Fast degree of aggregation filter for reducing impulse noise in satellite images
Srinivasa Rao Gantenapalli, Praveen B. Choppala, Vandana Gullipalli, James Stephen Meka, Paul D. Teal · 2022 IEEE 6th Conference on Information and Communication Technology (CICT) · 2022
The interest of this paper is the reduction of impulse noise in satellite images. The vector median filters are by far the most popular filters for reducing impulse noise in images. A recent variant of this class of filters is the degree of aggregation filter that operates by associating the degree of aggregation as the weight of a pixel in representing the signal component in the image. This approach can lead to improved filtering as pixels that are more similar obtain higher weights and vice versa. However the problem in this approach is that filtering should be performed over all the pixels in a sequential fashion thus leading to high computational complexity. This article improves on this filter and proposes a fast degree of association algorithm. The efficacy of the proposal is measured using a combined measure of time and accuracy.