A Fast Template Matching Algorithm Based on Wavelet Coefficient Projection Transform

Abstract:

As the electronic industry advances rapidly toward automatic manufacturing smaller, faster, and cheaper products, computer vision play more important role in IC packaging technology than before. One of the important tasks of computer vision is finding target position through similarity matching. Similarity matching requires distance computation of feature vectors for each target image. In this paper we propose a projection transform of wavelet coefficient based multi resolution data-structure algorithm for faster template matching, a position sequence of local sharp variation points in such signals is recorded as features. The proposed approach reduces the number of computation by around 70% over multi resolution data structure algorithm. We use the proposed approach to match similarity between wavelet parameters histograms for image matching. It is noticeable that the proposed fast algorithm provides not only the same retrieval results as the exhaustive search, but also a faster searching ability than existing fast algorithms. The proposed approach can be easily combined with existing algorithms for further performance enhancement.

• Periodical: Applied Mechanics and Materials (Volumes 48-49)
• Pages: 21-24
• DOI: https://doi.org/10.4028/www.scientific.net/AMM.48-49.21
• Online since: February 2011
• Authors: Xian Ping Fu, Sheng Long Liao
• Keywords: Fast Matching, Machine Vision, Semiconductor, Wafer Bonding, Wavelet Transform (WT)

References

[1] M. Gharavi-Alkhansari, A fast globally optimal algorithm for template matching using low-resolution pruning, IEEE Trans Image Process, vol. 10, pp.526-33, (2001).

DOI: 10.1109/83.913587

[2] P. Ferragina, G. Landau, C. Hundt, and M. Liśkiewicz, Two-Dimensional Pattern Matching with Combined Scaling and Rotation, in Combinatorial Pattern Matching. vol. 5029: Springer Berlin / Heidelberg, 2008, pp.5-17.

DOI: 10.1007/978-3-540-69068-9_4

[3] R. S. Feris, V. Krueger, and R. M. Cesar, A wavelet subspace method for real-time face tracking, Real-Time Imaging, vol. 10, pp.339-350, (2004).

DOI: 10.1016/j.rti.2004.06.002

[4] Y. Hel-Or and H. Hel-Or, Real-time pattern matching using projection kernels, Ieee Transactions on Pattern Analysis and Machine Intelligence, vol. 27, pp.1430-1445, (2005).

DOI: 10.1109/tpami.2005.184

[5] D. Mohr and G. Zachmann, Silhouette Area Based Similarity Measure for Template Matching in Constant Time, in Articulated Motion and Deformable Objects, Proceedings. vol. 6169, F. J. Perales and R. B. Fisher, Eds. Berlin: Springer-Verlag Berlin, 2010, pp.43-54.

DOI: 10.1007/978-3-642-14061-7_5

[6] F. Essannouni, R. Thami, D. Aboutajdine, and A. Salam, Adjustable SAD matching algorithm using frequency domain, Journal of Real-Time Image Processing, vol. 1, pp.257-265, (2007).

DOI: 10.1007/s11554-007-0026-0

[7] L. Di Stefano and S. Mattoccia, Fast template matching using bounded partial correlation, Machine Vision and Applications, vol. 13, pp.213-221, (2003).

DOI: 10.1007/s00138-002-0070-5

[8] S. Sha, J. Chen, and S. Luo, A fast matching algorithm based on K-degree template, " in Computer Science & Education, 2009. ICCSE , 09. 4th International Conference on, 2009, p.1967-(1971).

DOI: 10.1109/iccse.2009.5228224

[9] M. L. Comer and E. J. Delp, Segmentation of textured images using a multiresolution Gaussian autoregressive model, IEEE Trans Image Process, vol. 8, pp.408-20, (1999).

DOI: 10.1109/83.748895