Efficient content-based indexing of large image databases

Abstract

Large image databases have emerged in various applications in recent years. A prime requisite of these databases is the means by which their contents can be indexed and retrieved. A multilevel signature file called the Two Signature Multi-Level Signature File (2SMLSF) is introduced as an efficient access structure for large image databases. The 2SMLSF encodes image information into binary signatures and creates a tree structure that can be efficiently searched to satisfy a user's query. Two types of signatures are generated. Type I signatures are used at all tree levels except the leaf level and are based only on the domain objects included in the image. Type II signatures, on the other hand, are stored at the leaf level and are based on the included domain objects and their spatial relationships. The 2SMLSF was compared analytically to existing signature file techniques. The 2SMLSF significantly reduces the storage requirements; the index structure can answer more queries; and the 2SMLSF performance significantly improves over current techniques. Both storage reduction and performance improvement increase with the number of objects per image and the number of images in the database. For an example large image databases, a storage reduction of 78% may be achieved while the performance improvement may reach 98%.

Effectively and Efficiency Consideration for Spatial Database

“…They are binary trees that perform repeatedly in partitioning a data set at each node into two subsets. Two standard partitioning modes are identified Uhlmann in [5]: line partitioning and ball partitioning.…”

“…The search structure can be described as in the following example as in [4,5]: There are K objects in a database D, if the object of the search process is to find an object(s) that are mostly relevant to the object of the query q. In other words, a metric space (D, M) can draw D and q, where D is a corpus (group) of objects and M is the metric distance which defines resemblance, when the distances are very small mean they are more resemblance.…”

“…Therefore, many papers in SDB focus only on effectively and ignoring efficiency which leads to poor effectively, efficiency is a secondary concern. For example, a recent compendium of seminal research papers [5,6] did not include a single paper whose primary focus was efficiency. A system that takes too long to send result to clients will simply not be used.…”

“…Effectively can be improved by responding to the queries (in term of similarity retrieval) [5]. They can be summarized as follows:…”

Focused image search in the social Web.

“…In [65], Weber et al showed that the performance of these tree-based methods degrades significantly as the number of dimensions increases and is even outperformed by a sequential scan whenever the dimensionality is above The VA-file structure adopts a signature-file like filtering method, thus trying to build a mechanism for fast scanning for nearest neighbors search. Other signature-file based indexing systems include work by Essam et al [16]. Although there is no corroborative evidence showing that the inverted-file methods should be expanded to multimedia retrieval area, scalability concerns have drawn attention to the potential of using inverted files for multimedia indexing.…”