Research on Image Retrieval Method Based on Shape Feature

Abstract:

To meet requirements of fast and accurate image retrieve, this paper present a new method to retrieve image based on shape feature. Firstly, the technique pre-processes the image and extracts the foreground object using multi-threshold image segmentation based on color feature. In division process, threshold value can be gotten by the quick sort made used of characteristics of the background and the target pixels difference in order to avoid massive computing of other algorithms. Secondly, contour tracking method is used to obtain the contour feature which described by Fourier descriptor. Finally, similarity is measured by Euclid distance measurement. In order to verify the effectiveness of the technique, a framework for image retrieve is constructed based on shape. The result of the experiment shows that the technique is effective for image retrieve based on shape.

• Periodical: Applied Mechanics and Materials (Volumes 353-356)
• Pages: 3520-3523
• DOI: https://doi.org/10.4028/www.scientific.net/AMM.353-356.3520
• Online since: August 2013
• Authors: Ning Ding*, Fei Cai, Xun Cai
• Keywords: Contour Tracking, Multilevel Threshold, Retrieval Technique, Shape Feature

References

[1] J.H. Han and D. S. Huang, A novel BP - based image retrieval system, The IEEE International Symposium on Circuits and Systems. Kobe, Japan, pp.1557-1560, (2005).

DOI: 10.1109/iscas.2005.1464898

[2] M. Flickner, H. Sawhney, W. Niblack, J. Ashley, Q. Huang, B. Dom et al, Query by image and video content: the QBIC system, IEEE Computer. United States, vol. 28, pp.23-32, September (2005).

DOI: 10.1109/2.410146

[3] S. Wu, G. Xiong, G. Wang and G. Zhang, Digital-image retrieval based on shape-feature and MPEG-7 standard, International Conference on Consumer Electronics, Communications and Networks. China, pp.3095-3097, April (2011).

DOI: 10.1109/cecnet.2011.5768297

[4] A. K Jain and A. Vailaya, Shape-based retrieval: a case study with trademark image database, Pattern Recognition, United Kingdom, vol. 31, pp.1369-1390, September (1998).

DOI: 10.1016/s0031-3203(97)00131-3

[5] T.C. Lu, and C.C. Chang, Color image retrieval technique based on color features and image bitmap, Information Processing and Management. Elsevier, vol. 43, p.461–472, March (2007).

DOI: 10.1016/j.ipm.2006.07.014

[6] Y.D. Chun, N.C. Kim, and I.H. Jang, Content-based image retrieval using multiresolution color and texture features, IEEE Transactions onMultimedia. United States, vol. 10, p.1073–1084, October (2008).

DOI: 10.1109/tmm.2008.2001357