Automated identification of type and evaluation of noise parameter with multifractal indices

Authors

  • Марина Вячеславовна Полякова Odesa national polytechnic university Boulevard of Shevchenko,1, Odesa, Ukraine, 65044, Ukraine
  • Юрий Владимирович Емец Odesa national polytechnic university Boulevard of Shevchenko,1, Odesa, Ukraine, 65044, Ukraine

DOI:

https://doi.org/10.15587/1729-4061.2013.14842

Keywords:

multifractal indices, authentication of type of noise, evaluation of parameter of noise, additive-fluctuation, impulsive and multiplicative noise

Abstract

The method of the automated authentication of type and evaluation of parameter of additive-fluctuation, impulsive and multiplicative noise is worked out and investigated, and also the mixed noise on images by means of multifractal indices. Descriptions of the worked out method are investigated on test images.

Method of determination of type and evaluation of parameter of noise are used in such areas of processing of images as medicine, astronomy, radio-location, non-destructive control, technical diagnostics and other areas.

 A method consists of  the next stages: choice of homogeneous area dark-and-light by a man-operator; calculation of multifractal indices as a vector of signs(Нx, Hy, Hxy, Сх, Су, Cxy);  determination of type of hindrance dark-and-light; evaluation of parameter of noise in the areas of multifractal indices and output of identifier of type and value of parameter of noise.

On the conducted results  of experiment an offered method on the basis of multifractal indexes is better  on 2,78 %, what  base a method with the use of neural network. Worked out method at the evaluation of parameter of impulsive noise by value multifractal index it is recommended to apply at values the amounts of failure pixels of impulsive noise of 5% and higher. The results of evaluation of parameter of multiplicative and gauss noise on a multifractal index showed the worked out method, that it is expedient to apply it for  the choice of parameter of rough-down of image

Author Biographies

Марина Вячеславовна Полякова, Odesa national polytechnic university Boulevard of Shevchenko,1, Odesa, Ukraine, 65044

Candidate of engineering sciences, associate professor

Department of the applied mathematics and information technologies in business 

Юрий Владимирович Емец, Odesa national polytechnic university Boulevard of Shevchenko,1, Odesa, Ukraine, 65044

