Browsing a Classification of an Image Collection
نویسنده
چکیده
Ce document a été préparé avec L AT E X2ε et la classe these-LINA version 0.92 de l'as-sociation de jeunes chercheurs en informatique LOGIN, Université de Nantes. La classe these-LINA est disponible à l'adresse : Acknowledgements The first persons I wish to thank for this work are my two directors, Prof. José MARTI-NEZ from the Polytechnic School of the University of Nantes on the French hand, and Prof. Hiroshi ISHIKAWA from Tokyo Metropolitan University on the Japanese hand, as well as Prof. Noureddine MOUADDIB from the Polytechnic School of the University of Nantes. Preparing a Ph.D. between two universities can be both administratively and scientifically difficult, but these three professors offered me their time and energy for my Ph.D. studies to be such a great experience. Also, I am grateful to the Japanese Ministry of Research and Education (Monbukagaku-sho) that granted me a scholarship while most Japanese students have to fund their studies by themselves. I acknowledge assistant professors of Tokyo Metropolitan University, Dr. Manabu OHTA and Dr. Kaoru KATAYAMA for their help during my studies. I would also like to thank my Japanese colleagues for the insights that they gave to me: Shohei YOKOYAMA who made me taste every single Japanese dish, and Takuya WATANABE who was my guide into the Japanese otaku culture. More generally, all these years I appreciated all the members of Prof. Ishikawa's laboratory, first for welcoming me so kindly despite the poor Japanese level I had when I arrive, then for the numerous karaoke parties on Wednesday night. Next, let me have a kind thought to all the developers of the open source softwares that I used to implement my work and run the experiments (Ruby, gcc, ImageMagick, L A T E X, Linux, gnuplot.. .). Last, but no least, I lasted my family and my friends who were waiting for me in France and encouraged me all along this project. I thank them all for their permanent support.
منابع مشابه
Image Browsing and Navigation Using Hierarchical Classification
In this paper, we attempt to address the inadequacies of query-based image retrieval, and argue the importance of structured browsing and navigation methods. We propose to build a hierarchical classification model, which is the basic structure for image browsing and navigation. Currently a prototype system has been implemented, called CHROMA. CHROMA provides both a visual tool to browse or navi...
متن کاملBrowsing and Analyzing the Command-Level Structure of Large Collections of Image Manipulation Tutorials
Many people rely on Web-based tutorials to learn how to use complex software. Yet, it remains difficult for users to systematically explore the set of tutorials available online. We present Sifter, an interface for browsing, comparing and analyzing large collections of image manipulation tutorials based on their command-level structure. Sifter first applies supervised machine learning to identi...
متن کاملIncremental and hierarchical classification of a personal image collection on mobile devices
Browsing multimedia collection on mobile devices raises the needs for new multimedia indexing solutions. In this paper, we focus on the management of personal image collections. We propose a method to simplify the browsing task on such a collection. The contributions reside in an incremental hierarchical algorithm, a method to provide a textual representation of the groups obtained and an algor...
متن کاملOptimization of Brain Tumor MR Image Classification Accuracy Using Optimal Threshold, PCA and Training ANFIS with Different Repetitions
Background: One of the leading causes of death is brain tumors. Accurate tumor classification leads to appropriate decision making and providing the most efficient treatment to the patients. This study aims to optimize brain tumor MR images classification accuracy using optimal threshold, PCA and training Adaptive Neuro Fuzzy Inference System (ANFIS) with different repetitions.Material and Meth...
متن کاملPIBE: Manage Your Images the Way
A customizable system for image browsing, named PIBE, is proposed. In details, PIBE provides the user with a set of browsing and personalization facilities that enable an effective and efficient exploration of the image collection. The approach is novel and appealing because: 1) the personalization actions over the hierarchical organization of images are local, 2) the storage of the browsing st...
متن کامل