Focusing on different types of queries and images in the iDWing (image data warehousing environment)

• Grant number: 11/21636-5
• Support Opportunities: Scholarships in Brazil - Scientific Initiation
• Start date: January 01, 2012
• End date: December 31, 2012
• Field of knowledge: Physical Sciences and Mathematics - Computer Science - Computing Methodologies and Techniques
• Principal Investigator: Cristina Dutra de Aguiar
• Grantee: Lucas de Carvalho Scabora
• Host Institution: Instituto de Ciências Matemáticas e de Computação (ICMC). Universidade de São Paulo (USP). São Carlos , SP, Brazil

Abstract

It is being under development by the research group of the supervisor of this project the iDWing, a novel non-conventional data warehousing environment aimed at storing and retrieving images in decision-support applications. In the current phase of development, iDWing introduces the following distinctive good properties: (i) it extends the star-schema of a conventional data warehouse to also incorporate a dimension table specifically designed to store image data; (ii) it defines a strategy aimed to extract, translate and load image data into the extended data warehouse; (iii) it introduces a range query algorithm to perform analytical queries based on similarity over the image data warehouse; and (iv) it is mainly directed to the decision-making in the medical field. This project focuses on introducing extensions to iDWing, allowing that this environment be used to perform other types of analytical queries based on similarity and be able to handle other types of images. In detail, the objectives of this project are: (i) to develop a k-nearest neighbor query algorithm; (ii) to develop a k-nearest diverse neighbor query algorithm; and (iii) to extend the iDWing to handle remote sensor images. The project also aims to perform experimental tests to evaluate the proposed extensions. (AU)