Construction of Binary Multi-grid Markov Random Field Prior Models from Training Images

Bayesian modeling requires the specification of prior and likelihood models. In reservoir characterization, it is common practice to estimate the prior from a training image. This paper considers a multi-grid approach for the construction of prior models for binary variables. On each grid level we adopt a Markov random field (MRF) conditioned on values in previous levels. Parameter estimation in MRFs is complicated by a computationally intractable normalizing constant. To cope with this problem, we generate a partially ordered Markov model (POMM) approximation to the MRF and use this in the model fitting procedure. Approximate unconditional simulation from the fitted model can easily be done by again adopting the POMM approximation to the fitted MRF. Approximate conditional simulation, for a given and easy to compute likelihood function, can also be performed either by the Metropolis–Hastings algorithm based on an approximation to the fitted MRF or by constructing a new POMM approximation to this approximate conditional distribution. The proposed methods are illustrated using three frequently used binary training images.     

Generation of Aquifer Heterogeneity Maps Using Two-Dimensional Spectral Texture Segmentation Techniques

Numerical models that solve the governing equations for subsurface fluid flow and transport require detailed quantitative maps of spatially variable hydraulic properties. Recently, there has been great interest in methods that can map the spatial variability of hydraulic properties such as porosity and hydraulic conductivity (permeability). Presently, only limited data on natural permeability spatial structure are available. These data are often based on extensive discrete sampling in outcrops or boreholes. Then methods are used to interpolate between data values to map aquifer heterogeneity. Interpolation methods often mask critical local or intermediate scale heterogeneities. As sediment texture is directly correlated with many hydraulic properties we developed two new texture segmentation algorithms based on a space-local two-dimensional wavenumber spectral method known as the S-Transform. Existing texture segmentation algorithms could not delineate the subtle and continuous texture variations that exist in natural sediments. The S-Transform algorithms successfully delineated geologic structures and grain size patterns in photographs of outcrops in a glacial fluvial deposit; thus, no interpolation methods were required to produce continuous two-dimensional maps of texture facies. The S-Transform method is robust and is insensitive to changes in light intensity, and moisture variations. This makes the algorithm particularly applicable to natural sedimentary outcrops. The effectiveness of our methods are tested by correlating measured relative grain sizes in the images with actual grain size measurements taken from the sedimentary outcrops.     

面向对象的喀斯特地区石漠化遥感信息提取研究——以贵州省大方地区为例

针对现有基于像素的监督和非监督分类方法在地质环境复杂、地形起伏较大、阴影明显的喀斯特石漠化地区难以满足石漠化信息提取精度要求的问题,采用基于纹理特征数据和地形数据辅助面向对象方法进行喀斯特地区石漠化信息的提取。该方法首先依据石漠化分布在TM/ETM+影像面积大小不均匀的特征,利用纹理和地形因子计算最优分割参数进行多尺度分割;然后根据植被覆盖率、岩石裸露率以及坡度因子构建石漠化分级指标;最后参照石漠化分级标准、光谱信息以及纹理特征等建立的分类规则提取喀斯特地区石漠化信息。选取贵州省石漠化严重的大方县时序TM/ETM+影像进行石漠化信息提取试验,结果表明:与基于像素的监督分类和非监督分类方法相比,基于面向对象的分类可以有效地减少因复杂地形导致石漠化信息提取结果"椒盐化"现象,提取精度明显优于基于像素的监督分类和非监督分类方法。      

基于遥感影像多尺度分割与地质因子评价的滑坡易发性区划

区域滑坡易发性的研究是滑坡空间预测的核心内容之一。从影像多尺度分割和面向对象的分类理论出发,以研究区遥感影像的熵、能量、相关性、对比度共4个参数作为影像纹理因子提取易发性特征,利用滑坡所处区域的库水影响等级、坡度、斜坡结构、工程岩组4类地质因子分析地质背景,搭建C5.0决策树的易发性分类模型,实现了对研究区内4类滑坡易发性单元的预测。结果表明:高易发性单元的工程岩组通常发育为软岩岩组和软硬相间岩组,且坡度在15°～30°之间;模型显示该区域训练样本和测试样本平均正确率达91.64%,Kappa系数分别为0.84,0.51,因此这种基于影像多尺度分割与地质因子分级的滑坡易发性分类研究具有一定的适用性。     

Textural segmentation of digital rock images into bedding units using texture energy and cluster labels

