U.P.B. Sci. Bull., Series C, Vol. 68, No. 1, 2006 AN INDEXING SCHEME FOR CONTENT-BASED RETRIEVAL OF IMAGES BY SHAPE Irina MOCANU ∗ Apariţia calculatoarelor personale şi a staţiilor de lucru performante, a standardelor de compresie audio/video precum şi existenţa diferitelor aplicaţii ce folosesc biblioteci digitale sau baze de date medicale au crescut rolul sistemelor de baze de date multimedia. Deoarece datele vizuale necesită o cantitate mare de memorie pentru stocare, precum şi o putere de calcul mare necesară procesării acestor informaţii au fost necesare structuri de indexare, precum şi algoritmi eficienţi de regăsire a acestor informaţii. În cadrul regăsirii imaginilor bazate pe conţinut, acestea sunt descrise pe baza caracteristicilor de nivel scăzut, cum ar fi culoare, forma, textură sau combinaţii ale acestora. Dintre aceste caracteristici forma reprezintă caracteristica cea mai importantă. Lucrarea prezintă o schemă de indexare pentru regăsirea imaginilor bazată pe forme. Multimedia database systems are becoming increasingly important with the advent of high-powered PC’s and workstations, audio/visual compression standards, and many applications such as digital libraries or medical databases. Because visual data require a large amount of memory and computing power for storage and processing, it is greatly desired to efficiently index and retrieve the visual information from image database systems. In content-based image retrieval, several low-level image features, such as color, texture, shape or the combination of these features, describe image. Shape is an important low-level image feature. This paper presents an indexing scheme for image retrieval by shape. Keywords: software architecture, image retrieval, shape, distance histogram, content-based indexing, R*-tree. Introduction In the recent years, image database systems are increasingly important with the advent of broadband networks, high-powered PC’s and workstations and many applications such as digital libraries, medical databases, trademark and copyright databases. Because images data require a large amount of memory and computing power for storage and processing, it is greatly desired to efficiently index and retrieve the visual information from image database systems. ∗ Assist., Dept. of Computer Science, University “Politehnica” of Bucharest, Romania