基于岩土材料层析成像的孔隙网络构建算法

孔隙网络模型是研究多孔介质和其中颗粒的关键。现有的构建孔隙网络模型的算法有许多,但人们开始越来越多地研究基于图像的算法。将讨论一种基于中轴线的孔隙网络构建算法,并且对原有算法中孔隙中心表示方法不合理及缺少中轴线上各点坐标的问题提出改进,改进后的算法提出了孔隙中心的另一种表示方法,并且中轴线上任一点的坐标都被标出。在构建孔隙网络模型后利用该模型计算孔隙体积和喉径,并且针对过度分割的问题提出基于欧氏距离及基于分水岭算法的两种融合算法对孔隙和孔喉进行融合。     

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。      

孔壁钻孔图像的结构面参数提取方法研究

为了提升岩体结构面参数提取的自动化程度以及准确性,结合钻孔图像中岩体结构面的特性,开展了适用于结构面图像预处理和结构面参数提取的方法研究。首先,采用自适应中值滤波法对钻孔图像进行去噪处理,通过对梯度算子法和最大类间方差法的图像分割进行比较,提出了更适用于实际钻孔图像分割的改进方差法,即结合梯度算子与最大类间方差法;然后对分割后的图像进行Canny算子边缘检测;最后,通过多项式曲线拟合获得结构面的正弦曲线,从而计算出岩体结构面的参数——倾向、倾角和隙宽。通过将提取方法与传统的手动识别结果进行对比,其结果表明:提取方法获取的参数较为精确。从而证明了该结构面参数提取方法具有可行性和准确性,且提取效率高于传统的手动识别。     

基于小波变换的遥感数据融合与边缘检测

介绍了小波图像的分解和重构方法以及小波融合过程。采用Symlet小波变换融合方法对西安地区鲸鱼沟水库的TM4、TM5子图进行单尺度二维离散小波变换融合，并进行水库的边缘检测。对图像通过TM4分解的低频图与TM5分解的高频图像融合再与TM4影像原图比较，融合图像提取细节效果明显优于TM4原图像。     

基于图像描述的煤岩裂隙CT图像多尺度特征

基于煤岩孔隙系统多尺度结构特征对深入认识多尺度流体运移机制的重要性,提出了基于图像描述的煤岩CT图像孔隙结构的多尺度精细描述方法。采用了图像的多策略分割技术提取目标,利用Freeman链码对目标的边界进行表达,研究了由形态学、统计矩、链码、计盒维数构造目标之间的关系、目标占有区域与边界的图像描绘子、以及分形描绘子;综合运用上述方法对煤岩CT图像中的大尺度宏观裂纹目标、小尺度细观裂隙目标进行了识别。结果表明,宏观裂纹可由灰度阈值法实现目标提取;小尺度细观裂隙需采用较复杂的分割策略,如基于索贝尔梯度算子的分水岭变换;进一步应用链码表达、图像描绘子和分形描绘子,实现了煤岩孔隙结构在欧氏空间与分形空间的多尺度精确描述。     

消除噪声的RNF图像融合算法

在多光谱遥感图像中有时也会存在较严重的随机点噪声的干扰，这种随机点噪声严重影响了地物光谱特征提取和识别的精度，降低了各种遥感定量分析方法技术的有效性。常用的遥感图像随机噪声消除或压抑方法是Fourier变换频率域方法和采用平滑模板对图像进行卷积处理的空间域方法，但它们往往会损失图像信息。文章探讨了消除或压抑噪声的图像融合方法，即RNF融合法。RNF融合算法先对参与融合的多光谱图像进行低通滤波，对全色波段进行高通滤波。然后将滤波后的全色波段与多光谱图像用HSI变换法进行融合，融合后的图像消除了噪声。     