The texture of digital rock images, as recorded, for instance, with borehole imaging devices, is shown to reflect different bedding types. Textural segmentation of borehole images, therefore, subdivides the recorded sequence into bedding units. We show that a textural segmentation algorithm based on the concept of texture energy achieves good results when compared with synthetic as well as real data in which petroleum geologists have performed zonations on cores. Texture energy involves filtering of the original image with a set of texture sensitive masks. The filtering is done as a finite convolution over the size of the masks. On the resulting images the variance is computed over a relatively large sliding window, which, in its practical implementation, covers the full width of the image. The resulting nine one-dimensional curves are then clustered hierarchically into a user-determined number of image texture or lithological bedding classes. Principal component analysis previous to clustering can be used to reduce redundancy in the data. A recurring and relatively ill-defined problem in this field are macro-textures, i.e., the cyclic interbedding of two or more bedding types. We show that sliding Fourier transforms and variable mask scale can successfully address the zonation of macro-textures. In general, the method gives best results with mask sizes equivalent to 2–4 centimeters, reflecting the length scale at which the investigated geological bedding seems to have its highest variation.     

复杂结构构造区遥感图像的地质信息提取方法——基于小波变换的多层次图像分割

针对基岩裸露的造山带变形强烈、地层组合复杂、解译难度大的特点，提出了一种有效结合纹理信息和颜色信息的小波帧包变换的多层次图像分割方法，克服了以往图像处理方法将低层特征孤立起来的缺陷，在南天山东段铜花山－硫磺山地区地表地质结构的遥感影像解译中获得了令人满意的效果，为强烈变形的复杂结构构造区从遥感图像上提取地质信息提供了有效的途径。     

Automatic lithofacies segmentation using the wavelet transform modulus maxima lines combined with the detrended fluctuation analysis

The main goal of this paper is to design and develop a new technique and software tool that help automatic lithofacies segmentation from well logs data. Lithofacies is a crucial problem in reservoir characterization and our study intends to prove that techniques like wavelet transform modulus maxima lines (WTMM) and detrended fluctuation analysis (DFA) approaches allow geological lithology segmentation from well logs data. On the one hand, WTMM prove to be useful for delimitation of each layer. We based on its sensitivity on the presence of more than one texture. On the other hand, DFA is used to enhance the estimation of the roughness coefficient of each facies. We have used them jointly to segment the lithofacies of boreholes located in the Algerian Sahara. Obtained results are encouraging to publish this method because the principal benefit is economic.     

贵州西南部威-紫-罗断裂带构造特征及演化

威-紫-罗断裂带作为扬子板块南缘陆内较大的断裂, 控制着周围地区的盆地沉积及构造样式。本文通过野外地质勘查、平衡剖面技术及流体包裹体测温, 探讨了威-紫-罗断裂带的空间结构特征及构造演化历史。结果显示, NW-SE向威-紫-罗断裂带空间分段特征清楚, 运动性质以逆冲和左旋走滑活动为主。流体包裹体测温结果指示该断裂变形环境位于上地壳, 以脆性变形为主, 温度小于200 ℃。剖面复原结果显示, 断裂带在早泥盆世-中三叠世时期处于引张伸展构造背景下, 控制了断裂两侧沉积序列。断裂带挤压变形起始于中三叠世晚期的印支运动时期, 强烈变形发生在晚侏罗世早燕山运动时期, 与华南地块的向西挤压有关。      

一种面向对象的最优分割尺度计算模型

作为信息提取和分类的前提,面向对象的影像分割尺度参数的设置直接影响到提取和分类的精度。本文以GF-2影像数据为例,在已有分割理论和方法的基础上提出一种基于最优分割尺度的计算模型（OS模型）。该模型以主成分分析所得的主成分以及新建的归一化植被指数（normalized vegetation index,NDVI）特征层作为分割参考层,综合考虑均质因子的影响,构建加权尺度评价指数,插值拟合最优分割尺度。构建误差系数（E_c）对模型进行评价,结果表明：OS模型误差系数（E_c=1.15%）小于传统模型（E_c=3.28%）,且分割对象更均匀、与实际地物更接近。     

卷积神经网络和视觉注意力语义分割模型在高分辨率遥感影像分类中的性能分析

随着深度学习语义分割的快速发展,基于计算机视觉语义分割模型的高分辨率遥感影像分类方法也大量涌现。为系统定量地研究经典的和先进的视觉语义分割模型在遥感影像分类中的性能,在总结深度学习语义分割进展的基础上,选择9种基于卷积神经网络(CNN)和视觉注意力的语义分割算法,对米级和厘米级2个尺度的遥感数据集进行分析研究。在模型构建上基于计算机视觉通用的语义分割框架,训练时采用红绿蓝3波段遥感图像并基于ImageNet预训练权重进行迁移学习训练。研究结果表明：通用的语义分割模型通过常规训练设置进行训练能取得较好的遥感影像分类效果,部分地物的交并比(IoU)可以达到90%以上;基于视觉注意力的遥感影像分类模型的精度普遍高于基于CNN的模型,且MaskFormer能更有效地提取离散的地物信息;不同类别的精度最高值并不全在总体最优模型中,部分会存在于次优模型中;类似的地物在更高分辨率遥感数据集中可以获得更高的精度。     

