BibTex RIS Cite

ÇOKLU ENTROPİ VE YAPAY SİNİR AĞI YÖNTEMİ (ÇE-YSA) KULLANARAK RNA VİRÜSLERİNİN OTOMATİK BULUNMASI

Year 2014, Volume: 9 Issue: 3, 1 - 13, 01.04.2014

Esin Doğantekin Derya Avcı Mustafa Poyraz Akif Doğantekin Oznur Erkuş

Abstract

Son zamanlarda, mikrobiyoloji teşhis literatürü üzerine birçok çalışma yapılmıştır. Bu çalışmada RNA virüs görüntüsünün otomatik bulunması için çoklu entropi ve yapay sinir ağı (ÇE-YSA) sistemi sunulmuştur. Bu sistem dört adımdan oluşmaktadır. Bunlar sırasıyla ön işleme, özellik çıkarma, sınıflandırma ve ÇE-YSA yöntemin doğruluğunun başarısının test edilmesidir. Ön işleme adımında merkez kenar değişim yöntemi kullanılmıştır. Bu yöntemde görüntü üzerindeki bir nesnenin merkez piksellerinden nesnenin kenarına olan Euclidean mesafesi hesaplanmıştır. Bu yüzden mesafe vektörü elde edilmiştir. Bu hesaplamalar çalışmada kullanılan RNA virüs görüntülerinin her biri için tekrar edilmiştir. Özellik çıkarma adımında özellik vektöründen norm, logaritmik entropi ve eşik entropi değerleri hesaplanmıştır. Elde edilen bu özellikler RNA virüs görüntülerinin döndürülmesi ve ölçeklenmesinde değişmezdir. Sınıflandırma adımında elde edilen özellik vektörleri YSA sınıflandırıcıya verilir. Son aşamada RNA virüsleri için ÇE-YSA algoritmasının doğruluğunun başarı oranının test edilmesi gerçekleştirilmiştir. Önerilen sistemin başarı oranı %94.02`dir.

Keywords

RNA Virüs Görüntüleri Merkez-Kenar Değişim Yöntemi Entropi YSA Sınıflandırma Ölçekleme ve Döndürmede Değişmez Özellikler

References

  • C3%B6l%C3%BCm4_goruntu_isleme.pdf, last accessed: 24 April 2009. http://www.teknohaber.net/ makale.php?id=50801, last accessed: 23 March 2009. http://ieeexplore.ieee.org/stamp/ stamp. jsp?tp=&arnumber=1567656, last accessed: 02 April 2009.
  • Gonzalez, R.C. and Woods, R.E., (1993). “Digital Image
  • Processing”, Addison-Wesley Publishing Compan. http://www.pages.drexel. edu/ ~weg22 /can_tut.html, last accessed: 08 April 2009.
  • Kulkarni, A.D., (2001). Computer Vision and Fuzzy – Neural
  • Systems, Prentice Hall PTR, USA, s.s. 509. Shannon, C.E., (1948). “A mathematical theory of communication”, Bell System Technology Journal, 27,379-423.
  • Tonga, S., Bezerianosa, A., Paula, J., Zhub, Y., and Thakora, N., (2002). “Nonextensive entropy measure of EEG following brainin jury from cardiac arrest”, Elsevier Physica A, 305, 619 – 628.
  • Principe, J.C., Euliano, N.R., and Lefebvre, W.C., (2000).
  • Neural and Adaptive Systems. John Wiley & Sons, 1th Press, New York, 656p. Overwijk, M.H.F. and Reefman, D., (2000). “Maximum-entropy deconvolution applied to electron energy-loss spectroscopy”, Pergamon Micron, 31, 325–331.
  • Li, X., (2000). “Edge directed statistical inference with applications to image processing”, PhD Thesis, Princeton University, 131p.
  • Coifman, R.R. and Wickerhauser, M.V., (1992). “Entropy- based algorithms for best basis selection”, IEEE
  • Transaction on Information Theory, 38, 2, 713-718. The Math Works Company Inc., Ver: Matlab 6.5.

THE AUTOMAIC DETECTION OF RNA VIRUSES USING MULTI-ENTROPY AND ARTIFICIAL NEURAL NETWORK METHOD (ME-ANN)

Year 2014, Volume: 9 Issue: 3, 1 - 13, 01.04.2014

Esin Doğantekin Derya Avcı Mustafa Poyraz Akif Doğantekin Oznur Erkuş

Abstract

Nowadays, there are many studies on microbiologic diagnosis literature. In this study, the Multi-entropy and Artificial Neural Network (ME-ANN) system is presented for automatic detection of RNA virus images. This system consists of four stages. They are respectively pre-processing, feature extraction, classification and test of correct detection ratio of this ME-ANN method. In pre-processing stage, it is used the center - edge changing method. In this method, Euclidian distances are calculated the from center pixells of an object on image to edges of this object. Therefore, the distance vector has been obtained. This calculating is repeated for each of RNA virus images used in this study. In feature extraction stage, the norm, the logarithmic energy and threshold entropy values are calculated as feature vector. The obtained these features are invariant from rotation and scale of these RNA virus images. In classification stage, these obtained feature vector is given to the ANN classifier. Finally the test stage is performed for evaluation the correct detection ratio of ME-ANN algorithm for RNA virus images. The correct detection ratio of the proposed system is 94.02%.

