Distance mapping for image filtering

Documento adjunto(s)

Anexidades

• AS_14_n2_209_ROB.pdf (2,1M)

Abstract

This paper deals with particle disconnection on a grey-level image. The first approach is to compute a good threshold allowing to obtain two classes of points, one for particles and the other for background. As the particle density is very high on studied images (particles can even cover each other), thresholding does not give good results. The second approach is to compute the gradient image associated with the initial image. All the information about particle boundaries is represented on such an image, but the gradient image is very noisy. Filtering a gradient image in order to determine the particle boundaries is a difficult problem to solve. Classical methods such as thresholding and elimination of small objects have been used but they do not give good results. The proposed method in this paper is based on a combination of two distance maps, one being computed from the thresholded initial image and the other from a particle map. This particle map is obtained by selecting points with high gradient values as points of background on the thresholded image. The particle map and the thresholded image are binary images and a distance map can be computed on each of them. Then, for each point belonging to the background of the particle map (but belonging to a particle on the thresholded image), the corresponding values on the distance maps are checked, in order to decide if the point is noisy or not. This method allows to disconnect most of the particles. This algorithm has been implemented for an application in civil engineering. The aim was to determine the granulometry of a riprap (set of big stones covering an embankment) by image processing. Stone diameters are then computed on the particle map after noise elimination.

Para citar este artículo

Acerca de: Frédérique Robert

Acerca de: Guy Lefebvre