纹理特征辅助的S AR影像冰川识别

青藏高原的冰川监测对气候变化研究有着重要的意义,通过遥感图像可以大范围长时间的监测冰川的变化,识别冰川边界是研究的重点。为了研究SAR影像纹理特征在冰川识别中的作用,以喀喇昆仑山地区的克勒青河上游为研究区,利用2018年Sentinel-1A数据进行干涉处理得到相干系数,然后基于相干系数提取了均值、方差、同质性、反差、相异性、熵、相关性共7种纹理特征,并对不同纹理特征组合之间的提取效果进行了比较。结果表明VV极化方式下均值、方差、同质性、相异性的特征组合冰川识别效果最好。据此提取了克勒青河上游区域的冰川边界,最高精度达到91.36%,该方法明显优于基于相干系数图的阈值分割法和基于光学影像的波段比值法,冰川识别精度提高了约2%。     

边界层参数化影响“梅花”台风的敏感性试验

以GRAPES-TCM为试验模式,对1109台风“梅花”进行了36次72 h的预报试验,通过试验分析了2种边界层参数化方案——MRF方案与YSU方案在不同情况下对台风预报的影响.结果显示:“梅花”路径与强度对边界层方案的变化都表现出一定的敏感性,敏感性大小与对流参数化方案、台风的初始强度等因素有关,强度的敏感性比路径更明显;对弱台风的路径与强度,YSU方案的总体预报效果优于MRF方案,对于强台风,2种边界层方案中MRF方案的路径预报效果更好,哪种方案的强度预报效果更好与对流参数化方案有关;无论何种情况,YSU方案预报的“梅花”强度都明显强于MRF方案,YSU方案预报的降水及感热通量与潜热通量总体上大于MRF方案;YSU方案时更多的感热通量和潜热通量与该方案时边界层更强的湍流混合有关,更多的潜热通量导致更多的降水,从而释放更多的潜热,更多的潜热释放以及更多的感热通量导致台风强度更强.     

基于面向对象分类技术的沈阳市土地利用动态变化研究

基于RS和GIS技术,采用面向对象分类技术,根据沈阳市的实际情况,建立土地利用类型解译标志,利用多尺度分割方法进行影像分割,然后利用决策树法和最近邻的分类方法,建立分类规则,提取区域土地利用信息,得到该区域的两期土地利用/土地覆被数据,并对研究区的土地利用动态变化及引起这种变化的驱动力进行分析.研究中所采用的面向对象的分类方法在地表景观信息提取中充分利用了地物的光谱信息、纹理特征等特征,最大程度上克服由于不同地物光谱信息相似、相同地物光谱信息不同而造成的混分现象.     

Precise estimation of pressure–temperature paths from zoned minerals using Markov random field modeling: theory and synthetic inversion

The chemical zoning profile in metamorphic minerals is often used to deduce the pressure–temperature (P–T) history of rock. However, it remains difficult to restore detailed paths from zoned minerals because thermobarometric evaluation of metamorphic conditions involves several uncertainties, including measurement errors and geological noise. We propose a new stochastic framework for estimating precise P–T paths from a chemical zoning structure using the Markov random field (MRF) model, which is a type of Bayesian stochastic method that is often applied to image analysis. The continuity of pressure and temperature during mineral growth is incorporated by Gaussian Markov chains as prior probabilities in order to apply the MRF model to the P–T path inversion. The most probable P–T path can be obtained by maximizing the posterior probability of the sequential set of P and T given the observed compositions of zoned minerals. Synthetic P–T inversion tests were conducted in order to investigate the effectiveness and validity of the proposed model from zoned Mg–Fe–Ca garnet in the divariant KNCFMASH system. In the present study, the steepest descent method was implemented in order to maximize the posterior probability using the Markov chain Monte Carlo algorithm. The proposed method successfully reproduced the detailed shape of the synthetic P–T path by eliminating appropriately the statistical compositional noises without operator’s subjectivity and prior knowledge. It was also used to simultaneously evaluate the uncertainty of pressure, temperature, and mineral compositions for all measurement points. The MRF method may have potential to deal with several geological uncertainties, which cause cumbersome systematic errors, by its Bayesian approach and flexible formalism, so that it comprises potentially powerful tools for various inverse problems in petrology.     

