Multidimensional Image Indexing Using SR-Tree Algorithm for Content-Based Image Retrieval System

Abstract

Indexing is used to reduce the time required for query operation. It will minimize the time of average case and also the worst case. It also support dynamic insertion and deletion. For applying this technique to the huge databases, we need to efficiently create multi dimensional index structures, supporting nearest neighbor query. The feature extraction of the image and multidimensional indexing is used to perform content-based Image retrieval system. The SS-tree which stands for Sphere Sphere Tree occupy much larger volume than R*-Tree (Rectangle tree) and K-D-B Tree which reduces search efficiency. SR-Tree which stands for Sphere Rectangle Tree combines the bounding sphere and rectangle. A ‘region’ of the SR-Tree is small volume and a short diameter by the region intersection with bounding sphere and bounding rectangle which enhances disjoint among the regions and improves performance on the nearest neighbor queries especially for non uniform data. Here, the total process time and number of time disk access the leaf and internal node measures the index structure performance. The performance test results show SR-Tree performed most efficiently among other similarity indexing structure. The Proposed CBIR system is using SR-Tree algorithm after the process of extracting color and spatial feature and stored the values of Hue, Saturation, Value, Color Histogram, and Color cooccurrence matrix with the images.