Master's degree, graduate student

Department of the applied mathematics and information technologies in business

References

  1. Lee, I.S. Digital image enhancement and noise filtering by use of local statistics /I. S. Lee // IEEE Trans. on PAMI. – 1980. – v. 2, № 3. – P. 165 – 168.
  2. Kuan, D.T. Adaptive noise smoothing filter for images with signal dependent noise / D. T.Kuan. // IEEE Trans. on PAMI. – 1985. – v. 7, № 3. – P. 165 – 177.
  3. Canny, I. A computational approach to edge detection /I. Canny // IEEE Trans. on PAMI. – 1986. – v. 8, № 6. – P. 679 – 693.
  4. Deriche, R. Using Canny’s criteria to derive an optimal edge detector recursively implemented /R. Deriche // International Gournal on Computer Vision. – 1987. – v. 1. – P. 167 – 187.
  5. Kundur, D. Blinol image deconvolution/D.Kundur, D. Hatzinakos // IEEE signal Processing Magazine. – 1996. – v. 13(3), № 5. – P. 43 – 64.
  6. Davis, A.Multifractal characterizations of nonstationary and intermittency in geophysical fields: observed, retrieved or simulated / A. Davis, A. Marshak, W. Wiscombe, R. Cahalan. // Journal of Geophysical Research. – 1994. – V. 99, № D4. – P. 8055 – 8072.
  7. Дискриминантный анализ [Электронный ресурс] / Москва. Лаборатория рекламы и маркетинга и PublicRelations.–Режим доступа: WWW/ URL: http://http://www.advlab.ru/articles/article531.htm/ – 2006 г.
  8. Айвазян, С.А. Прикладная статистика: исследование зависимостей / С.А. Айвазян, И.С. Енюков, Л.Д. Мешалкин // Финансы и статистика, 1985. – 524 с.
  9. Santhanam, T. A novel approach to classify noises in images using artificial neural network /T.Santhanam, S. Radhika // J. of Computer Science. – 2010. – Vol.6, № 5. – P. 506 – 510.
  10. Крылов, В.Н. Оценивание параметра гауссовского шума на изображении с помощью мультифрактальных показателей / В.Н. Крылов, Н.И. Науменко, М.В. Полякова, Ю.В. Емец // Военный институт Киевского национального университета имени Тараса Шевченко. – К.: ВИКНУ, 2009. – Вып. № 24.– С. 127 – 134.
  11. Полякова, М.В. Оценивание параметра импульсного шума на изображении с помощью мультифрактальных показателей / М.В. Полякова, В.Н. Крылов, Ю.В. Емец // Методы и системы оптико-электронной и цифровой обработки изображений и сигналов. – Винница, 2010. – № 1 (19). – С. 5 – 13.
  12. Емец, Ю.В. Оценивание параметра мультипликативного шума на изображении с помощью мультифрактальных показателей / Ю.В. Емец, // Искусственный интеллект. – Донецк, 2010. – № 4. – С. 212 – 220.
  13. Lee, I.S. (1980). Digital image enhancement and noise filtering by use of local statistics. IEEE Trans. on PAMI, 2(3), 165 – 168.
  14. Kuan, D. T. (1985). Adaptive noise smoothing filter for images with signal dependent noise. IEEE Trans. on PAMI, 7(3), 165 – 177.
  15. Canny, I. (1986). A computational approach to edge detection. IEEE Trans. on PAMI, 8(6), 679 – 693.
  16. Deriche, R. (1987). Using Canny’s criteria to derive an optimal edge detector recursively implemented. International Gournal on Computer Vision, 1, 167 – 187.
  17. Kundur, D., Hatzinakos, D. (1996). Blinol image deconvolution. IEEE signal Processing Magazine, 13(3), 43 – 64.
  18. Davis, A., Marshak, A., Wiscombe, W., Cahalan, R. (1994). Multifractal characterizations of nonstationary and intermittency in geophysical fields: observed, retrieved or simulated. Journal of Geophysical Research, 99(4), 8055 – 8072.
  19. Discriminant analysis [Electronic resource]. (2006). Moscow. A laboratory of advertisement and marketing and Public Relations. - is access Mode : of WWW/ URL : http:// http://www.advlab.ru/articles/article531.htm.
  20. Aivozan, С. А., Enukov, I. C., Mechalkin, L. D. (1985). Applied statistics: research of dependences. Finances and statistics, 524.
  21. Santhanam, T., Radnika, S. (2010). A novel approach to classify noises in images using artificial neural network. Journal of Computer Science, 1.6(5), 506 – 510.
  22. Krulov, V.N., Naymenko, N.I., Polyakova, M.V., Emets, Y.V. (2009). Evaluation of parameter of Gausse noise dark-and-light by means of multifractal indexes. The Military institute of the Kyiv national university of the name of Taras Shevchenko. Kyiv, 24, 127 – 134.
  23. Polyakova, M.V., Krulov, V.N., Emets, Y.V. (2010). Evaluation of parameter of impulsive noise dark-and-light by means of multifractal indexes. Methods and system optical-electronic and digital processing of images and signal.Vinnytsya, 1(19), 5 – 13.
  24. Emets, Y.V. (2010). Evaluation of parameter of multiplicative noise dark-and-light by means of multifractal indexes. Artificial intelligence. Donetsk, 4, 212 – 220.

Published

2013-06-18

How to Cite

Полякова, М. В., & Емец, Ю. В. (2013). Automated identification of type and evaluation of noise parameter with multifractal indices. Eastern-European Journal of Enterprise Technologies, 3(9(63), 13–17. https://doi.org/10.15587/1729-4061.2013.14842

Issue

Section

Information and controlling system