Double Gaussian mixture model for image segmentation with spatial relationships☆
Introduction
The goal of image segmentation [1] is to label image pixels and divide them into clusters according to similarity of attributes. Image segmentation has been widely applied to image processing, video surveillance systems and medical image analysis. Because of its widespread applications, image segmentation is receiving greater attention. In recent years, many different algorithms for image segmentation have been developed. Among these algorithms, representative techniques include region splitting and merging [2], [3], normalized cut [4], [5], mean shift [6], [7], and level sets [8], [9], [10]. Statistical algorithms based on clustering have also been successfully applied to image segmentation.
Among the statistical models, the finite mixture model (FMM) [11] is receiving increasing attention because of its simple form and ease of implementation. The FMM comes in several different forms. Two representative FMMs are the Gaussian mixture model (GMM) [12], [13] and Student’s-t mixture model (SMM) [14], [15]. The component functions for these models are the Gaussian distribution and Student’s t-distribution, respectively. The expectation maximization (EM) [16], [17] algorithm is generally used to infer the parameters of these models. The GMM and SMM obtain good segmentation results when applied to image segmentation [13], [15].
However, experiments show that FMM is not robust against noise, with unsatisfactory segmentation results obtained when images are degraded by noise. The experiments show that FMM is not robust against noise. The main cause of this is that FMM supposes that the pixels are statistically independent and so the spatial information of pixels is not taken into account.
To overcome the aforementioned drawback and improve the quality of image segmentation for FMM, the Markov random field (MRF) [18], [19] model which incorporates the spatial relationships between image pixels has been proposed. MRF models have been widely applied in image processing [20], [21] and image segmentation [22]. The MRF models are divided into two types according to the methods used to incorporate spatial information. One approach imposes spatial information on pixel labels [22], [23]. These MRF models obtain better segmentation results than FMM because of the spatial information, but this also increases the burden of computation cost for the MRF model. To improve the computational efficiency of MRF model, a Bethe approximation [24] is used to approximate the Gibbs free energy function. However, the computational cost in [24] still remains very high because of the complex object function.
Another type of MRF, the spatially variant FMM (SVFMM) was proposed in [25]. In this model, the contextual mixing proportion is assumed to be a random variable upon which a spatial smoothness prior is imposed. The EM algorithm is used to infer the parameters of the model. The model obtains better segmentation results than FMM. However, the contextual mixing proportion cannot be obtained in closed form in the M-step of the EM algorithm. In other words, the contextual mixing proportion is a probabilistic vector and must be nonnegative and satisfy the constraints . To obtain a closed form solution for , the gradient projection algorithm is added in the M-step [25]. Blekas et al. [26] improved the SVFMM proposed in [25] using convex quadratic programming instead of gradient projection. The primary drawback of the model in [25], [26] is that the contextual mixing proportion cannot be obtained directly from the given data. To resolve this problem, a prior distribution based on the Gauss–Markov random field is imposed on the contextual mixing proportion [27]. The advantage of this model is that a closed form equation can be obtained by the EM algorithm. To preserve the region boundaries of the image segmentation results, two models were proposed in [28]. The first of these integrates a binary-Bernoulli-distributed line process (BLP) with MRF, and the other incorporates a continuous-Gamma-distributed line process (CLP). However, one drawback of the models in [27], [28] is that the contextual mixing proportion still cannot be obtained directly as a probabilistic vector.
To improve the computational efficiency of the model, a new representation of the contextual mixing proportion was given in [29]. In this model, the contextual mixing proportion is explicitly modelled as a probabilistic vector. Therefore, a closed form for the contextual mixing proportion can be obtained directly. However, this model can only be applied to grayscale image segmentation. A detail-preserving mixture model for image segmentation was proposed in [30]. Different weight values are assigned to different pixels in each pixel’s neighborhood. However, the computational complexity of the model in [30] is still very high because of the complex representation of the log-likelihood function.
To simplify the representation of the model and improve the computational efficiency, we propose a mixture model for image segmentation in this paper. The proposed model is very different from most of the aforementioned models. First, the proposed model is based on GMM. Therefore, it has a simple form and is easy to implement. Second, the proposed model incorporates spatial relationships between the pixels. Therefore, it is more robust against noise than FMM. Third, fewer parameters need to be estimated in the proposed model than in other models based on MRF, which improves the computational efficiency of the proposed model. Finally, the EM algorithm and gradient descent method are used to estimate the parameters of the proposed model directly. Therefore, the inference process is much simpler than in some other models based on MRF.
The remainder of this paper is organized as follows. In Section 2, the theoretical background related to our proposed model is introduced briefly. In Section 3, a detailed description of the proposed model is presented. Experimental results obtained by our model for various synthetic noisy grayscale images and natural color images and some discussions are presented in Section 4. Our conclusions are given in Section 5.
