基于小波域分层Markov模型的纹理分割

提出了一种新的小波域分层Markov模型。该模型使用高斯马尔可夫随机场(Gauss Markov randomfield,GMRF)模型描述每一尺度小波系数向量的分布,考虑了同一尺度特征之间的相互作用;利用尺度间的因果马尔可夫随机场(Markov random field,MRF)模型和尺度内的非因果MRF模型来描述标记场的局部作用关系,以此确定标记场的先验信息。根据贝叶斯准则,利用多目标问题优化技术,给出了此模型相应的纹理分割算法。最后,与经典模型的分割算法进行了对比实验,验证了所提出算法的有效性。     

基于立体环形邻域的GMRF模型及其在遥感影像纹理识别中的应用

对传统的GMRF模型进行了扩展,建立了基于立体环形邻域的GMRF模型,并设计了分步模型参数求解方法。这种新的模型不仅全面考虑了多光谱高分辨率影像中各波段内像元之间的空间相关性,而且还顾及了波段间像元的相关性。与传统的GMRF模型相比,基于立体环形邻域的GMRF模型提取的纹理信息更为丰富。采用Forrest彩色纹理图像和QuickBird卫星遥感影像进行了实验验证,实验结果表明本文提出的基于立体环形邻域GM-RF模型的纹理识别算法具有较强的普遍适用性,对不同情况的纹理影像(即空间相关性主要存在于波段间还是波段内)进行识别均能获得较好的分类效果。     

Bayesian analysis of zero inflated spatiotemporal HIV/TB child mortality data through the INLA and SPDE approaches: Applied to data observed between 1992 and 2010 in rural North East South Africa

Longitudinal mortality data with few deaths usually have problems of zero-inflation. This paper presents and applies two Bayesian models which cater for zero-inflation, spatial and temporal random effects. To reduce the computational burden experienced when a large number of geo-locations are treated as a Gaussian field (GF) we transformed the field to a Gaussian Markov Random Fields (GMRF) by triangulation. We then modelled the spatial random effects using the Stochastic Partial Differential Equations (SPDEs). Inference was done using a computationally efficient alternative to Markov chain Monte Carlo (MCMC) called Integrated Nested Laplace Approximation (INLA) suited for GMRF. The models were applied to data from 71,057 children aged 0 to under 10 years from rural north-east South Africa living in 15,703 households over the years 1992–2010. We found protective effects on HIV/TB mortality due to greater birth weight, older age and more antenatal clinic visits during pregnancy (adjusted RR (95% CI)): 0.73(0.53;0.99), 0.18(0.14;0.22) and 0.96(0.94;0.97) respectively. Therefore childhood HIV/TB mortality could be reduced if mothers are better catered for during pregnancy as this can reduce mother-to-child transmissions and contribute to improved birth weights. The INLA and SPDE approaches are computationally good alternatives in modelling large multilevel spatiotemporal GMRF data structures.     

Sonar image segmentation based on GMRF and level-set models

We propose two new level-set models to address the segmentation problem in sonar images. Local texture features, extracted using the Gauss-Markov random field model, are integrated into level-set energy functions to dynamically select regions of interest. Then, new two-phase level-set and multiphase level-set models are obtained by minimizing each new energy function, and the selection of model parameters is analyzed. The proposed models do not require re-initialization, which is usually a very costly procedure. Segmentation experiments on both synthetic and real sonar images show that the proposed two level-set models are accurate and robust when they are applied to noisy sonar images.     

基于高斯马尔可夫随机场混合模型的纹理图像分割

针对以高斯马尔可夫随机场中的邻域像素互作用参数为特征、以高斯混合模型为分类器的二步纹理图像分割方法,提出一个两者相互结合的一步模型———高斯马尔可夫随机场混合模型,并给出其EM算法的迭代计算公式。利用该模型进行纹理图像的分割实验,发现该算法在纹理图像分割的精度上比前者有较大程度的提高。     

无人机遥感影像林地单株立木信息提取

针对无人机遥感技术在提取单株立木信息的限制性问题,提出一种新的自动单株立木信息提取方法。对原始无人机影像进行光谱信息增强处理以突出局部细节特征;通过引入DBI指数自动化确定K-means聚类方法的最优聚类数目,进而对影像像素进行标记;通过利用高斯马尔可夫随机场模型进一步对影像进行分割;使用数学形态学算子等方法对分割结果进行后处理得到单株立木树冠信息,通过图像几何矩原理计算得到单株立木位置以作为其识别的依据。结果表明,应用该提取方法,油松林区和樟子松林区单株立木识别总体精度分别为89.52%和95.65%、单木树冠提取精度分别为81.90%和95.65%,均具有较好地适用性。该方法不需要大量的人工干预和先验知识的输入,大大提高提取方法的自动化程度。