一种改进的遥感图像变化检测算法

袁琪; 赵荣椿

doi:10.3724/SP.J.1146.2007.00614

一种改进的遥感图像变化检测算法

doi: 10.3724/SP.J.1146.2007.00614

袁琪,
赵荣椿

基金项目:

国家自然科学基金(60472072)和博士点基金(20040690034)资助课题

计量
- 文章访问数: 3030
- HTML全文浏览量: 72
- PDF下载量: 1123
- 被引次数: 0
出版历程
- 收稿日期: 2007-04-23
- 修回日期: 2007-09-24
- 刊出日期: 2008-11-19

An Improved Approach to Change Detection in Multitemporal Remote-Sensing Images

摘要

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。
- 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
Abstract: 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.
- M /
- u /
- l /
- t

HTML全文

参考文献(1)

[1] Carlotto M J. Detection and analysis of change in remotelysensed imagery with application to wide area surveillance[J].Image Processing.1997, 6(1):189-202 [2] Bruzzone L, and Prieto D F. Automatic analysis of thedifference image for unsupervised change detection. IEEETrans. on Geosci and Remote Sensing, 2000, 38(8):1170-1182. [3] Fung T. An assessment of TM imagery for land-coverchange detection[J].IEEE Trans. on Geosci. Remote Sensing.1990, 28(12):681-684 [4] Singh A. Digital change detection techniques usingremotely sensed data[J].Int. J. Remote Sensing.1989, 10(6):989-1003 [5] Townshend Jr R G and Justice C O. Spatial variability ofimages and the monitoring of changes in the normalizeddifference vegetation index[J].Int. J. Remote Sensing.1995,16(12):2187-2195 [6] Melgani F and Moser G. Unsupervised change-detectionmethods for remote-sensing images[J].Opt. Eng.2002, 41(12):3288-3297 [7] Nielsen A A, Conradsen K, and Simpson J J. Multivariatealteration detection(MAD) and MAF processing in multispectral,bitemporal image data: New approaches to changedetection studies[J].Remote Sensing Environ.1998, 64(1):1-19 [8] Bruzzone L and Prieto D F. An adaptive semiparametric andcontext-based approach to unsupervised change detectionmultitemporal remote-sensing Images. IEEE Trans. on ImageProcessing, 2002, 11(4): 452-466. [9] Bazi Y, Bruzzone L, and Melgani F. An unsupervisedapproach based on the generalized Gaussian model toautomatic change detection in multitemporal SAR images[J].IEEE Trans. on Geosci and Remote Sensing.2005, 43(4):874-887 [10] Wiemker D. An iterative spectral-spatial Bayesian labelingapproach for unsupervised robust change detection onremotely sensed multispectral imagery[J].Proceedings of the7th Int. Conf. on Computer Analysis of Images and Patterns.1997, 1296:263-270 [11] Kasetkasem T and Varshney P K. An image change detectionalgorithm based on Markov random field models[J].IEEE Trans.on Geosci and Remote Sensing.2002, 40(8):1815-1823 [12] Hammersley J M and Clifford P. Markov field on finite graphsand lattices. Unpublished manuscript, 1971. [13] Kashyap R and Chellappa R. Estimation and choice ofneighbors in spatial interaction models of images, IEEETrans. on Inf. Theory, 1983, IT-29(1): 60-72. [14] Geman S and Geman D. Stochastic relaxation, Gibbsdistributions and the Bayesian restoration of images. IEEETrans. on Pattern Anal. Machine Intell, 1984, MPAI-6(6):721-741. [15] Li S Z, Rao C R, and Shanbhag D N. Modeling image analysisproblems using Markov random fields. Stochastic Processes:Modeling and Simulation, 2000, 20(5): 1-43. [16] Jianchang M and Anil K J. Texture classification andsegmentation using multiresolution simultaneousautoregressive models [J].Pattern Recognition.1992, 25(2):173-188 [17] Kirkpatrick S, Gelatt C D, and Vecchi Jr M P. Optimizationby simulated annealing[J].Science.1983, 220(4598):671-680

施引文献

资源附件(0)

访问统计