Random-Valued Impulse Noise Reduction in Medical Images by Using Double-Checked Fast Adaptive Median Filter

Median-filters have been used for a very long time as an influential non-linear technique to removal of impulse noise. There have been many suggestions regarding the filters that can get rid of this noise while keeping the necessary details. However, these filters have one thing in common: They are only effective at low noise rate and thus are practical only for low-noise in image. Another thing that is also obvious for these filters is that they produce a blur and blotch that is undesired in high-noise in image. This method is fast, simple, and additionally there is no need to adjust it for different applications. Furthermore, the method can remove the impulse noise effectively from the image, and at the same time can preserve the details in the image, even when the input image is very highly corrupted by the impulse noise. The presented method can be performed in noise removal in medical images and real time image processing. Keyword: Impulse noise; Median-based filters, Digital images, Noise detector.