Sensitivity of MM5 and WRF mesoscale model predictions of surface winds in a typhoon to planetary boundary layer parameterizations

Sea surface winds and coastal winds, which have a significant influence on the ocean environment, are very difficult to predict. Although most planetary boundary layer (PBL) parameterizations have demonstrated the capability to represent many meteorological phenomena, little attention has been paid to the precise prediction of winds at the lowest PBL level. In this study, the ability to simulate sea winds of two widely used mesoscale models, fifth-generation mesoscale model (MM5) and weather research and forecasting model (WRF), were compared. In addition, PBL sensitivity experiments were performed using Medium-Range Forecasts (MRF), Eta, Blackadar, Yonsei University (YSU), and Mellor–Yamada–Janjic (MYJ) during Typhoon Ewiniar in 2006 to investigate the optimal PBL parameterizations for predicting sea winds accurately. The horizontal distributions of winds were analyzed to discover the spatial features. The time-series analysis of wind speed from five sensitivity experimental cases was compared by correlation analysis with surface observations. For the verification of sea surface winds, QuikSCAT satellite 10-m daily mean wind data were used in root-mean-square error (RMSE) and bias error (BE) analysis. The MRF PBL using MM5 produced relatively smaller wind speeds, whereas YSU and MYJ using WRF produced relatively greater wind speeds. The hourly surface observations revealed increasingly strong winds after 0300 UTC, July 10, with most of the experiments reproducing observations reliably. YSU and MYJ using WRF showed the best agreements with observations. However, MRF using MM5 demonstrated underestimated winds. The conclusions from the correlation analysis and the RMSE and BE analysis were compatible with the above-mentioned results. However, some shortcomings were identified in the improvements of wind prediction. The data assimilation of topographical data and asynoptic observations along coast lines and satellite data in sparsely observed ocean areas should make it possible to improve the accuracy of sea surface wind predictions.     

煤矿掘进工作面低照度视频增强技术研究

针对煤矿掘进工作面视频光照较低、亮度不均、纹理模糊、噪声较多等问题,提出一种煤矿掘进工作面低照度视频增强算法。首先,利用卷积的可分离性将视频图像进行一维水平卷积与垂直卷积,再利用完美反射法实现视频图像自动白平衡,并使用图像混合增强技术提高视频图像整体亮度。然后,基于大气散射模型与暗通道先验方法,通过递归分割将图像分割为高光区、中间调和暗调区,并求取对应区间通道像素最大值,将其3者均值作为大气光照估计值,引入调节因子对透射率进行调整优化,并使用拉普拉斯锐化操作,增加图像高频成分、抑制图像低频成分,提高图像对比度。最后,基于改进的大气散射模型对掘进工作面低照度视频进行去雾处理。实验结果表明,视频增强算法能够对煤矿掘进工作面低照度视频进行实时增强、去雾处理,避免了视频图像暗淡、失真、模糊和突变等问题。相较于Retinex算法、ALTM算法和暗通道先验算法,视频增强算法大幅度提高了视频图像的信息熵、标准差和平均梯度,且具有较好的实时处理速度,能够为掘进工作面视频的目标识别、目标跟踪、目标监测和图像分割等后续处理提供优质、可靠的支撑。      

湖北宜昌石板溪金矿区断层空间结构及其控矿规律研究

断层空间结构包括变形结构和几何结构两部分。本文在解析石板溪金矿区主要断层变形结构和几何结构的基础上，探讨了其空间结构模式和控矿规律。     

On the role of the Planetary Boundary Layer in the numerical simulation of a severe cyclonic storm Nargis using a mesoscale model

The very severe cyclonic storm Nargis of 2008 was a strong tropical cyclone that caused the deadliest natural disaster in the history of Myanmar. The time tested NCAR/PSU MM5 model has been used to simulate the Nargis cyclone, which is designed to have two domains covering the Bay of Bengal with horizontal resolutions of 90 and 30?km. The physics options chosen are Kain?CFritsch 2 for convection, Blackadar (BLA), Burk?CThompson, medium range forecast (MRF), Eta Mellor?CYamada (Eta MY) and Gayno?CSeaman (GS) for Planetary Boundary Layer (PBL) and Simple Ice for explicit cloud physics processes. The experiment was conducted with the model integration starting from April 27, 2008, to May 3, 2008. The performance of the five PBL schemes is evaluated in terms of radius height cross-section of the three component winds, surface heat fluxes of sensible heat and latent heat, equivalent potential temperature (?? _e ), precipitation, track and variation of Central Surface Pressure and wind speed with time. The numerical results show a large impact of the PBL schemes on the intensity and movement of the system. The intensity of the storm is examined in terms of pressure drop, strength of the surface wind and rainfall associated with the storm. The results are compared to the India Meteorological Department observations. These experiments indicate that the intensity of the storm is well simulated with the Eta MY and BLA with finer resolution. The simulated track with MRF compared well with the Joint Typhoon Warning Center observation at landfall position both with the 90 and 30?km resolutions.