améliorés radiométriquement et spatialement. Pour atteindre cet objectif, la thèse approche le problème de la modélisation des images RSO et propose une nouvelle solution fondée sur l'estimation du problème inverse pour l'extraction d'information. Le problème de la sélection du modèle est géré par le taux de distorsion, en raison de sa correspondance avec le cadre de l'inférence bayésienne. Nous commençons l'analyse avec l'extension de la famille de champs aléatoires de Gauss-Markov linéaires a des données à valeurs complexes, qui s'applique aux variables aléatoires à valeurs complexes : la distribution normale à plusieurs variables complexes et le modèle paramétriques des champs aléatoires de Gauss-Markov en cas de variables aléatoires correctes et incorrectes. La méthode proposée est une régularisation de Tikhonov dans le domaine complexe. Le speckle est traité comme un processus aléatoire à valeurs réelles. L'approche dans le domaine complexe permet de gérer la formation de l'image cohérente comme information ou comme incertitude dans le cas de structures ou de textures de la scène. Dans le contexte de l'optimisation des paramètres pour l'extraction de caractéristiques, a fenêtre d'analyse optimale (moyenne) et l'ordre optimal (moyen) du processus d'auto-régression sont estimés à l'aide du taux de distorsion. Cela confirme que le taux de distorsion est une bonne méthode basée sur l'entropie pour la sélection de modèle.

Selon le Groupe Intergouvernemental d'experts sur l'Évolution du Climat (GIEC), il est important de connaitre le climat passé afin de replacer le changement climatique actuel dans son contexte. Ainsi, de nombreux chercheurs ont travaillé à l'établissement de procédures permettant de reconstituer les températures ou les précipitations passées à l'aide d'indicateurs climatiques indirects. Ces procédures sont généralement basées sur des méthodes statistiques mais l'estimation des incertitudes associées à ces reconstructions reste une difficulté majeure. L'objectif principal de cette thèse est donc de proposer de nouvelles méthodes statistiques permettant une estimation précise des erreurs commises, en particulier dans le cadre de reconstructions à partir de données sur les cernes d'arbres.De manière générale, les reconstructions climatiques à partir de mesures de cernes d'arbres se déroulent en deux étapes : l'estimation d'une variable cachée, commune à un ensemble de séries de mesures de cernes, et supposée climatique puis l'estimation de la relation existante entre cette variable cachée et certaines variables climatiques. Dans les deux cas, nous avons développé une nouvelle procédure basée sur des modèles bayésiens semi- paramétriques. Tout d'abord, concernant l'extraction du signal commun, nous proposons un modèle hiérarchique semi-paramétrique qui offre la possibilité de capturer les hautes et les basses fréquences contenues dans les cernes d'arbres, ce qui était difficile dans les études dendroclimatologiques passées. Ensuite, nous avons développé un modèle additif généralisé afin de modéliser le lien entre le signal extrait et certaines variables climatiques, permettant ainsi l'existence de relations non-linéaires contrairement aux méthodes classiques de la dendrochronologie. Ces nouvelles méthodes sont à chaque fois comparées aux méthodes utilisées traditionnellement par les dendrochronologues afin de comprendre ce qu'elles peuvent apporter à ces derniers. / As stated by the Intergovernmental Panel on Climate Change (IPCC), it is important to reconstruct past climate to accurately assess the actual climatic change. A large number of researchers have worked to develop procedures to reconstruct past temperatures or precipitation with indirect climatic indicators. These methods are generally based on statistical arguments but the estimation of uncertainties associated to these reconstructions remains an active research field in statistics and in climate studies. The main goal of this thesis is to propose and study novel statistical methods that allow a precise estimation of uncertainties when reconstructing from tree-ring measurements data. Generally, climatic reconstructions from tree-ring observations are based on two steps. Firstly, a hidden environmental hidden variable, common to a collection of tree-ring measurements series, has to be adequately inferred. Secondly, this extracted signal has to be explained with the relevant climatic variables. For these two steps, we have opted to work within a semi-parametric bayesian framework that reduces the number of assumptions and allows to include prior information from the practitioner. Concerning the extraction of the common signal, we propose a model which can catch high and low frequencies contained in tree-rings. This was not possible with previous dendroclimatological methods. For the second step, we have developed a bayesian Generalized Additive Model (GAM) to explore potential links between the extracted signal and some climatic variables. This allows the modeling of non-linear relationships among variables and strongly differs from past dendrochronological methods. From a statistical perspective, a new selection scheme for bayesien GAM was also proposed and studied.