Integration of Gibbs Markov Random Field and Hopfield-Type Neural Networks for Unsupervised Change Detection in Remotely Sensed Multitemporal Images

被引:63
作者
Ghosh, Ashish [1 ]
Subudhi, Badri Narayan [1 ]
Bruzzone, Lorenzo [2 ]
机构
[1] Indian Stat Inst, Machine Intelligence Unit, Kolkata 700108, India
[2] Univ Trento, Dept Informat & Commun Technol, I-38050 Trento, Italy
关键词
Change detection; markov random field (MRF); maximum a posteriori probability (MAP) estimation; hopfield neural network; multitemporal images; remote sensing; CLASSIFICATION;
D O I
10.1109/TIP.2013.2259833
中图分类号
TP18 [人工智能理论];
学科分类号
081104 ; 0812 ; 0835 ; 1405 ;
摘要
In this paper, a spatiocontextual unsupervised change detection technique for multitemporal, multispectral remote sensing images is proposed. The technique uses a Gibbs Markov random field (GMRF) to model the spatial regularity between the neighboring pixels of the multitemporal difference image. The difference image is generated by change vector analysis applied to images acquired on the same geographical area at different times. The change detection problem is solved using the maximum a posteriori probability (MAP) estimation principle. The MAP estimator of the GMRF used to model the difference image is exponential in nature, thus a modified Hopfield type neural network (HTNN) is exploited for estimating the MAP. In the considered Hopfield type network, a single neuron is assigned to each pixel of the difference image and is assumed to be connected only to its neighbors. Initial values of the neurons are set by histogram thresholding. An expectation-maximization algorithm is used to estimate the GMRF model parameters. Experiments are carried out on three-multispectral and multitemporal remote sensing images. Results of the proposed change detection scheme are compared with those of the manual-trial-and-error technique, automatic change detection scheme based on GMRF model and iterated conditional mode algorithm, a context sensitive change detection scheme based on HTNN, the GMRF model, and a graph-cut algorithm. A comparison points out that the proposed method provides more accurate change detection maps than other methods.
引用
收藏
页码:3087 / 3096
页数:10
相关论文
共 22 条
[1]  
[Anonymous], 1982, Digital Picture Processing. Computer Science and Applied Mathematics
[2]   Unsupervised Change Detection in Multispectral Remotely Sensed Imagery With Level Set Methods [J].
Bazi, Yakoub ;
Melgani, Farid ;
Al-Sharari, Hamed D. .
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING, 2010, 48 (08) :3178-3187
[3]  
Bovolo F, 2008, IEEE T GEOSCI REMOTE, V46, P2647
[4]   Automatic analysis of the difference image for unsupervised change detection [J].
Bruzzone, L ;
Prieto, DF .
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING, 2000, 38 (03) :1171-1182
[5]   Kernel-based framework for multitemporal and multisource remote sensing data classification and change detection [J].
Camps-Valls, Gustavo ;
Gomez-Chova, Luis ;
Munoz-Mari, Jordi ;
Rojo-Alvarez, Jose Luis ;
Martinez-Ramon, Manel .
IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING, 2008, 46 (06) :1822-1835
[6]   Detection and analysis of change in remotely sensed imagery with application to wide area surveillance [J].
Carlotto, MJ .
IEEE TRANSACTIONS ON IMAGE PROCESSING, 1997, 6 (01) :189-202
[7]  
Chavez P. S, 1994, PHOTOGRAMM ENG REMOT, V60, P1285
[8]  
Fung T., 1988, PHOTOGRAMM ENG REMOT, V54, P1449
[9]   IMAGE SEGMENTATION USING A NEURAL NETWORK [J].
GHOSH, A ;
PAL, NR ;
PAL, SK .
BIOLOGICAL CYBERNETICS, 1991, 66 (02) :151-158
[10]   Object Detection From Videos Captured by Moving Camera by Fuzzy Edge Incorporated Markov Random Field and Local Histogram Matching [J].
Ghosh, Ashish ;
Subudhi, Badri Narayan ;
Ghosh, Susmita .
IEEE TRANSACTIONS ON CIRCUITS AND SYSTEMS FOR VIDEO TECHNOLOGY, 2012, 22 (08) :1127-1135