Developing a Framework for Content-based Image Retrieval Integrating New Distance Functions, Relevance Feedback e User Profile to Answer Perceptual Similarity Queries

• Grant number: 08/00485-6
• Support Opportunities: Scholarships in Brazil - Doctorate
• Start date: May 01, 2008
• End date: December 31, 2010
• Field of knowledge: Physical Sciences and Mathematics - Computer Science - Computing Methodologies and Techniques
• Principal Investigator: Agma Juci Machado Traina
• Grantee: Pedro Henrique Bugatti
• Host Institution: Instituto de Ciências Matemáticas e de Computação (ICMC). Universidade de São Paulo (USP). São Carlos , SP, Brazil

Abstract

This project aims at creating a consistent support to perform similarity queries over medical images, maintaining the semantics of a given query desired by the user. For this, it is proposed the development of new distance functions that can properly quantify the similarity between pairs of images, according to the user's expectations.Taking advantage of Relevance Feedback techniques, the distance functions will be tuned through an interactive process, and the user's profile will be used for the system to maintain the settings desired for each one of them, or for a specific class of users. The new algorithms and distance functions developed will be integrated to the access method Slim-tree and will be part of cbPACS.

Academic Publications

(References retrieved automatically from State of São Paulo Research Institutions)

BUGATTI, Pedro Henrique. Content-based image retrieval aimed at reaching user´s perception. 2012. Doctoral Thesis - Universidade de São Paulo (USP). Instituto de Ciências Matemáticas e de Computação (ICMC/SB) São Carlos.