基于变分高斯混合模型的图像分割算法
PDF下载 (297)张媛媛,钟意伟.基于变分高斯混合模型的图像分割算法[J].宁波大学学报(理工版),2013,27(01):23-28.DOI:
ZHANG Yuan-yuan,ZHONG Yi-wei.Image Segmentation via Variational Mixture of Gaussions[J].Journal of Ningbo University(Natural Science & Engineering Edition),2013,27(01):23-28.DOI:
| Title: | Image Segmentation via Variational Mixture of Gaussions |
| 作者: | 张媛媛, 钟意伟 |
| Author(s): | ZHANG Yuan-yuan, ZHONG Yi-wei |
| 关键词: | 图像分割; 变分推断; 高斯混合模型; 期望最大化算法 |
| Keywords: | image segmentation; variational inference; Gaussian mixture models; expectation-maximization |
| 分类号: | TP391.41 |
| 文献标识码: | A |
| 摘要: | 提出了一种基于变分推断的高斯混合模型的图像分割算法. 该算法首先用贝叶斯混合高斯模型对图像的特征进行建模, 并针对模型的参数学习问题, 利用变分推断算法估计模型的参数及其后验概率; 这种方法比采样法的计算量更少, 而且能够根据图像数据自动优化混合个数, 实现了模型的自动选择. 最后, 该算法在Berkeley的自然图像集上进行的实验结果与经典的图像分割算法进行了比较, 结果表明此方法得到的图像分割结果精度较高, 具有较好的性能. |
| Abstract: | Gaussian mixture model (GMM) has been effectively used in image segmentation. In this case, the features of an image are described by a mixture model with different components. However, how to choose the number of mixture components and estimate model parameters are still short of solutions. Current algorithms such as maximum likelihood and sampling methods are known for their own limitations. So we present an alternative algorithm based on Bayesian variational method and apply it in image segmentation. This method works at less computational cost than sampling methods, and can also naturally handle the model selection problem. In the model’s iterative process, the algorithm can automatically determine the number of mixture components in view of the data collected. By comparing our method against other classical segmentation methods on natural images acquired from Berkeley Segmentation Data Set, it suggests that our method provides better performance on image segmentation. |
| 参考文献 /References: | [1] Yang A Y, Wright J, Ma Yi, et al. Unsupervised segmentation of natural images via lossy data com- pression[J]. Computer Vision and Image Understanding, 2008, 110(2):212-225.[2] Haim P, Joseph F, Ian J. A study of Gaussian mixture models of color and texture features for image classi- fication and segmentation[J]. Pattern Recognition, 2006, 39(4):695-706.[3] Konstantinos B, Aristidis L, Galatsanos N P, et al. A spatially constrained mixture model for image segmen- tation[J]. Neural Networks, 2005, 16(2):494-498.[4] Omer R, Hayit G, Jacob G. Combining region and edge cues for image segmentation in a probabilistic Gaussian mixture framework[C]. Computer Vision and Pattern Recognition, MN: Minneapolis, 2007. [5] Bishop C M. Pattern recognition and machine learning [M]. New York: Springer Presses, 2007.[6] Constantinos C, Titsias M K, Aristidis L. Bayesian feature and model selection for Gaussian mixture models[J]. Pattern Analysis and Machine Intelligence, 2006, 28(6): 1013-1018.[7] Cathy M, Bertrand M. Data-driven penalty calibration: a case study for Gaussian mixture model selection[J]. ESAIM: Probability and Statistics, 2011, 15:320-339.[8] Figueiredo M A T, Jain A K. Unsupervised learning of finite mixture models[J]. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2002, 24(3):381-396.[9] Steele R J, Raftery A E. Performance of Bayesian model selection criteria for Gaussian mixture models[M]. Washington: Dept Stat Univ Washington D C Tech Rep, 2009: 559.[10] Alexander A, Thomas L. On image segmentation using information theoretic criteria[J]. The Annals of Statistics, 2011, 39(6):2912-2935.[11] Bradley P S, Fayyad U M. Refining initial points for K-Means clustering[C]. Proceedings of the 15th Inter- national Conference on Machine Learning, Canada: Montreal, 1998:91-99.[12] Jaakkola T S, Jordan M I. Bayesian parameter estimation via variational methods[J]. Statistics and Computing, 2000, 10(1):25-37.[13] Ghahramani Z, Beal M J. Variational inference for Bayesian mixtures of factor analysers[C]. In Advances in Neural Information Processing Systems, 2000:449-455.[14] Tzikas D G, Likas A C, Galatsanos N P. The variational approximation for Bayesian inference[J]. Signal Process- ing Magazine, 2008, 25(6):131-146.[15] Wainwright M J, Jordan M I. Graphical models, exponential families, and variational inference[J]. Foun- dations and Trends in Machine Learning, 2008, 1(1/2): 1-305.[16] Orbanz P, Buhmann J M. Nonparametric Bayesian image segmentation[J]. International Journal of Computer Vision, 2008, 77(1/3):25-45.[17] Blei D M, Jordan M I. Variational inference for Dirichlet process mixtures[J]. Bayesian Analysis, 2006, 1(1):121- 144.[18] John P, David B, Jordan M I. Variational Bayesian inference with stochastic search[C]. Proceedings of the 29th International Conference on Machine Learning, UK: Edinburgh, Scotland, 2012.[19] Hoffman M, Blei D M, Wang Chong, et al. Stochastic variational inference[J]. Journal of Machine Learning Research, 2012, 14:1303-1347. |
| 备注/Memo: | Received date: 2013?06?09. JOURNAL OF NINGBO UNIVERSITY ( NSEE ): http://journallg.nbu.edu.cn/Foundation items: Supported by the National Natural Science Foundation of China (61175026); Discipline Project of Ningbo University (XKL09154).The first author: ZHANG Yuan-yuan ( 1981? ), female, Ninghai Zhejiang, PhD, experimentalist, research domain: machine learning and computer vision.E-mail: zhangyuanyuan@nbu.edu.cn 宁波大学学报（理工版）网址：http://journallg.nbu.edu.cn/ |
