Abstract

We have to construct a Stacked Adaptive Median Filter for removing Salt & Pepper Noise. Stack Filter is used for easy implementation of VLSI design. In Stack Filter we are having two properties they are
1. Threshold decomposition and the stacking property
2. Ordering property.

Motivated by the success of rank-order filters in a wide variety of applications and by the case with which they can now be implemented, we consider in this paper a new class of filters called stack filters. They share the threshold decomposition and stacking properties of rank ordered filters.

Before decomposing the image we have to find the corrupted and uncorrupted pixels in that image and set the flag 0 if the pixel is uncorrupted, set the flag 1 if the pixel is corrupted. After setting the flag we have to decompose the image using threshold decomposition technique.

In those threshold binary images we have to check if the pixel is uncorrupted or not, if it is uncorrupted that value is given to the output. If the pixel is corrupted then see the neighboring pixels, in that if there are any uncorrupted pixels then find the median of that pixels that will be placed and change the flag to 0 other wise there is no uncorrupted pixels on the neighboring pixels increase the window size by 2 and check the size of the window is less than the size. It means that we have to check the uncorrupted pixels other wise replace the old value and set the flag to 1.This process will be continued until all the pixels becomes true.

In this way we have to do this process for all the binary images. After getting all the results of the binary images we get the decimal result. In Stack Filter each binary image is unrelated so we have to do the filtering operation in parallel. So we do the higher level of parallelism in stack filter, the binary median filter is easier than the normal median filter

<< back