A Comparative Evaluation of Feature Extraction and Similarity Measurement Methods for Content-based Image Retrieval

Full Text (PDF, 534KB), PP.19-32

Views: 0 Downloads: 0

Author(s)

S.M. Mohidul Islam 1,* Rameswar Debnath 1

1. Computer Science and Engineering Discipline, Khulna University, Khulna, Bangladesh

* Corresponding author.

DOI: https://doi.org/10.5815/ijigsp.2020.06.03

Received: 15 Mar. 2020 / Revised: 6 May 2020 / Accepted: 2 Jul. 2020 / Published: 8 Dec. 2020

Index Terms

RST invariant, color, texture, shape, similarity, comparative evaluation

Abstract

Content-based image retrieval is the popular approach for image data searching because in this case, the searching process analyses the actual contents of the image rather than the metadata associated with the image. It is not clear from prior research which feature or which similarity measure performs better among the many available alternatives as well as what are the best combinations of them in content-based image retrieval. We performed a systematic and comprehensive evaluation of several visual feature extraction methods as well as several similarity measurement methods for this case. A feature vector is created after color and/or texture and/or shape features extraction. Then similar images are retrieved using different similarity measures. From experimental results, we found that color moment and wavelet packet entropy features are most effective whereas color autocorrelogram, wavelet moment, and invariant moment features show narrow result. As a similarity measure, cosine and correlation measures are robust in maximum cases; Standardized L2 in few cases and on average, city block measure retrieves more similar images whereas L1 and Mahalanobis measures are less effective in maximum cases. This is the first such system to be informed by a rigorous comparative analysis of the total six features and twelve similarity measures.

Cite This Paper

S.M. Mohidul Islam, Rameswar Debnath, " A Comparative Evaluation of Feature Extraction and Similarity Measurement Methods for Content-based Image Retrieval", International Journal of Image, Graphics and Signal Processing(IJIGSP), Vol.12, No.6, pp. 19-32, 2020. DOI: 10.5815/ijigsp.2020.06.03

Reference

[1] Yu, C., Brandenburg, T.: ‘Multimedia Database Applications: Issues and Concerns for Classroom Teaching’, IJMA, 2011, 3, (1), pp. 1-9.

[2] Sclaroff, S., Taycher, L., Cascia, M.L.: ‘Imagerover: A content-based image browser for the World Wide Web’. Proc. IEEE Workshop on Content-Based Access of Image and Video Libraries, San Juan, Puerto Rico, June 1997, pp. 2–9.

[3] Prathiba, T., Darathi, G.S.: ‘An efficient content based image retrieval using local pattern’, IJAREEIE, 2013, 2, (10), pp. 5040-5046.

[4] Duanmu, X.: ‘Image retrieval using color moment invariant. Proc. Int. Con. Information Technology’, Las Vegas, USA, April 2010, pp. 200-203.

[5] Chitkara, V.; Nascimento, M.A.; Mastaller, C.: ‘Content-based image retrieval using binary signatures’, Department of Computing Science, University of Alberta, Canada, September 2000.

[6] Singla, A., Garg, M.: ‘CBIR approach based on combined HSV, auto correlogram, color moments and Gabor wavelet’, IJECS, 2014, 3, (10), pp. 9007-9012.

[7] Han, J., Ma, K.: ‘Fuzzy Color Histogram and Its Use in Color Image Retrieval’, IEEE Trans. Image Process., 2002, 11, (8), pp. 944-952.

[8] Stricker, M.A., Orengo, M.: ‘Similarity of color images’, Proc. SPIE Con. the International Society for Optical Engineering, San Jose, February 1995, pp. 381-392.

[9] Tamura, H., Mori, S., Yamawaki, T.: ‘Texture Features Corresponding to Visual Perception’, IEEE Trans. Syst., Man, Cybern., 1978, 8, (6), pp. 460-473.

[10] Sakhare, S.V., Nasre, V.G.: ‘Design of feature extraction in Content Based Image Retrieval (CBIR) using color and texture’, IJCSIS, 2011, 1, (2), pp. 57-61.

[11] Castelli, V., Bergman, L.D.: ‘Image Databases: Search and Retrieval of Digital Imagery’, (John Wiley & Sons, 1st Edn., 2002).

[12] Haralick, R.M., Shanmugam, K., Dinstein, I.: ‘Textural Features for Image Classification’, IEEE Trans. Syst., Man, Cybern., 1973, 3, (6), pp. 610–621.

[13] Latecki, L.J., Lak¨amper, R., Wolter, D.: ‘Shape similarity and visual parts’, Proc. Int. Con. Discrete Geometry for Computer Imagery, Naples, Italy, November 2003, pp. 34–51.

[14] ‘Wang Groups: Modeling Objects, Concepts, Aesthetics, and Emotions in Big Visual Data’, http://wang.ist.psu.edu/docs/home.shtml, Accessed 10 July 2018.

[15] Tungkasthan, A., Intarasema, S., Premchaiswadi, W.: ‘Spatial Color Indexing using ACC Algorithm’, Proc. Int. Con. ICT and Knowledge Engineering, Bangkok, Thailand, December 2009, pp. 113-117.

[16] Dua, S., Acharya, U.R., Chowriappa, P., et al.: ‘Wavelet-based energy features for glaucomatous image classification’, IEEE Trans. Inf. Technol. Biomed., 2012, 16, (1), pp. 80-87.

[17] Langley, P.: ‘Wavelet entropy as a measure of Ventricular beat suppression from the Electrocardiogram in Atrial Fibrillation’, Entropy, 2015, 17, (9), pp. 6397–6411.

[18] Hu, M.K.: ‘Visual Pattern Recognition by Moment Invariants’, IRE Trans. Info. Theory, 1962, 8, (2), pp. 179–187.

[19] Mahalanobis P.C.: ‘On the generalized distance in statistics’, Proc. National Institute of Sciences of India, Calcutta, India, April 1936, pp. 49–55,

[20] Cantrell, C.D.: ‘Modern Mathematical Methods for Physicists and Engineers’, (Cambridge University Press, 1st Edn. 2006).

[21] ‘Distance metrics’, http://numerics.mathdotnet.com, Accessed 19 July 2018.

[22] Chandha, A., Mallik, S., Johar, R.: ‘Comparative study and optimization of feature extraction techniques for Content Based Image Retrieval’, IJCA, 2012, 52, (20), pp. 35-42.