Segmentation of partially-blurred images using wavelet transform

2013;
: pp. 118 – 122
Authors: 
Piontko N., Karpinski M.

N. Piontko1, M. Karpinski2

  1. Ternopil Ivan Pul’uj National Technical University (Ukraine).
  2. Akademia Techniczno-Humanistyczna w Bielsku-Białej (Poland).

In the research a method for segmentation of partially-blurred images using the wavelet transformation, particularly coiflet of the order L=3 is presented. The entropy is used as a segmentation criterion based on wallet transformation. K-means method is used for image segmentation. Developed method was tested and has shown good results of his work; it correctly performs more than 82 % of pixels of image, and in many individual cases, more than 90 %.

1. Gonzalez R. D igital Image Processing (2nd Edition) / R. Gonzalez, R. W oods. – P rentice Hall, 2002. – P. 793. – ISBN 978-0201180756. 2. Методы компьютерной обработки изображений / [Гашников М.В., Глумов Н.И., Ильясова Н.Ю. и др.]; под ред. В.А. Сойфера. – [2-e изд.]. – М.: Физматлит, 2003. – 784 с. 3. Liu R. Image Parti al Blur Detec tion and Class ification / R. Liu, Z. Li , J. Jia // IEEE Conference on Computer Vision and P attern Recognition (CVPR), 2008. 4. S u B. B lurred Image Region Detection and Classification / B. Su // ACM Multimedia. – 2011. – P. 1397-1400. 5. Tong H. Blur detection for digital images using wavelet transform / H. Tong / / Multimedia and Expo, 2004. ICME '04. 2004 I EEE International Conference on. 6. Anchuri A. I mage Bl ur Metrics / A. A nchuri // Applied Vision and Imaging Systems, Standford University, PSYCH 221. – 2011 – P.1 – 8. 7. Mallat S. Singularity Detection and Processing with Wawelet // S. Mallat, W. L. Hwang // IEEE Trans. On Information Theory. – March 1992 – P.617- 643. 8. Daubechies I., Ten Lectures on Wavelets / I. Dau bechies – SIAM: Society for Industrial and Applied Mathematics, 1992 – P. 377. – ISBN 978-0898712742. 9. Forsyth D . Computer Vision: A Modern Approach (2nd Edition) / D. Forsyth, J . Ponce – Prentice Hall, 2011. – P. 792. – ISBN 978-0136085928.