Entity

Time filter

Source Type


Wu Y.,Xidian University | Xiao P.,Shaanxi Bureau of Surveying and Mapping | Wang C.,Xidian University | Li M.,Xidian University
Guangxue Xuebao/Acta Optica Sinica | Year: 2010

In view of the speckle noise in the synthetic aperture radar (SAR) images, and based on the Contourlet's advantages of multiscale, localization, directionality, and anisotropy, a new SAR image fusion segmentation algorithm based on the persistence and clustering in the Contourlet domain is proposed. The algorithm captures the persistence and clustering of the Contourlet transform, which is modeled by hidden Markov tree (HMT) and Markov random field (MRF), respectively. Then, these two models are fused by fuzzy logic, resulting in a Contourlet domain HMT-MRF fusion model. Finally, the maximum a posterior (MAP) segmentation equation for the new fusion model is deduced. The algorithm is used to emulate the real SAR images. Simulation results and analysis indicate that the proposed algorithm effectively reduces the influence of multiplicative speckle noise, improves the segmentation accuracy and provides a better visual quality for SAR images over the algorithms based on HMT-MRF in the wavelet domain, HMT and MRF in the Contourlet domain, respectly. Source


Zhang P.,Xidian University | Li M.,Xidian University | Wu Y.,Xidian University | Gan L.,Xidian University | And 2 more authors.
Pattern Recognition Letters | Year: 2012

Particle filter (PF) is an effective approach to nonlinear and non-Gaussian Bayesian state estimation and has been successfully applied to wavelet-based synthetic aperture radar (SAR) image despeckling. In this paper, we propose an improved PF despeckling algorithm based on Markov Random Field (MRF) model that can preserve the edge, textural information and structural features of SAR images well. First, we show that the wavelet coefficients of SAR images which exhibit significantly non-Gaussian statistics can be described accurately by generalized Gaussian distribution (GGD) in stationary wavelet domain. Secondly, to amend the weight deviation, MRF model parameters are introduced to redefine the importance weight of the particles. At last, region-divided processing is implemented for the real time application of the proposed algorithm. The effectiveness of the proposed algorithm is demonstrated by application to simulated images and real SAR images. © 2012 Elsevier B.V. All rights reserved. Source


Wu Y.,Xidian University | Li M.,Xidian University | Zhang P.,Xidian University | Zong H.,Xidian University | And 2 more authors.
Pattern Recognition Letters | Year: 2011

Non-Gaussian triplet Markov random fields (TMF) model is suitable for dealing with multi-class segmentation of nonstationary and non-Gaussian synthetic aperture radar (SAR) images. However, the segmentation of SAR images utilizing this model still fails to resolve the misclassifications due to the inaccuracy of edge location. In this paper, we propose a new unsupervised multi-class segmentation algorithm by fusing the traditional energy function of TMF model with the principle of edge penalty. Through the introduction of the penalty function based on local edge strength information, the new energy function could prevent segment from smoothing across boundaries. Then we optimize the objective function that stems from the new energy function to obtain an iterative multi-region merging Bayesian maximum posterior mode (MPM) segmentation equation for the new segmentation algorithm. The effectiveness of the proposed algorithm is demonstrated by application to simulated data and real SAR images. © 2011 Elsevier B.V. All rights reserved. Source


Wu Y.,Xidian University | Wang X.,Xidian University | Xiao P.,Shaanxi Bureau of Surveying and Mapping | Gan L.,Xidian University | And 2 more authors.
Science China Information Sciences | Year: 2011

Non-Gaussian triplet Markov fields (TMF) model is suitable for dealing with multi-class segmentation of non-stationary and non-Gaussian synthetic aperture radar (SAR) images. Considering the complexity of the model and algorithm, as well as the requirement of real-time, and robust and efficient processing of SAR images, a fast algorithm based on TMF for unsupervised multi-class segmentation of SAR images is proposed in this paper. For the speckle noise in SAR images, numerical characteristic, threshold selection and QuadTree decomposition criterion are researched firstly. With the new method, a SAR image can quickly be mapped into an edge-based pixon-representation, which results in a coarse decomposition in smooth regions, and a fine decomposition in edges. Then by combining TMF model with the pixon-representation of SAR image, a new potential energy function of TMF based on pixon-representation is derived. Finally, the segmentation is finished by Bayesian maximum posterior mode (MPM). The effectiveness of the fast TMF algorithm is demonstrated by applying it to simulated data and real SAR images. © 2011 Science China Press and Springer-Verlag Berlin Heidelberg. Source


Zhang P.,Xidian University | Li M.,Xidian University | Wu Y.,Xidian University | Gan L.,Xidian University | Xiao P.,Shaanxi Bureau of Surveying and Mapping
Tien Tzu Hsueh Pao/Acta Electronica Sinica | Year: 2011

The particle filter (PF) algorithm has been successfully applied to synthetic aperture radar (SAR) image despeckling. In this paper, we propose a modified PF despeckling algorithm based on Markov random field (MRF) in stationary wavelet domain. It is shown that the wavelet coefficients of SAR images which exhibit significantly non-Gaussian statistics can be described accurately by generalized Gaussian distribution (GGD) in stationary wavelet domain. MRF is introduced to redefine the weight of the particles to amend the weight deviation. Furthermore, the sampling interval is updated according to the new weight. To enhance the efficiency of the proposed algorithm, region-divided processing is implemented. Experiment results and analysis demonstrate the ascendant performance of the proposed algorithm in noise reduction, preservation of the textural features, single target and edges of SAR images. Source

Discover hidden collaborations