Change Detection In SAR Images Based On Triplet Markov Fields In Wavelet Domain

Synthetic aperture radar(SAR)is insensitive to atmospheric and illumination conditions,it can monitors on targets all day and all weather.With the rapid development of SAR imaging,earth observation remote sensing data of different resolutions emerges in large amounts,which provides significant support for the theoretical study and practical application of remote sensing.SAR image change detection is an important branch of SAR image applications,it is widely used in military and civil fields.Due to the coherent imaging mechanism,the SAR images will inevitably introduce in speckle noise which will bring error to change detection results.Therefore,how to suppress the speckle noise in the SAR images has been an urgent problem in SAR image change detection.Aiming at the effect of speckle noise,some scholars put forward change detection algorithms based on markov random field(MRF),because MRF model can carry on the spatial context information of the images,which greatly improved the accuracy of change detection.The whole image in MRF model is considered as a large homogeneous region.Therefore,the MRF model is not proper to model nonstationary images.The method based on triplet Markov fields(TMF)model is proposed to overcome this defect by introducing a auxiliary filed U to describe different nonstationarities in the image.The traditional TMF model uses sample 4-nighborhood systems,it is impossible to define target edges accurately if the speckle noise in SAR images is strong.For this,this paper introduces an improved SAR image change detection algorithm based on TMF,this algorithm will extend the 4-nighborhood system in order to use more spatial context information of the images.On this basis,this paper puts forward a SAR image change detection algorithm based on triplet markov fields in wavelet domain.Firstly,the difference image is divided into four layers of stationary wavelet transform,as a result,four low frequency components of different scales are obtained.Among them,the low frequency components in low scale can preserve the edge information,and the low frequency components in high scale have less speckle noise.Then,for each low frequency component,the improved SAR image change detection algorithm based on TMF is used to obtain change image.Finally,the four change images are fused to get the final change detection result.The advantage of this algorithm is that it can reduce the speckle noise as well as improve the accuracy of location.In this paper,the measured SAR image data are used to simulate the proposed algorithm,and the effectiveness of the proposed algorithm is verified by the detailed analysis of the experiment results.