Conclusions and Future Work
5.1 Conclusions
With advances in medical technology and the rapid development of electronic products, the researchers recently offer various ways to help medical personnel to diagnose specific diseases easily. Therefore, the treatment can be accurately provided to patients. In the field of medical imaging, the calibration is a very important issue.
Through some suitable calibration methods, the medical images can be normalized to a same reference template before medical personnel determine or make more analyses.
In this research paper, we emphasized on the determination of the CT images in the medical pictures. We can compare the differences between the processed images and the target images. The image registration can help us to easily compare the different parts.
We used the ICP and AT to increase the degree of image registration. In conclusion, the experiment tells that the accuracy of this method is better than in either solely ICP or AT.
5.2 Future Work
The result has been accomplished in this thesis is only a preliminary study of image registration. Much of work is needed to be done in the future.
(1) Using the Nonlinear Methods
The algorithms we used in this paper are all linear functions, which have many limitations. Thus, we can move toward the non-linear method in future (ex: bilinear
interpolation [24, 25, 26]).
(2) Towards 3D Images
In this paper, we use 2D images. But a person's body or head are 3D. Therefore, we can learn how to convert 2D into 3D [27] in the future first. We can also study the brain [1] [28] first because the brain is a rigid image, which can easily help us to do 3D imaging registration.
(3) Computational Speed
We spent a lot of time in running these experimental data when doing the image registration. Therefore, we must think of a new method to minimize the time spending.
References
[1] 蔡明倫，民 91，“三度空間腦部結構校準”, 國立交通大學資訊科學研究所碩士 論文。
[2] R. C. Gonzalez and R. E. Woods, Digital Image Processing Second Edition.New Jersey: Prentice Hall, 2002.
[3] R. C. Gonzalez and R. E. Woods, Digital Image Processing Third Edition. New Jersey: Prentice Hall, 2008.
[4] P. J. Besl and N. D. McKay, “A method for registration of 3-D shapes,” IEEE Trans. Patt. Anal. Machine Intell., vol. 14, no. 2, pp. 239-256, Feb., 1992.
[5] D. Arthur and S. Vassilvitskii, “Worst-case and smoothed analysis of the ICP algorithm, with an application to the k-means method,” FOCS, pp.153-164, 2006.
[6] S. Y. Du, N. N. Zheng, S. H. Ying, Q. B. You, and Y. Wu, “An extension of the ICP algorithm considering scale factor,” IEEE International Conference on Image Processing (ICIP), pp.193-196, 2007.
[7] T. Jost and H. Hugli, “A multi-resolution scheme ICP algorithm for fast shape registration,” Proc 1st Int. Conf. 3D Data Processing Visualization and Transmission, 2002, pp. 540-543.
[8] T. Zinsser, J. Schmidt, and H. Niemann, “A refined ICP algorithm for robust 3-D correspondence estimation,” IEEE International Conference on Image Processing (ICIP), Sept. 2003.
[9] C. V. Stewart, C. L. Tsai, and B. Roysam, “The dual-bootstrap iterative closest point algorithm with application to retinal image registration,” IEEE Trans. Med.
Imag., Nov, 2003.
[10] N. Gelfand, L. Ikemoto, S. Rusinkiewicz, and M. Levoy, “3-D digital imaging and modeling,” Proc 4th Int. Conf. 3DIM 2003.
[11] R. F. Hashimoto, “An extension of an algorithm for finding sequential decomposition of erosions and dilations,” Comput. Vision Graphics Image Processing, 1998.
[12] J. Gil and R. Kimmel, “Efficient dilation, erosion, opening, and closing algorithms,” IEEE Trans. Patt. Anal. Machine Intell., vol. 24, no. 12, Dec, 2002.
[13] N. Desikachari and R. M. Haralick, “Recursive binary dilation and erosion using digital line structuring elements in arbitrary orientations,” IEEE Trans. Image Processing, vol. 9, no. 5, May, 2000.
