AN INTEGRATED IMAGE ANALYSIS AND IMAGE DATABASE MANAGEMENT SYSTEM

NING-SAN CHANG, Purdue University

Abstract

Record-keeping and decision-making are increasingly based on data stored in computer processable databases. The potential for applying computers to vast amounts of images, such as earth resources satellite images (LANDSAT), computer tomography and medical images, leads to the necessity for carefully designing image database or pictorial database systems. An Integrated Image Analysis and Image Database Management System (IMAID) is presented as an integrated database system interfaced with an image understanding system for the efficient storage, retrieval and manipulation of pictorial information. By using image processing and pattern recognition manipulation functions, structures and features of images are extracted and integrated into relational databases. A relational query language, Query-by-Pictorial-Example (QPE) is introduced for manipulating queries regarding pictorial relations as well as conventional queries. In addition to the manipulating capabilities of the usually query languages, queries can also be expressed in terms of pictorial examples through a display terminal. A parallel parsing algorithm for tree languages is presented for the implementation of a more efficient Error-Correcting Tree Automata (ECTA). It is then used to recognize features from images. The conversion of a relational graph to relational database is discussed. Capabilities of QPE are discussed and data representation and manipulation algorithms for QPE are presented. Illustrative examples include the application of the proposed parallel parsing algorithms to the interpretation of LANDSAT images and the classification of textures. An experimental IMAID is implemented for LANDSAT images.

Degree

Ph.D.

Subject Area

Electrical engineering|Computer science

Off-Campus Purdue Users:
To access this dissertation, please log in to our
proxy server
.

Share

COinS