Image Registration Based on Normalized Cross Correlation and Discrete Cosine Transform

AUTHOR(S):

TITLE

Image Registration Based on Normalized Cross Correlation and Discrete Cosine Transform

ABSTRACT

There have been great advancements in recent years regarding computer vision, medical imaging, cartography, astronomy and similar image acquisition methods. Therefore, there is certainly need for efficient image registration techniques. Image registration is a process in which two images, that represent the same scene but are captured from different angles, sensors or time periods, are geometrically aligned to each other. This is one of the fundamental image processing techniques and is very useful when combining images captured from various sensors or time periods. The basic concept is to find points of interest by comparing unregistered image with source image. Points of interest are parts of source image that are highly similar to unregistered image. In this paper, points of interest are obtained by two approaches - by using normalized cross correlation (NCC) and discrete cosine transform (DCT). The proposed method was applied to satellite images. Tests have been successfully concluded even with high resolution images with some form of local distortion. Testing shows that DCT approach is quicker and more accurate.

KEYWORDS

normalized cross correlation, discrete cosine transform, image registration, root mean squared error, points of interest extraction

REFERENCES

[1] E. Tuba, M. Tuba, and R. Jovanovic, “An algorithm for automated segmentation for bleeding detection in endoscopic images,” in International Joint Conference on Neural Networks (IJCNN). IEEE, 2017, pp. 4579–4586.

[2] A. Stojak, E. Tuba, and M. Tuba, “Framework for abnormality detection in magnetic resonance brain images,” in 24th Telecommunications Forum TELFOR. IEEE, 2016, pp. 687–690.

[3] N. Grice, C. Christian, A. Nota, and P. Greenfield, “3D printing technology: A unique way of making hubble space telescope images accessible to non-visual learners.” Journal of Blindness Innovation & Research, vol. 5, no. 1, 2015.

[4] B. Cook, D. Proctor, R. Bromberg, N. E. La- Pointe, S. C. Feinstein, and L. Wilson, “Digital quantification of neurite outgrowth and retraction by phase-contrast microscopy: A tau perspective,” in Methods in cell biology. Elsevier, 2017, vol. 141, pp. 217–228.

[5] E. Tuba, M. Tuba, and E. Dolicanin, “Adjusted fireworks algorithm applied to retinal image registration,” Studies in Informatics and Control, vol. 26, no. 1, pp. 33–42, 2017.

[6] M.-S. Lee, M. Shen, A. Yoneyama, and C.- C. J. Kuo, “DCT-domain image registration techniques for compressed video,” in IEEE International Symposium on Circuits and Systems (ISCAS). IEEE, 2005, pp. 4562–4565.

[7] R. B. Karani and T. K. Sarode, “Image registration using discrete cosine transform and normalized cross correlation,” in Proceedings on International Conference and workshop on Emerging Trends in Technology, 2012, pp. 28–34.

[8] E. Tuba, I. Ribic, R. Capor-Hrosik, and M. Tuba, “Support vector machine optimized by elephant herding algorithm for erythemato-squamous diseases detection,” Procedia Computer Science, vol. 122, pp. 916–923, 2017.

[9] I. Tuba, E. Tuba, and M. Beko, “Digital image forgery detection based on shadow texture features,” in 24th Telecommunications Forum (TELFOR). IEEE, 2016, pp. 679–682.

[10] E. Tuba, L. Mrkela, and M. Tuba, “Retinal blood vessel segmentation by support vector machine classification,” in 27th International Conference Radioelektronika. IEEE, 2017, pp. 1–6.

[11] E. Tuba, M. Tuba, D. Simian, and R. Jovanovic, “JPEG quantization table optimization by guided fireworks algorithm,” in International Workshop on Combinatorial Image Analysis, LNCS, 2017, vol. 10256, pp. 294–307.

[12] NASA, “Landsat image gallery.” [Online]. Available: https://landsat.visibleearth.nasa.gov/

Cite this paper

Petar Rutesic, Zorana Stosic. (2018) Image Registration Based on Normalized Cross Correlation and Discrete Cosine Transform. International Journal of Signal Processing, 3, 16-20