A divisive hierarchical clustering-based method for indexing image information

TL;DR

This paper introduces a divisive hierarchical clustering-based indexing method designed for efficient image retrieval in high-dimensional feature spaces, addressing the limitations of existing multi-dimensional indexing structures.

Contribution

It proposes a novel high-dimensional indexing structure using projection pursuit for effective data partitioning, improving retrieval efficiency.

Findings

Demonstrates superior performance on high-dimensional datasets

Outperforms existing indexing methods in efficiency

Validates effectiveness through extensive experiments

Abstract

In most practical applications of image retrieval, high-dimensional feature vectors are required, but current multi-dimensional indexing structures lose their efficiency with growth of dimensions. Our goal is to propose a divisive hierarchical clustering-based multi-dimensional indexing structure which is efficient in high-dimensional feature spaces. A projection pursuit method has been used for finding a component of the data, which data's projections onto it maximizes the approximation of negentropy for preparing essential information in order to partitioning of the data space. Various tests and experimental results on high-dimensional datasets indicate the performance of proposed method in comparison with others.