Keywords

RNA Virus Images Center-Edge Changing Method Entropy ANN Classifier Invariant Features From Scaling and Rotating

References

  • C3%B6l%C3%BCm4_goruntu_isleme.pdf, last accessed: 24 April 2009. http://www.teknohaber.net/ makale.php?id=50801, last accessed: 23 March 2009. http://ieeexplore.ieee.org/stamp/ stamp. jsp?tp=&arnumber=1567656, last accessed: 02 April 2009.
  • Gonzalez, R.C. and Woods, R.E., (1993). “Digital Image
  • Processing”, Addison-Wesley Publishing Compan. http://www.pages.drexel. edu/ ~weg22 /can_tut.html, last accessed: 08 April 2009.
  • Kulkarni, A.D., (2001). Computer Vision and Fuzzy – Neural
  • Systems, Prentice Hall PTR, USA, s.s. 509. Shannon, C.E., (1948). “A mathematical theory of communication”, Bell System Technology Journal, 27,379-423.
  • Tonga, S., Bezerianosa, A., Paula, J., Zhub, Y., and Thakora, N., (2002). “Nonextensive entropy measure of EEG following brainin jury from cardiac arrest”, Elsevier Physica A, 305, 619 – 628.
  • Principe, J.C., Euliano, N.R., and Lefebvre, W.C., (2000).
  • Neural and Adaptive Systems. John Wiley & Sons, 1th Press, New York, 656p. Overwijk, M.H.F. and Reefman, D., (2000). “Maximum-entropy deconvolution applied to electron energy-loss spectroscopy”, Pergamon Micron, 31, 325–331.
  • Li, X., (2000). “Edge directed statistical inference with applications to image processing”, PhD Thesis, Princeton University, 131p.
  • Coifman, R.R. and Wickerhauser, M.V., (1992). “Entropy- based algorithms for best basis selection”, IEEE
  • Transaction on Information Theory, 38, 2, 713-718. The Math Works Company Inc., Ver: Matlab 6.5.
There are 11 citations in total.

Details

Primary Language Turkish
Journal Section Basic Medicine Sciences
Authors

Esin Doğantekin This is me

Derya Avcı This is me

Mustafa Poyraz This is me

Akif Doğantekin This is me

Oznur Erkuş This is me

Publication Date April 1, 2014
Published in Issue Year 2014 Volume: 9 Issue: 3

Cite

APA Doğantekin, E., Avcı, D., Poyraz, M., Doğantekin, A., et al. (2014). THE AUTOMAIC DETECTION OF RNA VIRUSES USING MULTI-ENTROPY AND ARTIFICIAL NEURAL NETWORK METHOD (ME-ANN). Medical Sciences, 9(3), 1-13. https://doi.org/10.12739/NWSA.2014.9.3.1B0038
AMA Doğantekin E, Avcı D, Poyraz M, Doğantekin A, Erkuş O. THE AUTOMAIC DETECTION OF RNA VIRUSES USING MULTI-ENTROPY AND ARTIFICIAL NEURAL NETWORK METHOD (ME-ANN). Medical Sciences. April 2014;9(3):1-13. doi:10.12739/NWSA.2014.9.3.1B0038
Chicago Doğantekin, Esin, Derya Avcı, Mustafa Poyraz, Akif Doğantekin, and Oznur Erkuş. “THE AUTOMAIC DETECTION OF RNA VIRUSES USING MULTI-ENTROPY AND ARTIFICIAL NEURAL NETWORK METHOD (ME-ANN)”. Medical Sciences 9, no. 3 (April 2014): 1-13. https://doi.org/10.12739/NWSA.2014.9.3.1B0038.
EndNote Doğantekin E, Avcı D, Poyraz M, Doğantekin A, Erkuş O (April 1, 2014) THE AUTOMAIC DETECTION OF RNA VIRUSES USING MULTI-ENTROPY AND ARTIFICIAL NEURAL NETWORK METHOD (ME-ANN). Medical Sciences 9 3 1–13.
IEEE E. Doğantekin, D. Avcı, M. Poyraz, A. Doğantekin, and O. Erkuş, “THE AUTOMAIC DETECTION OF RNA VIRUSES USING MULTI-ENTROPY AND ARTIFICIAL NEURAL NETWORK METHOD (ME-ANN)”, Medical Sciences, vol. 9, no. 3, pp. 1–13, 2014, doi: 10.12739/NWSA.2014.9.3.1B0038.
ISNAD Doğantekin, Esin et al. “THE AUTOMAIC DETECTION OF RNA VIRUSES USING MULTI-ENTROPY AND ARTIFICIAL NEURAL NETWORK METHOD (ME-ANN)”. Medical Sciences 9/3 (April 2014), 1-13. https://doi.org/10.12739/NWSA.2014.9.3.1B0038.
JAMA Doğantekin E, Avcı D, Poyraz M, Doğantekin A, Erkuş O. THE AUTOMAIC DETECTION OF RNA VIRUSES USING MULTI-ENTROPY AND ARTIFICIAL NEURAL NETWORK METHOD (ME-ANN). Medical Sciences. 2014;9:1–13.
MLA Doğantekin, Esin et al. “THE AUTOMAIC DETECTION OF RNA VIRUSES USING MULTI-ENTROPY AND ARTIFICIAL NEURAL NETWORK METHOD (ME-ANN)”. Medical Sciences, vol. 9, no. 3, 2014, pp. 1-13, doi:10.12739/NWSA.2014.9.3.1B0038.
Vancouver Doğantekin E, Avcı D, Poyraz M, Doğantekin A, Erkuş O. THE AUTOMAIC DETECTION OF RNA VIRUSES USING MULTI-ENTROPY AND ARTIFICIAL NEURAL NETWORK METHOD (ME-ANN). Medical Sciences. 2014;9(3):1-13.