• 主题
高级搜索  >
历史搜索 清空历史
首页> 美国政府科技报告 >Bayesian Color Image Segmentation Using Reversible Jump Markov Chain Monte Carlo.211 Probability, Networks and Algorithms
【24h】

Bayesian Color Image Segmentation Using Reversible Jump Markov Chain Monte Carlo.211 Probability, Networks and Algorithms

机译：基于可逆跳马尔可夫链的贝叶斯彩色图像分割monte Carlo.211概率，网络和算法

获取原文
页面导航
  • 摘要
  • 著录项
  • 相似文献
  • 相关主题

摘要

This paper deal with the problem of unsupervised image segmentation. Our goal is211u001eto propose a method which is able to segment a color image without any human 211u001eintervention. The only input is the observed image, all other parameters are 211u001eestimated during the segmentation process. Our method is model-based, we use a 211u001efirst order Markov random field (MRF) model (also known as the Potts model) where 211u001ethe singleton energies derive from a multivariate Gaussian distribution and 211u001esecond order potentials favor similar classes in neighboring pixels. The most 211u001edifficult part is the estimation of the number of pixel classes or in other 211u001ewords, the estimation of the number of Gaussian mixture components. Reversible 211u001ejump Markov chain Monte Carlo (MCMC) is used to solve this problem. These jumps 211u001eenable the possible splitting and merging of classes. The algorithm finds the 211u001emost likely number of classes, their associated model parameters and generates a 211u001esegmentation of the image by classifying the pixels into these classes. The 211u001eestimation is done according to the Maximum A Posteriori (MAP) criteria. 211u001eExperimental results are promising, we have obtained accurate results on a 211u001evariety of real color images.

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱：kefu@zhangqiaokeyan.com

京公网安备：11010802029741号 ICP备案号：京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有

  • 客服微信

  • 服务号