Normalised Euclidean Distance Based Image Retrieval Using Coefficient Analysis

摘要

The article presented a novel method based on normalized Euclidean distance using application of discrete wavelet transform and bins intensity measurement, which is then coupled to a parameterized framework for content-based image retrieval. The discrete wavelet transform captures both frequency and location information and make image retrieval efficient. It further facilitates to incorporate recent research work on feature based coefficient distributions. We demonstrate the applicability of the proposed method in the context of color texture retrieval on different image databases and compare retrieval performance to a collection of state-of-the-art approaches in the area. Our experiment results on a large database further include a thorough analysis of computations of the main building blocks and runtime measurements of images.