Comparative performance study of various filters on high density impulse noise
Jeet Kumar · 2017
The paper compares the performance of various filters of different sizes on impulse noise in spatial domain. In this paper median and adaptive median filters are used from non-linear category. For a low density impulse noise, smaller size median filter is required while for a high density impulse noise, either a larger size median filter is required or a series of smaller size median filter is required. The circular averaging filter also performs well in this case but at the expense of more blurring. Moreover alpha trimmed mean filter works well with a moderate density impulse noise. Finally the adaptive median filter works best on impulse noise. The adaptive median filter can recover the image even if the image is not visible in case of very high density impulse noise. The paper demonstrates some novel procedures involving a series of filters to handle high density impulse noise.