This Paper presents an efficient algorithm for median filtering with a 3x3 filter kernel with only about 9 comparisons per pixel using spatial coherence between neighbouring filter computations. The basic algorithm calculates two medians in one step and reuses sorted slices of three vertical neighbouring pixels. An extension of this algorithm for 2D spatial coherence is also examined, which cal...

Removal of random value impulse noise is essential to help successive process like image segmentation, analysis, pattern recognition etc. To remove uniform/randomvalued impulse noise without compromising the image detail information such as sharp edges, Modified Directional Weighted Median Filter (MDWMF) is proposed. The proposed filter utilizes the second order difference (SOD) along with a th...

Median filter is widely used to remove impulse noise without blurring sharp edges. However, when noise level increased, or with thin edges, median filter may work poorly. This paper proposes a new filter, which will detect edges along four possible directions, and then replace noise corrupted pixel with estimated noise-free edge median value. Simulations show that the proposed multi-stage direc...

This paper presents a new approach for region-based video mosaicing, treating moving objects separately from the background, and with improved ghost-like noise elimination. The mosaic images show the moving objects superimposed over a stationary background. Conventional technologies can reduce the ghost-like noise that occurs from moving objects by using temporal median filtering, but its effic...

A novel extension of the classical signal -adaptive median lter (SAM) is proposed in this paper. It is the so-called morphological signal -adaptive median lter (MSAM). Two mod-iications are introduced in the SAM lter aiming at: (1) enhancing SAM impulse detection mechanism so that it detects not only impulses of a constant amplitude but randomly-valued impulses as well, (2) employing an anisotr...