碎屑锆石U-Pb年龄有效性初探

碎屑锆石U-Pb 年代学是限定沉积地层年龄、追踪沉积物源、揭示区域构造演化历史等重要的手段,在地质学和地理学领域应用广泛。近年来,单颗粒碎屑锆石数据获取技术日渐成熟,数据量激增,然而,部分邻近甚至相同地区不同研究组获得的原始数据存在差异甚至相互矛盾。如何高效严谨地获取及处理数据,并挖掘其中所蕴涵的丰富的地质信息,越来越受到研究者的重视,但目前相关方面的研究较少。以青藏高原东北缘酒西盆地新生代疏勒河组砂岩碎屑锆石为例,采用标准实验流程,通过试验数据之间的横向和纵向对比,确定合理的碎屑锆石颗粒测试数量,利用碎屑锆石颗粒外部形态与内部结构特征相结合的方法,选择有效的测试点,发现单颗粒锆石阴极发光（CL）图像在数据测试、分析和应用方面极具前景。具体取得如下认识:1）碎屑锆石U-Pb测年前处理时需采用标准的重矿物分选流程,以保持所采集的碎屑样品中单颗粒锆石分布的全貌,避免引起年龄组分丢失或偏差;制作样品靶及选择测年颗粒时,需将全部锆石单颗粒在体视显微镜及扫描电镜下的特征紧密结合,得到最优有效数量的单颗粒年龄,提高效率及数据的准确性、完整性;2）数据处理、分析及应用中,必须结合单颗粒锆石CL图像,修正或剔除异常数据,避免由于激光剥蚀点或离子束跨越特征不同的生长环带产生无意义的“混合年龄”。该研究为碎屑锆石U-Pb年代学大数据处理及应用提供了基础技术支撑,且为有效甄别U-Pb测年数据质量进而合理、准确地应用于地学研究提供了新的视角。     

基于国产高分辨率光学遥感影像的水体提取

遥感图像中地表水体同山体、建筑物等地物产生的阴影在光谱特征上存在较高的类间相似性,导致提取过程中容易出现混淆和错分的情况。针对此问题,提出一种基于面向对象和人工蜂群的地表水体提取方法。该方法首先对遥感图像进行分割以获取分割对象的光谱、比率、几何形状等统计特征,以弥补高分遥感图像波段数目少,信息量不足的缺陷;并借助人工蜂群算法在解决复杂问题最优化方面的优势,选取水体同阴影二值分类的几何平均正确率作为算法的适应度函数,最终获取地表水体的最优化提取规则。选取厦门市大嶝岛和湖南省资兴市部分区域,基于国产高分一号、二号遥感数据进行水体提取,并与传统SVM分类结果进行比较。实验结果表明本算法提取水体的总体精度和Kappa系数均优于传统SVM分类器,表明该方法可应用于高分遥感图像的地表水体提取。     

基于2DPCA的遥感图像融合算法研究

基于PCA的遥感图像融合算法思想简单,实现起来较容易,而且遥感图像融合的性能也较好.但这种方法也有一些缺点:①不能有效利用图像的结构信,息;②光谱信息损失较多;③灵活性较差.针对这些问题,首次提出了一种基于2DPCA的遥感图像融合算法.与PCA融合算法相比,基于2DPCA的融合算法的主要特点是:①直接对图像矩阵进行2DPCA分析,因而可以有效利用图像的二维结构信息;②多光谱图像的特征矩阵主成分的替换个数可以是一个或多个,这样就可以调节光谱信息的保持程度和空同分辨率的改善程度,从而获得不同融合质量的图像,具有更好的灵活性;③融合后的图像不仅光谱信息得到了较好保持,空间分辨率获得了明显改善,而且图像色彩也得到了增强.     

基于正交偏光序列消光特征的岩石薄片颗粒分割与孔隙提取

在岩石薄片正交偏光显微镜下角度域序列图像采集的基础上,分析了不同岩石组分在消光角度域上的光学特征及其差异性,并据此提出一种新的岩石颗粒分割和孔隙分析的方法。充分利用岩石颗粒赋存状态及其晶体光轴产状的复杂性、岩石颗粒空间排列及其接触关系的多样性,对岩石薄片在消光角维度上进行像素尺度的相关分析,并提出相关系数均值、相关系数标准差以及相关系数均差比等敏感参数,实现了岩石颗粒的分割和孔隙的提取。研究表明岩石颗粒内部的像素点灰度和RGB值在角度域上的相关性较强,在颗粒边缘及孔隙内部填隙物分布区域,其相关系数较低,且相关系数标准差要明显高于岩石颗粒内部。该方法从消光特征出发论证了角度域信息完整性的意义,提取的颗粒边缘较为清晰,孔隙结构骨架得以刻画,颗粒分割的效果好于Sobel和Canny等方法。     

