Non-linear Diffusion for Interactive Multi-scale Watershed Segmentation

Dam, Erik B.; Nielsen, Mads

doi:10.1007/978-3-540-40899-4_22

Cited by 14 publications

(8 citation statements)

References 27 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Although the edge and region approaches are considered to yield good results, some researchers have argued the impossibility of extracting complete information of either aspect of the image independent of the other and have proposed hybrid approaches which integrate boundary and region information (Moigne & Tilton, 1995;Pal & Pal, 1993). One example of these approaches is the watershed algorithm which sees an image as a topographic surface and, after identifying maxima (ridges) and minima (valleys), attempts 'flooding' the terrain to obtain catchment regions (Dam, 2000).…”

Section: Image Segmentation In Computer Visionmentioning

confidence: 99%

Segmentation of Remotely Sensed Imagery: Moving from Sharp Objects to Fuzzy Regions

Lizarazo¹,

Elsner²

2011

Image Segmentation

View full text Add to dashboard Cite

Section: Image Segmentation In Computer Visionmentioning

confidence: 99%

Segmentation of Remotely Sensed Imagery: Moving from Sharp Objects to Fuzzy Regions

Lizarazo¹,

Elsner²

2011

Image Segmentation

View full text Add to dashboard Cite

“…The watershed transform has been widely used in many fields of image processing, including medical image segmentation, due to the number of advantages that it possesses: it is a simple intuitive method, it is fast and can be parallelized (in [2], an almost linear speedup was reported for a number of processors up to 64) and it produces a complete division of the image in separated regions even if the contrast is poor, thus avoiding the need for any kind of contour joining. Furthermore, several researchers have proposed techniques to ember the watershed transform in a multiscale framework, thus providing the advantages of these representations [3][4][5]. Some important draw backs also exist, and they have been widely treated in the related literature.…”

Section: Introductionmentioning

confidence: 98%

A New Segmentation Method using Watersheds on Grey Level Images

Rao

Kumar²,

Nagaraju³

2006

2006 International Conference on Advanced Computing and Communications

View full text Add to dashboard Cite

Segmentation is one of the most challenging tasks in image processing as it requires to some extent a semantic understanding of the image.The morphological segmentation paradigm, based on the watershed transform and markers has been extremely successful, both for interactive as for automatic segmentation. Though many algorithms were proposed for watersheds in the literature, but most of them are based on binary images, for finding water catchments basins. In the present study a new method of finding watersheds on grey level image is proposed on 5 X 5 and 3 X 3 masks. To count the number of watersheds on grey levels, a new scheme is proposed for selecting the height of the peak points from where water starts falling, to the central point. The algorithm is applied on human images and textures and a good segmentation is resulted. The present study compares the proposed segmentation scheme with number of watersheds taken into account on different mask sizes.

show abstract

“…Especially in biomedical image applications this is often the case. Therefore those theories were extended to non-linear diffusion processes, see [NVWV97, NVW + 99, DN00]. A drawback of these approaches is that their analysis of scales is not fully automatic and can only be used in a forward approach, thus going from fine to coarse scales and then trying to find a backward relation.…”

Section: Introductionmentioning

confidence: 99%

Multiscale Segmentation via Bregman Distances and Nonlinear Spectral Analysis

Zeune¹,

Dalum²,

Terstappen³

et al. 2017

SIAM J. Imaging Sci.

View full text Add to dashboard Cite

In biomedical imaging reliable segmentation of objects (e.g. from small cells up to large organs) is of fundamental importance for automated medical diagnosis. New approaches for multi-scale segmentation can considerably improve performance in case of natural variations in intensity, size and shape. This paper aims at segmenting objects of interest based on shape contours and automatically finding multiple objects with different scales. The overall strategy of this work is to combine nonlinear segmentation with scales spaces and spectral decompositions. We generalize a variational segmentation model based on total variation using Bregman distances to construct an inverse scale space. This offers the new model to be accomplished by a scale analysis approach based on a spectral decomposition of the total variation. As a result we obtain a very efficient, (nearly) parameter-free multiscale segmentation method that comes with an adaptive regularization parameter choice. To address the variety of shapes and scales present in biomedical imaging we analyze synthetic cases clarifying the role of scale and the relationship of Wulff shapes and eigenfunctions. To underline the potential of our approach and to show its wide applicability we address three different experimental biomedical applications. In particular, we demonstrate the added benefit for identifying and classifying circulating tumor cells and present interesting results for network analysis in retina imaging. Due to the nature of nonlinear diffusion underlying, the mathematical concepts in this work offer promising extensions to nonlocal classification problems.

show abstract

Non-linear Diffusion for Interactive Multi-scale Watershed Segmentation

Cited by 14 publications

References 27 publications

Segmentation of Remotely Sensed Imagery: Moving from Sharp Objects to Fuzzy Regions

Segmentation of Remotely Sensed Imagery: Moving from Sharp Objects to Fuzzy Regions

A New Segmentation Method using Watersheds on Grey Level Images

Multiscale Segmentation via Bregman Distances and Nonlinear Spectral Analysis

Contact Info

Product

Resources

About