利用灰度共生矩阵提取纹理属性的研究以及沉积相划分
[Abstract]:At present, 3D seismic attributes have been widely used in seismic interpretation, mainly in geological structure and geological interpretation, as well as lithology and pore fluid identification, reservoir characterization and so on. With the development of high speed digital electronic computer, the application computer can process and interpret the digital seismic record in a variety of ways. There are many kinds of seismic attributes, such as square root amplitude attribute, frequency attribute, coherent volume attribute, AVO attribute, wave impedance attribute and so on. At present, a large number of attributes are mainly used for reservoir prediction and reservoir characterization, which can identify lithology, fluid, fault and channel sand body, respectively. Therefore, 3D seismic attributes can help geophysical and geological engineers to make 3D visual interpretation, and improve the accuracy and efficiency of interpretation. This paper first summarizes the texture features, and introduces four methods of texture feature analysis from the point of view of texture analysis technology. Because of the uniqueness of seismic texture attributes compared with traditional seismic attributes, so in the seismic texture attribute analysis, The extraction of seismic texture attributes is particularly important. This paper briefly introduces three methods of texture extraction, emphasizes on the principle of attribute extraction based on gray level co-occurrence matrix, and explains its superiority over the other two methods. The traditional C1 algorithm is based on the statistical cross-correlation theory to calculate the coherence of seismic data along the line number and trace direction. Based on the theory of gray level co-occurrence matrix, a better coherence algorithm, seismic texture coherence attribute algorithm, is introduced. This algorithm not only takes into account the seismic coherent response characteristics along the direction of line number and trace sign, but also includes sum line number. The coherent information of the track in the angular direction makes full use of the coherence information of the seismic trace in the four directions. Moreover, the coherence between two or three adjacent channels considered by the traditional C1 algorithm can be extended to multi-channel coherence. The texture attribute parameters are also analyzed in this paper. Through the response of texture eigenvalues to different profiles under different parameters, the effect of different parameters on texture attribute resolution is discussed. In addition, the processing effect of 3D data volume shows that, Compared with the traditional C _ 1 and C _ 3 coherent algorithms, this method has higher lateral resolution and can effectively identify fault and channel boundaries. Finally, based on the structural sedimentary data, Huan 2 block is selected to study the seismic texture attributes of 4 sub-layers in the lower part of the second member of Shahejie formation, and good results have been obtained.
【学位授予单位】:长江大学
【学位级别】:硕士
【学位授予年份】:2015
【分类号】:P631.4
【相似文献】
中国期刊全文数据库 前4条
1 桑桑;郝鹏翼;丁友东;石蕴玉;;基于纹理和轮廓的铅笔素描画生成方法[J];上海大学学报(自然科学版);2010年03期
2 覃家美;周瑞霞;廖孟扬;;用多分辨率图象锥提取纹理方向性测度特征检测纹理边缘[J];武汉大学学报(自然科学版);1991年02期
3 何珏;赵鹏;李浩;;基于纹理的林区影像匹配窗口设置方法探讨[J];遥感信息;2013年04期
4 ;[J];;年期
中国博士学位论文全文数据库 前10条
1 张军;基于局部结构分布的统计纹理表征方法研究[D];西安电子科技大学;2014年
2 江巨浪;曲面纹理生成方法及实现的研究[D];合肥工业大学;2006年
3 赵洋;基于局部描述子的纹理识别方法及其在叶片识别方面的应用[D];中国科学技术大学;2013年
4 苗艳凤;木材山峰状纹理的视觉特性研究[D];南京林业大学;2013年
5 朱为;基于纹理合成的数字图像修补技术研究[D];国防科学技术大学;2010年
6 韩守东;纹理建模与图切分优化方法研究[D];华中科技大学;2010年
7 钱文华;基于纹理的非真实感绘制技术研究[D];云南大学;2010年
8 桂彦;可视媒体编辑与重用关键技术研究[D];上海交通大学;2012年
9 普园媛;云南重彩画艺术风格的数字模拟及合成技术研究[D];云南大学;2010年
10 张岩;纹理合成技术的研究及其应用[D];吉林大学;2006年
中国硕士学位论文全文数据库 前10条
1 吕秋丽;基于共生扩展八邻域矩阵的纹理识别方法[D];西安电子科技大学;2014年
2 李明;基于实际测量的纹理力触觉装置及建模方案改进研究[D];东南大学;2015年
3 张正;基于织物表面纹理的疵点分割方法研究[D];西安工程大学;2015年
4 潘翔;基于复用计算的肝脏软组织体纹理合成方法研究[D];福州大学;2014年
5 吴昊;利用灰度共生矩阵提取纹理属性的研究以及沉积相划分[D];长江大学;2015年
6 陆华锋;纹理合成算法的研究与应用[D];合肥工业大学;2009年
7 王元龙;纹理生成映射技术的研究及应用[D];太原科技大学;2010年
8 周航军;纹理合成算法的研究与应用[D];南京理工大学;2004年
9 颜星;基于能量优化的图像与视频纹理替换技术研究[D];北京理工大学;2010年
10 刘利娟;纹理标准性度量及近似周期性纹理合成方法探讨[D];西北工业大学;2005年
,本文编号:2417324
本文链接:https://www.wllwen.com/kejilunwen/diqiudizhi/2417324.html