Open Access Journal

ISSN : 2394-2320 (Online)

International Journal of Engineering Research in Computer Science and Engineering (IJERCSE)

Monthly Journal for Computer Science and Engineering

Open Access Journal

International Journal of Engineering Research in Computer Science and Engineering (IJERCSE)

Monthly Journal for Computer Science and Engineering

ISSN : 2394-2320 (Online)

Sparse Vector Quantization and Genetic Algorithm Similarity Measure based Image Indexing Technique for CBIR

Author : D.Kiran 1 Dr.T.Venu Gopal 2 Dr.C.H.Suresh Babu 3

Date of Publication :14th November 2017

Abstract: Content-Based Image Retrieval (CBIR) plays a significant role in the image processing field. Based on image content, CBIR extracts images that are relevant to the given query image from large image archives. The effective utilization of data stored in digital library is possible through effective image indexing and retrieval techniques. Vector Quantization (VQ) method appears to be a good candidate for image indexing in CBIR, however, this method is inefficient to work on large-scale image databases. Sparse Vector Quantization (SPQ) is employed to encode the high-dimensional vector of image features, where the sparse coding technique is introduced into approximate nearest neighbor search using soft assignment technique, apart from hard assignment employed VQ. In addition, the computation of similarity amongst the query and target image using simple Euclidean Distance is also very expensive. Thus, the Genetic Algorithm-based similarity measure is performed between the query image features and the database image features in Sparse Vector Quantization. Hence, from the proposed Sparse Vector Quantization and GeneticAlgorithm-based similarity measure Image Indexing technique, the database images that are relevant to the given query image are retrieved. The performance efficiency of the proposed approach is analyzed using three image datasets such as Corel-1K, Corel-10K and Oxford-5K and showed that this approach has good precision-recall curve and maximum f-score values when compared to the existing compared CBIR approaches

Reference :

    1. Tejaswi Potluri, T. Sravani, B. Ramakrishna, Gnaneswara Rao Nitta, “Content-Based Video Retrieval Using Dominant Color and Shape Feature”, Proceedings of the First International Conference on Computational Intelligence and Informatics, pp 373-380, 2016, Springer.
    2. Tejaswi Potluri1,Gnaneswararao Nitta, “Content Based Video Retrieval UsingDominant Color Of The Truncated BlocksOf Frame”, Journal of Theoretical and Applied Information Technology, Vol.85. No.2, 2016
    3. Hsin-Hui C., Jian-Jiun D., and Hsin-Teng S., “Image retrieval based on quad tree classified vector quantization” Springer Science, Vol. 72, pp 1961- 1984,2013.
    4. Prerana K., and Nitin S., “A Survey on Content based Image Retrieval using Vector Quantization”, International Journal of Computer Applications, pp.18-22,2013.
    5. Li Jun-yi, Li Jian-hua Sparse spectral hashing for content-based image retrieval, International Journal of Intelligent Information Systems, 2015; 4(2-2): 1-4.
    6. AbirGallas, Walid Barhoumi✉, NeilaKacem, EzzeddineZagrouba, Locality-sensitive hashing for region-based large-scale image indexing, IET Image Processing, 2015, Vol. 9, Iss. 9, pp. 804–810
    7.  L. Yang, M. Di, X. Huang, and F. Duan, “Blockb-tree: A new indexstructure combined compact b+-tree with block distance,” in 2015 8thInternational Congress on Image and Signal Processing (CISP). IEEE,2015, pp. 533–538.
    8. Nhat Quang Doan1,2, Thi Phuong Nghiem1,3, Giang Son Tran1, Dynamic Indexing for Content-Based Image Retrieval Systems using Hierarchical and Topological Network, 2016 Eighth International Conference on Knowledge and Systems Engineering (KSE)

Recent Article