Texture Image Retrieval Using A Set of Dual-Tree Rotated Complex Wavelet Filter And Dual-Tree-Complex Wavelet Transform

Journal Title: International Journal of Engineering and Science Invention - Year 2018, Vol 7, Issue 6

Abstract

Most of the texture image retrieval systems are still incapable of providing retrieval result with high retrieval accuracy and less computational complexity. This is the limitation of conventional image retrieval system. To overcome this problem, we propose a new approach of extracting texture features for texture image retrieval by using a combined set of dual-tree rotated complex wavelet filter (DT-RCWF) and dual-tree-complex wavelet transform (DT-CWT) together, which obtains texture analysis in 12 different directions. A set of twodimensional (2-D) rotated complex wavelet filters (RCWFs) are designed with complex wavelet filter coefficients, which gives texture information strongly oriented in six different directions (450 apart from complex wavelet transform). The 2-D RCWFs are non-separable and oriented, which improves characterization of oriented textures. The information provided by DT-RCWF complements the information generated by DTCWT. Features are obtained by computing the energy and standard deviation on each subband of the decomposed image. To check the retrieval performance, texture database D of 500 texture images is created. Experimental results on database D consisting of 500 texture images are categorized into 5 different classes indicates that our method significantly improves retrieval performance by considering African, bus and dinosaur category examples is 81.25%, 87.50%, and 87.50% respectively by using DT-RCWF +DT-CWT, compared with from 25%, 25% and 25% of traditional approach of Gabor wavelet-based features, respectively. Experimental result also shows retrieval time for query image using proposed method is 0.08 second which is very less than that using Gabor wavelets based method having 0.146 second. The proposed method also retains comparable levels of computational complexity.

Authors and Affiliations

Sachin S. Mahulkar, Dr. Ajay P. Thakare, Ajaykumar K. Kakde

Keywords

Related Articles

Improved Particle Filter For Nor-Linear Noisy Observation Series

The particle filtering (PF) theory is a very method to denoise the non-linear disturbed signals. It could disposal signals with non-Gauss noise which can’t be done by Kalman filtering (KF). The theory is widely used in t...

Durability studies and NDT of High Grade Concrete Replacing Main Ingredients By Quarry Dust & Silicafume

The durability of concrete is one of its most important properties because it is essential that especially the high grade concrete should be capable of withstanding the conditions its intended life. The durability and st...

Suitability of Awka-North (Nigeria) Sedimentary stones as Coarse Aggregate for Building Construction

The research was conducted to investigate properties and strength behaviour of concrete incorporating Awka-North sedimentary stones (locally known as‘ironstone’) as coarse aggregate in concrete. Ironstone is abundantly a...

Ranking Models for Selection of Cloud Services Based On the Fuzzy Logic

The third Millennium generation has been geared to operate on large recourses with verity of choices of services to satisfy the user’s requirement. Hence in the recent past, there is wide increase in the demand of cloud...

Keywords - School, Socialization, Education, Sociology, Students.

Social progress is not always so simple. It usually comes about very gradually. But the gradual evolutionary pace of social progress is not fast enough. As a social system the school aims to prepare the students to occup...

Download PDF file
  • EP ID EP397403
  • DOI -
  • Views 88
  • Downloads 0

How To Cite

Sachin S. Mahulkar, Dr. Ajay P. Thakare, Ajaykumar K. Kakde (2018). Texture Image Retrieval Using A Set of Dual-Tree Rotated Complex Wavelet Filter And Dual-Tree-Complex Wavelet Transform. International Journal of Engineering and Science Invention, 7(6), 40-48. https://europub.co.uk/articles/-A-397403