Evolutionary Tree-Structured Filter for Impulse Noise Removal
A new evolutionary approach for construction of uniform impulse noise filter is presented. Genetic programming is used for combining the basic image transformations and filters into tree structure, which can accurately estimate noise map. Proposed detector is employed for building switching-scheme filter, where recursively implemented α-trimmed mean is used as the estimator of corrupted pixel values. The proposed evolutionary filtering structure shows very good results in removal of uniform impulse noise, for wide range of noise probabilities and different test images.
KeywordsNoisy Image Impulse Noise Primitive Function Noisy Pixel Corrupted Pixel
Unable to display preview. Download preview PDF.
- 1.Gonzales, C., Woods, E.: Digital Image Processing. Prentice-Hall, Englewood Cliffs (2002)Google Scholar
- 8.Crnojevic, V., Senk, V., Trpovski, Z.: Advanced Impulse Detection Based on Pixel-Wise MAD. IEEE Signal processing letters 11(7) (July 2004)Google Scholar
- 12.Aoki, S., Nagao, T.: Automatic Construction of Tree-structural Image Transformation using Genetic Programming. In: Proceedings of the 1999 International Conference on Image Processing (ICIP 1999), vol. 1, pp. 529–533 (1999)Google Scholar