The EM/MPM algorithm for segmentation of textured images: analysis and further experimental results

Comer, Mary L.; Delp, Edward J.

doi:10.1109/icip.1996.560955

Cited by 25 publications

(41 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Unlike previous approaches for incorporating Markov priors into the EM framework, SCEM does not limit the type of Markov model (as in [5]) nor the parameters that can be estimated (as in [4]). On a final note, we should mention that SCEM is not specific to nuclear segmentation, or even image segmentation, but instead is applicable to any task requiring the unsupervised estimation of MRF models.…”

Section: Resultsmentioning

confidence: 99%

“…Equation (2) follows from changing the order of summation and then summing out the superfluous variables. Note that the first term in (2) is the formulation proposed by Comer and Delp [4]. It is the second term that allows us to estimate the MRF parameters during the SCEM iteration.…”

Section: Derivation Of Spatially Constrained Expectation Maximizationmentioning

confidence: 99%

“…In an effort to maintain tractability, authors have invoked simplifying assumptions; unfortunately, these assumptions tended to reduce the generality of the solutions. For example, Comer and Delp [4] demanded that the Markov prior remain constant during the EM iteration; Nikou et al [5] imposed Gaussianity constraints; Marroquin [6] tied the implementation to a predetermined cost function, thus linking estimation (i.e. model fitting) and classification.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Image Segmentation with Implicit Color Standardization Using Spatially Constrained Expectation Maximization: Detection of Nuclei

Monaco

Hipp

Lucas

et al. 2012

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Abstract. Color nonstandardness -the propensity for similar objects to exhibit different color properties across images -poses a significant problem in the computerized analysis of histopathology. Though many papers propose means for improving color constancy, the vast majority assume image formation via reflective light instead of light transmission as in microscopy, and thus are inappropriate for histological analysis. Previously, we presented a novel Bayesian color segmentation algorithm for histological images that is highly robust to color nonstandardness; this algorithm employed the expectation maximization (EM) algorithm to dynamically estimate -for each individual image -the probability density functions that describe the colors of salient objects. However, our approach, like most EM-based algorithms, ignored important spatial constraints, such as those modeled by Markov random field (MRFs). Addressing this deficiency, we now present spatially-constrained EM (SCEM), a novel approach for incorporating Markov priors into the EM framework. With respect to our segmentation system, we replace EM with SCEM and then assess its improved ability to segment nuclei in H&E stained histopathology. Segmentation performance is evaluated over seven (nearly) identical sections of gastrointestinal tissue stained using different protocols (simulating severe color nonstandardness). Over this dataset, our system identifies nuclear regions with an area under the receiver operator characteristic curve (AUC) of 0.838. If we disregard spatial constraints, the AUC drops to 0.748.

show abstract

Section: Resultsmentioning

confidence: 99%

Section: Derivation Of Spatially Constrained Expectation Maximizationmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Image Segmentation with Implicit Color Standardization Using Spatially Constrained Expectation Maximization: Detection of Nuclei

Monaco

Hipp

Lucas

et al. 2012

Lecture Notes in Computer Science

View full text Add to dashboard Cite

show abstract

“…We now compare the proposed method with three other segmentation techniques: a texture-based, Markovian Random Field (MRF) [6], a traditional The lack of topological control in both techniques yields a great amount of small contours, many of which are of no anatomical meaning. In order to obtain the same behaviour as the one provided by the topological control, segmentation by region based techniques such as MRF and K-means would require some post-processing, which, in general, is highly user dependent.…”

Section: Application To Object Detection In Gray Scale Imagesmentioning

confidence: 99%

“…This edge-based segmentation process does not exhibit topology control, but ensures that a single, closed contour with no holes is obtained from an image. Traditional image processing and segmentation techniques based on edges or regions [3,11,4,22]; pattern recognition methods [23,5,13] and model-based such as Markovian Random Fields (MRF) and Fractals [6,25,19] do not exploit topological properties either.…”

Section: Related Workmentioning

confidence: 99%

Topological triangle characterization with application to object detection from images

Nonato

Lizier

Batista

et al. 2008

Image and Vision Computing

View full text Add to dashboard Cite

Estimating the number of pure chemical components in a mixture by maximum likelihood

Levina

Wagaman

Callender

et al. 2007

Journal of Chemometrics

View full text Add to dashboard Cite

This paper addresses the problem of determining the number of pure chemical components in a mixture by applying the maximum likelihood estimator (MLE) of intrinsic dimension. The application here is to Raman spectroscopy data, although the method is general and can be applied to any type of data from a chemical mixture. We show that the MLE produces superior results compared to other methods on both simulated and real chemical mixtures, and is accurate even when minor components are present. Even if the signal-to-noise (SN) ratio is very low, accurate estimates can still be obtained by smoothing the data before applying the estimator, this approach is illustrated on two real datasets with high noise levels. Since the MLE is computed locally at every data point, we also show how the local estimates can be used for other applications, such as segmenting the specimen into homogeneous regions.

show abstract

The EM/MPM algorithm for segmentation of textured images: analysis and further experimental results

Cited by 25 publications

References 16 publications

Image Segmentation with Implicit Color Standardization Using Spatially Constrained Expectation Maximization: Detection of Nuclei

Image Segmentation with Implicit Color Standardization Using Spatially Constrained Expectation Maximization: Detection of Nuclei

Topological triangle characterization with application to object detection from images

Estimating the number of pure chemical components in a mixture by maximum likelihood

Contact Info

Product

Resources

About