[14] S. Chen and R. M. Haralick, “Recursive erosion, dilation, opening, and closing transforms,” IEEE Trans. Image Processing, vol. 4, no. 3, March, 1995.
[15] A. M. Siddiqi, M. Saleem and A. Masud, “A local transformation function for images,” International Conference Electrical Engineering (ICEE), April 2007.
[16] N. Y. Lee, “Automatic generation of 3D vessels model using vessels image matching based on adaptive control points,” Proc. 6th International Conf.
Advanced Language Processing and Web Information Technology, 2007.
[17] L. M. G. Fonseca and C. S. Kenney, “Control point assessment for image registration,” Comput. Vision Graphics Image Processing, pp.125-132, 1999.
[18] W. E. Hart and M. H. Goldbaum, “Registering retinal images using automatically selected control point pairs,” Proc. IEEE Int. Conf. Image Processing, pp. 576–580, 1994.
[19] L. Zagorchev and A. Goshtasby, “A comparative study of transformation functions for nonrigid image registration,”, IEEE Trans. Image Processing, vol. 15, no. 3, March, 2006.
[20] X. Chen, J. Yang, J. Zhang, and A. Waibel, “Automatic detection of signs with affine transformation,” Proc. 6th IEEE Workshop Conf. Applications of Computer
[21] Y. Zhao and B. Yuan, “A new affine transformation: Its theory and application to image coding,” IEEE Trans. Circuits Syst . , vol. 8, no. 3, 1998.
[22] O. Camara, G. Delso, O. Colliot, A. Moreno-Ingelmo, and I. Bloch, “Explicit incorporation of prior anatomical information into a nonrigid registration of thoracic and abdominal CT and 18-FDG whole-body emission PET images,”
Medical Imaging, IEEE Trans., vol. 26, no. 2, pp. 164-178, Feb, 2007.
[23] K. T. Gribbon and D. G. Bailey, “A novel approach to real-time bilinear interpolation,” Proc. 2nd IEEE Workshop Conf. DELTA 2004, pp. 126 – 131, Jan, 2004.
[24] D. G. Bailey, A. Gilman, and R. Browne, “Bias characteristics of bilinear interpolation based registration,” Proc. IEEE Region 10 Conf. TENCON 2005, pp.
1-6, Nov, 2005.
[25] M. C. Tsai and P. Y. Huang, “Design of scan converter using the locally 2-D bilinear interpolation,” Proc IEEE Int. Conf. SMC '06, vol. 5, pp.3961-3966, Oct, 2006.
[26] S. Gefen, L. Bertrand, N. Kiryati, and J. Nissanov, “Localization of sections within the brain via 2D to 3D image registration,” Proc. IEEE Int. Conf. Acoustics, Speech, and Signal Processing, March, 2005, vol. 2, pp.733-736.
[27] G. Li, T. Liu, G. Young, L. Guo, and S. T. C. Wong, “Deformation invariant attribute vector for 3D image registration: method and validation,” 3rd IEEE Int.
Symposium. Biomedical Imaging: Nano to Macro, 2006. pp.442 - 445, April, 2006.
[28] Y. Lo and S. Lee, “Affine transformation and its application to antenna arrays,”
IEEE Trans. Antennas and Propagation, vol. 13, no. 6, Nov, 1965.
[29] K. S. Arun, T. S. Huang, and S. D. Blostein, “Least square fitting of two 3-d point sets,” IEEE Trans. Patt. Anal. Machine Intell., pp.698 - 700, 1987.
[30] B. K. P. Horn, “Closed-form solution of absolute orientation using unit
quaternions,” Journal of the Optical Society of America A, pp.629 - 642, April 1987.
[31] B. K. P. Horn, H. M. Hilden, and S. H. Negahdaripour, “Closed-form solution of absolute orientation using orthonormal matrices,” Journal of the Optical Society of America A, pp.1127 - 1135, July 1988.
[32] M. W. Walker, L. Shao, and R. A. Volz, “Estimating 3-d location parameters using dual number quaternions,” CVGIP: Image Understanding, pp.358 - 367, November 1991.