Similar Image Retrieval using DWT and LIM based Image Matching Technique

C. Lakshmi Devasena, M. Hemaltha

Abstract


The field of image retrieval and mining has become an energetic research area due to hasty enhancement in the size of digital image databases. As, a great part of information is in visual form; it is must and certainly pleasing to track for images by content. Image mining has diverse applications in various fields like remote sensing, defence, space research, medical diagnosis, biology, etc. This research work is to find out and get back the similar images when mining an image database and put forward a novel method for mining images using amalgamation of image feature extraction techniques. This approach combines Discrete Wavelet Transform, Histogram technique and Lorenz Information Measure to provide the healthy solution. It is designed and implemented on MATLAB and is tested with various images. Appropriate measures were formulated to evaluate the performance of the system. The performance of the proposed technique is noteworthy and comparable. The proposed work is suited for several day to day image mining (retrieval) and duplicate detection systems.

 


Keywords: Discrete Cosine Transform; Discrete Wavelet Transform; Histogram; Image Matching; Image Signature; Lorenz Information Measure


Full Text:

PDF


DOI: https://doi.org/10.26483/ijarcs.v3i2.1085

Refbacks

  • There are currently no refbacks.




Copyright (c) 2016 International Journal of Advanced Research in Computer Science