摘要
原有基于简单马尔可夫随机场(MRF)模型的变化检测算法基于全局一致性假设,这一假设往往与实际情况不符,影响到结果准确性。本文提出基于观察场与标号场互相关的改进MRF模型及相应的变化检测算法。以迭代条件模型解决后验概率最大化问题,为像素分类;根据当前分类,利用邻域中同类像素调整观察场中的像素特征值;以新的像素特征进一步优化分类。本文采用两段迭代算法,以多时相遥感图像的差值图像做为观察场。实验证明该算法能有效提高检测结果精度。
Traditional unsupervised change detection algorithms based on simple MRF model assume that subimages applied to extracting features are homogeneous, but that is not always true and causes low accuracy. Based on the fields Correlation Markov Random Field （CMRF） model, an adaptive algorithm is proposed in this paper. The labeling is obtained through solving a Maximum A Posterior （MAP） problem by Iteration Condition Model （ICM）. Features of each pixel are exacted by using only the pixels currently labeled as the same pattern. With the adapted features, the new labeling is obtained. Under the idea of two-stage iteration algorithm, we use the difference image of multitemporal remote-sensing images as observation field. The satisfied experimental confirm the effectiveness of proposed techniques.
出处
《电子与信息学报》
EI
CSCD
北大核心
2008年第11期2737-2741,共5页
Journal of Electronics & Information Technology
基金
国家自然科学基金(60472072)
博士点基金(20040690034)资助课题
关键词
多时相遥感图像
互相关马尔可夫随机场
最大后验概率
同步自回归模型
迭代条件模型
Multitemporal remote-sensing images
Correlation MRF （CMRF）
Maximum A Posterior （MAP）
Simultaneous auto-regressive, Iteration condition model
