International Journal of Scientific & Technology Research

Home About Us Scope Editorial Board Blog/Latest News Contact Us
10th percentile
Powered by  Scopus
Scopus coverage:
Nov 2018 to May 2020


IJSTR >> Volume 8 - Issue 10, October 2019 Edition

International Journal of Scientific & Technology Research  
International Journal of Scientific & Technology Research

Website: http://www.ijstr.org

ISSN 2277-8616

Hybrid Shape Based Image Retrieval-Performance Improvement Using Parallel Processing

[Full Text]



Girija G. Chiddarwar, S. Phani Kumar



Shape Based Image Retrieval, Graphics Processing Unit, Locality Sensitive Hashing, Global descriptors, Local descriptors



Shape-Based Image Retrieval (SBIR) is an image mining process which extracts images based on contents of the query image. SBIR is being used in several applications like medicine, digital libraries, biodiversity information systems, historical research, and crime prevention etc. Generally, SBIR is implemented using either local features (texture, color, intensity, etc.) or global features (edges, points, contours, shape, etc.). Computation of local features for image retrieval gives high precision but low performance; while computation using global features gives high performance by sacrificing precision. To overcome the limitation of local and global features, the proposed method uses hybrid technique. And to make SBIR faster, we tried towards adventure a GPU (graphics processing unit) to parallelize both feature extraction and the similarity matching process. Meanwhile GPUs have numerous processing units; we can apply enormous parallelism in both steps. In similarity matching process, locality sensitive hashing (LSH) is used to convert descriptors to bitsets, which is further used for similarity matching. To conclude, the proposed system has increased the accuracy, precision and retrieval performance as compared with other techniques.



[1] Acharyya, Mausumi, and Malay K. Kundu. ”An adaptive approach to unsupervised
[2] texture segmentation using M-Band wavelet transform.”, Signal Processing 81.7 (2001):1337-1356.
[3] Castelli, Vittorio, Alexander Thomasian, and Chung-Sheng Li. ”CSVD: Clusteringand singular value decomposition for approximate similarity search in high-dimensionalspaces.”, IEEE Transactions on knowledge and data engineering 15.3 (2003): 671-685.
[4] ]Celebi, M. Emre, and Y. Alp Aslandogan. ”A comparative study of three moment-basedshape descriptors.”Information Technology: Coding and Computing, 2005. ITCC 2005.International Conference on. Vol. 1. IEEE, 2005.
[5] Chaux, Caroline, Laurent Duval, and J-C. Pesquet. ”Image analysis using a dual-treeM-band wavelet transform.”,IEEE Transactions on Image Processing 15.8 (2006): 2397-2412.
[6] Do, Yen, et al. ”Image retrieval using wavelet transform and shape decomposition.”,Proceedingsof the 7th International Conference on Ubiquitous Information Management and Communication.ACM, 2013.
[7] mage Database: COIL-100.http://www1.cs.columbia.edu/CAVE/software/softlib/coil-100.php.
[8] Jian, Muwei, and Liang Xu. ”Trademark image retrieval using wavelet-based shapefeatures.” Intelligent Information Technology Application Workshops, 2008. IITAW’08.International Symposium on. IEEE, 2008.
[9] Kokare, Manesh, P. K. Biswas, and B. N. Chatterji. ”Rotation-invariant texture imageretrieval using rotated complex wavelet filters.”IEEE Transactions on Systems, Man, andCybernetics, Part B (Cybernetics)36.6 (2006): 1273-1282.
[10] Leila Kabbai, Mehrez Abdellaoui,Ali Douik. ”Content Based Image Retrieval using Lo-Dept. of Computer Engineering, SCOE 53 2017-18Hybrid Shape Based Image Retrieval-Performance Improvement using Parallel Processing G26cal and Global features descriptor,2nd International Conference on Advanced Technologiesfor Signal and Image Processing - ATSIP’2016 March 21-24, 2016, Monastir, Tunisia.
[11] Likas, Aristidis, Nikos Vlassis, and Jakob J. Verbeek. ”The global k-means clusteringalgorithm.”,Pattern recognition,36.2 (2003): 451-461.
[12] Lin, Chuen-Horng, Rong-Tai Chen, and Yung-Kuan Chan. ”A smart content-basedimage retrieval system based on color and texture feature.”,Image and Vision Computing.
[13] Yousuf, M., Mehmood, Z., Habib, H.A., Mahmood, T., Saba, T., Rehman, A. and Rashid, M., 2018. A novel technique based on visual words fusion analysis of sparse features for effective content-based image retrieval. Mathematical Problems in Engineering, 2018.
[14] Alsmadi, M.K., 2018. Query-sensitive similarity measure for content-based image retrieval using meta-heuristic algorithm. Journal of King Saud University-Computer and Information Sciences, 30(3), pp.373-381.
[15] Zheng, L., Yang, Y. and Tian, Q., 2018. SIFT meets CNN: A decade survey of instance retrieval. IEEE transactions on pattern analysis and machine intelligence, 40(5), pp.1224-1244.
[16] Asha, M.M. and Ranjani, J.J., 2013, December. Secure Image Retrieval using pyramid histogram of oriented gradient descriptor. In 2013 International Conference on Advanced Computing and Communication Systems (pp. 1-5). IEEE.
[17] Hu, R. and Collomosse, J., 2013. A performance evaluation of gradient field hog descriptor for sketch based image retrieval. Computer Vision and Image Understanding, 117(7), pp.790-806.
[18] https://github.com/kayzhu/LSHash
[19] http://scikit-learn.org/stable/tutorial/index.html
[20] https://bitbucket.org/rpetek/phog-descriptor/src
[21] https://www.pyimagesearch.com/2015/12/07/local-binary-patterns-with-python-opencv/