Random Walks for Image Segmentation

@article{Grady2006RandomWF,
  title={Random Walks for Image Segmentation},
  author={Leo J. Grady},
  journal={IEEE Transactions on Pattern Analysis and Machine Intelligence},
  year={2006},
  volume={28},
  pages={1768-1783}
}
  • L. Grady
  • Published 1 November 2006
  • Mathematics
  • IEEE Transactions on Pattern Analysis and Machine Intelligence
A novel method is proposed for performing multilabel, interactive image segmentation. Given a small number of pixels with user-defined (or predefined) labels, one can analytically and quickly determine the probability that a random walker starting at each unlabeled pixel will first reach one of the prelabeled pixels. By assigning each pixel to the label for which the greatest probability is calculated, a high-quality image segmentation may be obtained. Theoretical properties of this algorithm… 

Figures from this paper

Random walker watersheds: A new image segmentation approach

This work proposes a new graph-based approach for performing a multilabel, interactive image segmentation using the principle of random walks, and presents quantitative and qualitative results, comparing the new algorithm with the original random walker image segmentsation algorithm.

An Improved Random Walk Algorithm for Interactive Image Segmentation

The random walk algorithm is improved in two aspects: (1) label prior is taken into account when computing edge weights between adjacent pixels; (2) each unseeded pixel is assigned with the same label as the seed with maximum first arrival probability to reduce the bias effect of seed size.

A Seeded Image Segmentation Framework Unifying Graph Cuts And Random Walker Which Yields A New Algorithm

  • A. SinopL. Grady
  • Computer Science
    2007 IEEE 11th International Conference on Computer Vision
  • 2007
This work explores the segmentation algorithm defined by an linfin norm, provides a method for the optimization and shows that the resulting algorithm produces an accurate segmentation that demonstrates greater stability with respect to the number of seeds employed than either the graph cuts or random walker methods.

Interactive image segmentation using probabilistic hypergraphs

Spatial Sampling for Image Segmentation

A novel framework for image segmentation based on the maximum likelihood estimator, which allows for efficient segmentation methods able to deal with textured images and is naturally extended for combining different features.

Enhancing a diffusion algorithm for 4D image segmentation using local information

An interactively adjusted active contours slice by slice method considering local information, where one starts with one manually labeled slice and moves forward in any direction, which is superior with respect to accuracy towards the diffusion algorithm but inferior in the amount of tedious manual processing steps.

Image segmentation using Scale-Space Random Walks

An extension to Random Walks is proposed that improves its performance without significantly modifying the original algorithm, and is able to produce more accurate segmentations in the presence of noise while still retaining all of the properties of the original algorithms.

Semi-supervised image segmentation combining SSFCM and Random Walks

A semi-supervised image segmentation algorithm to segment the noisy image that includes a large amount of objects with the same color features and reduces the noise sensitivity of SSFCM but also avoids cumbersome operations that the user labels the seed points of all objects for Random Walks.

An improved image graph for semi-automatic segmentation

  • D. Freedman
  • Computer Science, Mathematics
    Signal Image Video Process.
  • 2012
This work proposes two improvements to the image graph used by the Random Walker method, including a new way of computing the edge weights, and substitutes a smaller, irregular graph based on Mean Shift oversegmentation.
...

References

SHOWING 1-10 OF 83 REFERENCES

Multilabel random walker image segmentation using prior models

  • L. Grady
  • Computer Science
    2005 IEEE Computer Society Conference on Computer Vision and Pattern Recognition (CVPR'05)
  • 2005
It is shown that incorporation of a nonparametric probability density model allows for an extended random walkers algorithm that can locate disconnected objects and does not require user-specified labels.

Multi-label Image Segmentation for Medical Applications Based on Graph-Theoretic Electrical Potentials

A novel method is proposed for performing multi-label, semi-automated image segmentation using combinatorial analogues of standard operators and principles from continuous potential theory, allowing it to be applied in arbitrary dimension.

A Seeded Image Segmentation Framework Unifying Graph Cuts And Random Walker Which Yields A New Algorithm

  • A. SinopL. Grady
  • Computer Science
    2007 IEEE 11th International Conference on Computer Vision
  • 2007
This work explores the segmentation algorithm defined by an linfin norm, provides a method for the optimization and shows that the resulting algorithm produces an accurate segmentation that demonstrates greater stability with respect to the number of seeds employed than either the graph cuts or random walker methods.

A New Algorithm for Energy Minimization with Discontinuities

This paper considers a natural class of energy functions that permits discontinuities, and develops a new approximation algorithm based on graph cuts that produces a local minimum with respect to a certain move space.

Interactive graph cuts for optimal boundary & region segmentation of objects in N-D images

  • Yuri BoykovM. Jolly
  • Computer Science
    Proceedings Eighth IEEE International Conference on Computer Vision. ICCV 2001
  • 2001
A new technique for general purpose interactive segmentation of N-dimensional images where the user marks certain pixels as "object" or "background" to provide hard constraints for segmentation.

Interactive Image Segmentation Using an Adaptive GMMRF Model

Estimation is performed by solving a graph cut problem for which very efficient algorithms have recently been developed, however the model depends on parameters which must be set by hand and the aim of this work is for those constants to be learned from image data.

"GrabCut": interactive foreground extraction using iterated graph cuts

A more powerful, iterative version of the optimisation of the graph-cut approach is developed and the power of the iterative algorithm is used to simplify substantially the user interaction needed for a given quality of result.

Image segmentation by data driven Markov chain Monte Carlo

  • Z. TuSong-Chun ZhuH. Shum
  • Computer Science
    Proceedings Eighth IEEE International Conference on Computer Vision. ICCV 2001
  • 2001
The DDM-CMC paradigm provides a unifying framework where the role of existing segmentation algorithms, such as; edge detection, clustering, region growing, split-merge, SNAKEs, region competition, are revealed as either realizing Markov chain dynamics or computing importance proposal probabilities.

Scale-Space and Edge Detection Using Anisotropic Diffusion

A new definition of scale-space is suggested, and a class of algorithms used to realize a diffusion process is introduced, chosen to vary spatially in such a way as to encourage intra Region smoothing rather than interregion smoothing.

Normalized cuts and image segmentation

  • Jianbo ShiJ. Malik
  • Computer Science
    Proceedings of IEEE Computer Society Conference on Computer Vision and Pattern Recognition
  • 1997
This work treats image segmentation as a graph partitioning problem and proposes a novel global criterion, the normalized cut, for segmenting the graph, which measures both the total dissimilarity between the different groups as well as the total similarity within the groups.
...