8 September 2011 Removing impulsive noise for infrared image with difference images and adaptive filter
Author Affiliations +
In order to overcome the disadvantage of traditional median filtering method in removing impulsive noise, according to the characteristics of impulsive noise image, a novel approach based on difference images and adaptive filter with multiple windows is proposed. Firstly, utilize the noise image to obtain three difference images, while the difference images are computing in three different directions which are level, vertical, diagonal (counterclockwise rotation of 45° from the horizontal direction), and then get the position of impulse noise points by detecting these three difference image with setting different threshold. Secondly, handle the detected impulse noise point using adaptive filter with multiple windows. Finally, Peak Signal-to-Noise Ratio is used to assess ability of this approach in removing noise. Experimental results indicate that when the, the effect of image restoration by this method can be increased by 8dB compared with median filter. Advantage in removing noise but preserving detail is obvious compared with conventional median filtering method.
© (2011) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
Ju-feng Zhao, Ju-feng Zhao, Hua-jun Feng, Hua-jun Feng, Zhi-hai Xu, Zhi-hai Xu, Qi Li, Qi Li, } "Removing impulsive noise for infrared image with difference images and adaptive filter", Proc. SPIE 8193, International Symposium on Photoelectronic Detection and Imaging 2011: Advances in Infrared Imaging and Applications, 81931E (8 September 2011); doi: 10.1117/12.899851; https://doi.org/10.1117/12.899851

Back to Top