Showing papers by "Fernand Meyer published in 2007"

Micro-viscous morphological operators

Abstract: In most operators of mathematical morphology source and destina- tion are the same: from pixels to pixels. In this paper we present adjunctions where source and destination are not the same. In ad- dition to the pixels of a grid, we also consider the centers of edges linking neighboring pixels. Interesting filters may be constructed using such operators, in particular bi-levelings, where the introduc- tion of some degree of viscosity permits to obtain higher levels of simplifications as with ordinary levelings.

Morphological operators for flooding, leveling and filtering images using graphs

Abstract: We define morphological operators on weighted graphs in order to speed up image transformations such as floodings, levelings and waterfall hierarchies. The image is represented by its region adjacency graph in which the nodes represent the catchment basins of the image and the edges link neighboring regions. The weights of the nodes represent the level of flooding in each catchment basin ; the weights of the edges represent the altitudes of the pass points between adjacent regions.

On the Regularization of the Watershed Transform

Abstract: Publisher Summary The watershed transform is one of the most popular judging from the great diversity of applications in which the method has been successfully applied. Since its introduction, various formulations of the watershed transform have appeared, providing elegant descriptions, revealing its properties, and resolving some difficulties. Since its origins, numerical image processing has developed along three main axes: (1) filtering theory and scale-space analysis, which corresponds to a robust framework for signal characterization, (2) segmentation for a precise identification of the shapes present in a scene, and (3) symbolic representations that seek to numerically restore a description as close as possible to our own vision. With the flat zones approach, the watershed transform is a fully automatic and parameter-free procedure. The most efficient computation method of the watershed transform is based on the hierarchical queues of pixels. This has two major benefits: first, it allows work on a narrow band near the lakes and second, it avoids a threshold by threshold processing.

News from viscousland.

Abstract: Viscous closings have been presented at ISMM'02 as an e cient tool for regularizing the watershed lines in gray-scale images. We consider now the problem of reconnecting several edge portions of a same object. In the binary case, this is very nicely solved via the computation of the distance function to the grains: the downstream of the saddle points reconnects the grains, and is known as the perceptual graph. As a particular case, overlapping particles may be separated by computing the watershed line of the inverse distance function. This paper extends the approach to grey-tone images using the concept of viscous dilations. Finally, combinations of both viscous dilations and viscous closings are proposed for segmenting objects with dotted and irregular contours.