Porous Structure Reconstruction Using Convolutional Neural Networks

The three-dimensional high-resolution imaging of rock samples is the basis for pore-scale characterization of reservoirs. Micro X-ray computed tomography (µ-CT) is considered the most direct means of obtaining the three-dimensional inner structure of porous media without deconstruction. The micrometer resolution of µ-CT, however, limits its application in the detection of small structures such as nanochannels, which are critical for fluid transportation. An effective strategy for solving this problem is applying numerical reconstruction methods to improve the resolution of the µ-CT images. In this paper, a convolutional neural network reconstruction method is introduced to reconstruct high-resolution porous structures based on low-resolution µ-CT images and high-resolution scanning electron microscope (SEM) images. The proposed method involves four steps. First, a three-dimensional low-resolution tomographic image of a rock sample is obtained by µ-CT scanning. Next, one or more sections in the rock sample are selected for scanning by SEM to obtain high-resolution two-dimensional images. The high-resolution segmented SEM images and their corresponding low-resolution µ-CT slices are then applied to train a convolutional neural network (CNN) model. Finally, the trained CNN model is used to reconstruct the entire low-resolution three-dimensional µ-CT image. Because the SEM images are segmented and have a higher resolution than the µ-CT image, this algorithm integrates the super-resolution and segmentation processes. The input data are low-resolution µ-CT images, and the output data are high-resolution segmented porous structures. The experimental results show that the proposed method can achieve state-of-the-art performance.     

基于面向对象分类的稀土开采区遥感信息提取方法研究

离子吸附型稀土矿是我国宝贵的矿产资源,运用遥感影像分类技术提取稀土开采区可以准确地实现对稀土开采状况的监测,但仅利用光谱信息往往难以保证分类精度。本文以江西寻乌稀土矿区为研究区,以IKONOS影像为数据源,应用面向对象分类方法提取了稀土开采区的遥感信息。针对稀土开采区的分布特点,选择基于边缘的分割算法进行影像分割;结合地形信息、光谱信息及几何信息建立规则集,进行特征提取;最后采用隶属度函数法实现面向对象分类,并与传统的光谱角填图分类进行对比分析。研究结果表明,面向对象分类法提取稀土开采区的总体精度为92.49%,Kappa系数为0.857 6,与传统监督分类方法相比有了很大的提高。     

基于多分辨率图像融合的多尺度多组分数字岩心构建

针对单一分辨率的数字岩心模型无法完整描述岩心不同尺度结构信息的问题,以砂岩样品为例,通过对岩心多分辨率CT成像,采用基于特征的图像配准方法实现了不同分辨率岩心图像的精确匹配。通过融合不同分辨率岩心扫描图像进行孔隙分割和矿物分割,构建多尺度、多组分的数字岩心模型。结果表明:多尺度多组分数字岩心模型可以表征跨尺度的岩心孔隙结构,孔隙分布与核磁测量结果有较好的一致性;与Qemscan相比,数字岩心矿物体积分数同X衍射测定结果更吻合,真实地还原了岩心不同组分的结构信息。     

Identification of Mineral Grains in a Petrographic Thin Section Using Phi- and Max-Images

A new approach to identifying grains in a petrographic thin section is presented in this paper. The mineral grain boundaries are detected using two synthetic images, created by mapping the maximum birefringence color intensity (max-image) and the corresponding angular rotation at which it occurs (phi-image), instead of original images obtained by rotating the section between crossed polarizers. Edge detection and image segmentation operations are first applied on the phi- and max-images separately. The two segmented images resulting from edge detection are then combined to generate a new segmented image, which preserves edges with higher reliabilities and eliminates those with lower reliabilities in the two former segmented images. The identification rate is thereby greatly improved. The method has been implemented in C++ in the Linux environment. Two sets of images are used to test the method. Each set has 200 images corresponding to 200 rotation angles between 0 and 180°.     

