Query Processing Issues in Image(Multimedia) Databases

• Venue: ICDE '99 Proceedings of the 15th International Conference on Data Engineering
• Year: 1999

Multimedia databases have attracted academic and industrial interest, and systems such as QBIC (Content Based Image Retrieval system from IBM) have been released. Such systems are essential to effectively and efficiently use the existing large collections of image data in the modern computing environment. The aim of such systems is to enable retrieval of images based on their contents. This problem has brought together the (decades old) database and image processing communities.As part of our research in this area, we are building a prototype CBIR system called CHITRA. This uses a four level data model, and we have defined a Fuzzy Object Query Language(FOQL) for this system. This system enables retrieval based on high level concepts, such as "retrieve images of MOUNTAINS", "retrieve images of MOUNTAINS and SUNSET".A problem faced in this system is processing of complex queries such as "retrieve all images that have similar color histogram and similar texture to the given example image". Such problems have attracted research attention in recent times, notably by Fagin, Chaudhury and Gravano. Fagin has given an algorithm for processing such queries and provided a probabilistic upper bound for the complexity of the algorithm (which has been implemented in IBM's Garlic project). In this paper we provide theoretical (probabilistic) analysis of the expected cost of this algorithm. We propose a new multi-step query processing algorithm and prove that it performs better than Fagin's algorithm in all cases. Our algorithm requires fewer database accesses. We have evaluated both algorithms against an image database of 1000 images on our CHITRA system. We have used color histogram and Gabor texture features. Our analysis presented and the reported experimental results validate our algorithm (which has significant performance improvement).