Image Retrieval using Autocorrelation Based Chordiogram Image Descriptor and Support Vector Machine
1A.Saravanan, Department of Computer and Information Science, Annamalai University, Annamalai Nagar, India.
2S.Sathiamoorthy, Tamil Virtual Academy, Chennai, India.
Manuscript received on 01 August 2019. | Revised Manuscript received on 08 August 2019. | Manuscript published on 30 September 2019. | PP: 6019-6023 | Volume-8 Issue-3 September 2019 | Retrieval Number: C5566098319/2019©BEIESP | DOI: 10.35940/ijrte.C5566.098319
Open Access | Ethics and Policies | Cite | Mendeley | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: Nowadays, evolving systems for indexing and organizing images is more important due to proliferation of images in all domains and it has made content-based image retrieval (CBIR) as significant research area. This paper uses autocorrelation based chordiogram image descriptor (ACID) for effective image representation and Support vector machine (SVM) for effective image classification. The ACID of images is computed from Haar wavelet based multiresolution domain and it exploits shape, texture and geometric details. The proposed combination of ACID and SVM is highly compatible and is comprehensively tested on benchmark datasets namely Gardens Point Walking and St. Lucia and experimental results prove that proposed combination outperforms significantly in terms of precision and recall.
Keywords: Chordiogram Image Descriptor, Autocorrelation, Edgels, Support Vector Machine.
Scope of the Article: Information Retrieval