2016 Second International Conference on Research in Computational Intelligence and Communication Networks (ICRCICN 2016)

28 / 2016-07-21 00:49:25

Real time median filter for Airborne applications system

Median,Noise

This paper proposes an algorithm to
Automatically detect and correct salt and pepper noise
in an image viewed in grey-scale mode, using a median
filter. The main aim of the filtering action is to remove
these unwanted corrupt pixels and restore the
detailing of the image. The algorithm becomes highly
useful during in air-surveillance system. The proposed
algorithm unlike the previous approaches, not only
increases computation efficiency but also saves
memory upto 50%. The median-module method so
proposed is highly optimised and customised to
execute on TMS320F2812 as well as TMS320F28335
DSP based video platform, in real time. It has been
verified for videos being captured according to the
PAL Standards.