储层微观孔喉网络图形识别方法

通过对大量岩样铸体薄片的图形特征反复观察与实验,设计了一种储层微观孔喉网络图形识别方法。该方法的关键是将原图片由RGB彩色空间转换到受亮度影响很小的YUV空间下,使待识别岩样孔喉网络产生很好的类聚,然后采用形态学分水岭分割方法识别出岩样铸体薄片图片中的孔隙与喉道。将该方法识别出的孔喉网络与RGB空间下一般图像分割技术识别出的孔喉网络对比,结果表明该方法识别出的孔喉网络比较完整与清晰;同时使用德国蔡司公司图像处理系统中的Axiovision4.0软件对待识别岩样薄片进行面孔率测算,结果表明该方法识别效果明显,相对误差降低了0.425%,大大提高了储层微观孔喉网络图形识别的准确度。     

基于蚁群和Canny边缘检测算子混合算法的二维岩石图像裂隙特征提取与修复研究

岩石中常含有裂缝、岩脉等细小而重要的微观结构,需要高分辨率的成像技术来提取其特征。在岩石图像处理中,特征提取可以提取出岩石的主要结构信息从而极大降低数据量,然而某些细线状岩石特征(如裂缝、裂隙等)往往无法被完整识别出来而存在间断现象。本文在采用蚁群算法修复岩石细线状特征的基础上,综合利用蚁群算法、DBSCAN聚类算法和Canny算子提出一种新的岩石图像修复算法。根据设定的阈值,对提取的图像间断部分进行搜索并连接,以修复在图像特征提取过程中缺失的部分重要信息,同时去除修复过程产生的多余特征。实际结果表明,利用混合算法处理的岩石特征提取图像得到了较好的修复,能够提高岩石裂隙特征提取的准确性。     

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。     

Analysis of soil erosion characteristics in small watersheds with particle swarm optimization, support vector machine, and artificial neuronal networks

Sand production by soil erosion in small watershed is a complex physical process. There are few physical models suitable to describe the characteristics of the intense erosion in domestic loess plateau. Introducing support vector machine (SVM) oriented to small sample data and possessing good extension property can be an effective approach to predict soil erosion because SVM has been applied in hydrological prediction to some extent. But there are no effective methods to select the rational parameters for SVM, which seriously limited the practical application of SVM. This paper explored the application of intelligence-based particle swarm optimization (PSO) algorithm in automatic selection of parameters for SVM, and proposed a prediction model by linking PSO and SVM for small sample data analysis. This method utilized the high efficiency optimization property and swarm paralleling property of PSO algorithm and the relatively strong learning and extending capacity of SVM. For an example of Huangfuchuan small watershed, its intensive fragmentation and intense erosion earn itself the name of “worst erosion in the world”. Using four characteristics selection algorithms of correlation feature selection, the primary affecting factors for soil erosion in this small watershed were determined to be the channel density, ravine area, sand rock proportion, and the total vegetation coverage. Based on the proposed PSO–SVM algorithm, the soil erosion modulus in the small watershed was predicted. The accuracy of the simulation and prediction was good, and the average error was 3.85%. The SVM predicting model was based on the monitoring data of sand production. The construction of the SVM erosion modulus prediction model for the small watershed comprehensively reflected the complex mechanism of soil erosion and sand production. It had certain advantage and relatively high practical value in small sample prediction in the discipline of soil erosion.     

Measurement of sand creep on a flat sand bed using a high-speed digital camera

In this study, a novel method for measuring sand creep along a flat bed was developed. Specifically, a high-speed digital camera was employed to record the sand creep movement. With an unsigned image subtraction, a new image queue could be formed from the original experimental images. The particle tracking was then calculated, after which the velocity and trajectory of the creep of sand grains was obtained. In addition, the creep fluxes of the different sand samples could also be calculated based on the velocities of the creep grains.     

Artificial Intelligence Identification of Multiple Microfossils from the Cambrian Kuanchuanpu Formation in Southern Shaanxi,China

The Cambrian Kuanchuanpu Formation in southern Shaanxi, China is a critical window for the understanding of the Cambrian explosion, because of abundant and various exceptionally preserved metazoans and embryo fossils yielded. The efficiency of traditional sample manually selecting with microscopes is quite low and hinder the discoveries of new species, thus recognition and classification of microfossils by artificial intelligence (AI) is substantially in the request. In this paper, we develop a procedure for fossil area segmentation in common multi-typed mixed photos by improved watershed algorithm. And for better fossil recognition, previous histogram of oriented grandient (HOG) algorithm is replaced by scale invariant feature transform (SIFT), which is feasible for the segmented images and increase the accuracy significantly. Thus, the scope of application of AI fossil recognition can be extended form single fossil image to multi-typed mixed images and the reliability is also secured, as the result of our test presents a high (at least 84%) accuracy of fossil recognition.