Segmentation of fuzzy images: a novel and fast two-step pseudo MAP method

Wang, Zhen Zhou; Zhang, Yuming

doi:10.1007/s00138-012-0441-5

Cited by 4 publications

(11 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For example, it estimates θx as {0. 18 [6], an iterative method based on the estimated parameter set is used to converge the segmentation result by EM. In this research work, the method in [6] is modified as a gradient descent iterative (GDI) method that iterates according to the following equation until it converges:…”

Section: Image Analysismentioning

confidence: 99%

“…where π m is the probability that the class label takes on the value m. The distribution of the entire observed image sequence is {Y n , 0 < n < MN + 1}, formulated as follows: In [4], [5], [6], and [9], the EM algorithm was used to find the parameter set θy = μ 0 , σ 2 0 , . .…”

Section: Image Analysismentioning

confidence: 99%

“…Due to the variety and complexity of the images captured in the real world, robust segmentation of the objects from the background remains a bottleneck for many types of images, e.g., segmentation of the remote sensing images that are usually characterized as noisy or textured. Much of the past research work has been directed at stochastic modeling methods [1], [2], [3], [4], [5], [6], [7], [8], [9], [10], [11], [12]. Usually, the stochastic modeling methods assume that the image is a Markov random field (MRF) which complies with Gibbs distribution [3], [4].…”

Section: Introductionmentioning

confidence: 99%

“…Thus, it belongs to the local property of the image. Markov chain Monte Carlo (MCMC) [6], [13] is a popular way to estimate the parameters of the Gibbs distribution. Unfortunately, it does not converge to correct labeling in most cases for the segmentation.…”

Section: Introductionmentioning

confidence: 99%

“…To deal with the intractability of computation complexity, different methods have been proposed, e.g. iterated conditional modes (ICM) [3], dynamic programming [4], and maximization of local energy and global energy [6]. Unfortunately, none of these methods has achieved satisfying accuracy and computation complexity in segmenting noisy or textured images.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

A New Approach for Robust Segmentation of the Noisy or Textured Images

Wang¹

2016

SIAM J. Imaging Sci.

Self Cite

View full text Add to dashboard Cite

Abstract. Segmentation of noisy or textured images remains challenging in both accuracy and computational efficiency. In this paper, we propose a new approach for segmentation of noisy or textured images that exist widely in real life. The proposed approach finds the mean values of different pixel classes more efficiently and accurately than the benchmark expectation maximization (EM) and K-means methods. With these mean values, the segmentation is achieved by clustering the pixels to its nearest mean. When too much noise is left for the presegmentation result or when textured objects are involved, we propose transforming the density distribution of labeled pixels into grayscale distribution by down-sampling the image with a bicubic function. An optimal threshold is automatically selected from the slope difference distribution of the histogram for the final segmentation. The extracted boundary is then refined by an energy minimization function with the detected edges when enough clear edges can be obtained. A large variety of images are used to validate the proposed approach, and the results verify its effectiveness in segmenting both noisy and textured images.

show abstract

Section: Image Analysismentioning

confidence: 99%