Section snippets
Theoretical background
In this section, the GMM and the mixture model based on MRF, both of which are closely related to our proposed model, are described briefly.
Proposed model
To overcome the drawbacks of mixture models based on MRF discussed above, we present a new GMM in this section. In the proposed model, a Gaussian distribution prior is incorporated into the contextual mixing proportion set Π, and is modelled explicitly as a probabilistic vector, so that the results of are obtained in closed form. For the proposed model, we first define a weight function for the nth pixel for each class as:where is the
Experiments
In this section, the results of comprehensive experiments are presented to quantitatively and visually evaluate the performance of our proposed model. We compare our model with several state-of-the-art image segmentation models. These models are K-means, GMM [12], SVFMM [25], [26], CA-SVFMM [27], CLP and BLP [28], and DPSMM [30]. The source code for K-means, GMM, CA-SVFMM, CLP and BLP can be downloaded from http://www.cs.uoi.gr/sfikas/matlabcode.html. The source code for SVFMM can be downloaded
Conclusions
In this paper, a new GMM is proposed for image segmentation. In the proposed model, the contextual mixing proportion of a pixel effectively incorporates the spatial relationships between pixels. Its representation is closely related to a pixel’s neighborhood system and its form is very simple. At the same time, there are fewer parameters in our model than in many other models based on MRF. Therefore, the proposed model is easy to implement. The computational complexity is reduced because the
Acknowledgments
The authors would like to thank the anonymous reviewers for their thorough and valuable comments and suggestions, which greatly helped to improve the quality of the paper. This work was supported by National Nature Science Foundation of China under Grant Nos. 61322203 and 61332002, and Foundation of Chengdu University of Information Technology (No. KYTZ201426).
References (41)
- et al.
Picture segmentation using a recursive region splitting method
Comput. Graph. Image Process.
(1978) - et al.
Fronts propagating with curvature-dependent speed: algorithms based on Hamilton–Jacobi formulation
J. Comput. Phys.
(1988) - et al.
EM procedures using mean field-like approximations for Markov model-based image segmentation
Pattern Recognit.
(2003) - et al.
Hierarchical Markovian segmentation of multispectral images for the reconstruction of water depth maps
Comput. Vision Image Underst.
(2004) - et al.
EM procedures using mean field-like approximations for Markov model-based image segmentation
Pattern Recognit.
(2003) Computer Vision Algorithms and Applications
(2010)- et al.
Integrating region growing and edge detection
IEEE Trans. Pattern Anal. Mach. Intell.
(1990) - et al.
Normalized cuts and image segmentation
IEEE Trans. Pattern Anal. Mach. Intell.
(2000) - et al.
Shape matching by segmentation averaging
IEEE Trans. Pattern Anal. Mach. Intell.
(2010) Mean shift, mode seeking, and clustering
IEEE Trans. Pattern Anal. Mach. Intell.
(1995)
Mean shift: a robust approach toward feature space analysis
IEEE Trans. Pattern Anal. Mach. Intell.
Active contours without edges
IEEE Trans. Image Process.
Minimization of region-scalable fitting energy for image segmentation
IEEE Trans. Image Process.
Finite Mixture Models
Pattern Recognition and Machine Learning
A finite mixture model for image segmentation
Stat Comput.
Robust mixture modeling using the t distribution
Stat. Comput.
Maximum likelihood from incomplete data via EM algorithm
J. Roy. Stat. Soc.
The EM Algorithm and Extensions
Cited by (16)
A new method for building adaptive Bayesian trees and its application in color image segmentation
2018, Expert Systems with ApplicationsCitation Excerpt :Furthermore, since the number of feature clusters is determined adaptively, the proposed method provides the flexibility of handling Gaussian mixtures with distinct numbers of multivariate Gaussian components for better feature cluster representation. It shall be emphasized that many clustering methods based on Gaussian mixtures estimation (Fu & Wang, 2012; Sefidpour & Bouguila, 2012; Xiong, Zhang, & Yi, 2016) assume a pre-defined number of Gaussian components to represent all feature clusters. Finally, the proposed method is application independent, and can be extended to approach generic non-supervised clustering problems.
Markov random field model and expectation of maximization for images segmentation
2023, Indonesian Journal of Electrical Engineering and Computer ScienceRobust segmentation algorithms for retinal blood vessels, optic disc, and optic cup of retinal images in medical imaging
2022, Big Data Analytics and Machine Intelligence in Biomedical and Health Informatics: Concepts, Methodologies, Tools and ApplicationsA new binary representation method for shape convexity and application to image segmentation
2022, Analysis and Applications
- ☆
This paper has been recommended for acceptance by M.T. Sun.