DocumentCode :
2264396
Title :
An Adaptive Morphological Filter Based on Multiple Structure and Multi-Scale Elements
Author :
Xu, Guo-bao ; Su, Zhi-bin ; Wang, Ji ; Yin, Yi-xin ; Shen, Yu-li
Author_Institution :
Inf. Sch., Guangdong Ocean Univ., Zhanjiang
Volume :
2
fYear :
2008
fDate :
20-22 Dec. 2008
Firstpage :
399
Lastpage :
403
Abstract :
To filter out various kinds of noise of different intensities in gray images, a highly adaptive morphological filter based on multiple structure and multi-scale elements was proposed. This algorithm employs a morphology filter based on multiple structure and multi-scale elements to filter the images, followed by image fusion processing with the weights calculated from the image entropies, before finally obtaining the filtered images. The experimental results show that the proposed filtering algorithm, compared with the traditional filtering algorithms using the mean filter, median filter and Wiener filter, has a better adaptability. The new algorithm can effectively filter many kinds of noise such as the salt and pepper noise, Gaussian noise and speckle noise, while preserving more image details.
Keywords :
Gaussian noise; Wiener filters; adaptive filters; image fusion; median filters; speckle; Gaussian noise; Wiener filter; adaptive morphological filter; filtering algorithm; gray images; image entropies; image filter; image fusion processing; mean filter; median filter; multi-scale elements; multiple structure; salt-and-pepper noise; speckle noise; Adaptive filters; Filtering algorithms; Gaussian noise; Image analysis; Image processing; Information technology; Intelligent structures; Morphology; Shape; Wiener filter; Entropy; Image filtering; Mathematical morphology; Multi-scale elements; Multiple structure elements;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Intelligent Information Technology Application, 2008. IITA '08. Second International Symposium on
Conference_Location :
Shanghai
Print_ISBN :
978-0-7695-3497-8
Type :
conf
DOI :
10.1109/IITA.2008.359
Filename :
4739794
Link To Document :
بازگشت