虚拟现实技术下交互式地震数据场可视化的研究与设计
发布时间：2018-09-18 17:47
【摘要】：地震勘探是当今世界范围内最为常用的油气藏勘探手段之一。近年来,随着地震勘探技术的飞速发展,地震勘探的深度越来越大,面对的地质构造也越来越复杂,采集到的地震数据体规模也越来越庞大。地震数据的价值在于它能反映出地下信息,将这些巨量的、杂乱的、抽象的数据以图形、图像的形式展示出来,对下一步的地震资料处理及解释工作提供有力帮助,本课题即是研究地震数据可视化方面的相关内容。二维图像可视化是最基本的科学计算可视化方法之一,在地震数据可视化领域中,二维剖面展示也是最为常用的可视化方法之一。然而随着三维地震勘探技术的不断进步,相比二维可视化过程中信息丢失的问题,三维可视化技术可以更加充分利用地震数据,以直观的三维图像对各种复杂结构进行展示,成为寻找油气藏的有效手段。在这种情况之下,如何利用计算机中央处理器CPU以及图像处理器GPU的高速运算能力,实现三维可视化算法,提高算法运行效率,丰富软件交互方式,是目前亟待解决的问题。对此众多地球物理勘探公司和科研机构都投入了大量的精力和物力,致力于地震数据可视化及交互技术的研究。在调研大量文献之后,本文总结了地震数据可视化背景所涉及的技术,包括科学计算可视化技术、地震数据采集及存储格式以及跨平台图像库Qt和开放图形库OpenGL,为可视化系统的实现提供了思路。针对地震数据可视化问题,本文从科学计算可视化技术出发,分析了通用的科学计算可视化处理流程及常用可视化算法。在此基础之上,针对海量地震数据,分析了三维数据网格化技术,为三维可视化建模问题提供了理想的解决办法。随后总结了现有二维剖面可视化和三维地震数据场可视化方法,并对可视化方法的优劣性进行分析,为系统可视化方法的最终实现提供了合理的参考。此外针对未来地震数据可视化的发展趋势——虚拟现实技术,本文也进行了简短的、前瞻性的探讨；本文针对地震数据场交互式技术,重点分析了系统缩放功能、色标设置功能、切面绘制功能以及层位拾取功能,并对这四种交互功能的实现原理进行了较为详细的介绍,为系统交互式设计提供了依据；最后系统按照软件工程思想进行开发,先后经过系统概述、系统需求分析、系统总体设计以及系统流程设计等步骤后,采用面向对象编程的开发方式,实现了在Linux平台下轻量级的交互式地震数据场可视化系统。该系统使用跨平台图像库Qt作为界面设计工具,采用C++作为主要的开发语言,实现了交互功能和可视化功能。其次,使用Ⅰ-710KB, Ⅱ-3.6MB, Ⅲ-91MB, Ⅳ-363MB,Ⅴ-3.0GB五个不同大小的地震数据体对系统功能、性能进行测试,测试结果表明该系统基本满足了地质人员对可视化功能的要求,但是相比同类的成熟商业软件,该系统功能还比较单一,且性能也存在较大的改进余地。然后,针对系统加载较大地震数据体时,绘制图像缓慢的问题,采用GPU高性能运算CUDA并行运算技术对绘制算法进行优化和加速,最终结果表明该技术可以极大的提高数据体的绘制速度,节约了地质人员的等待时间,取得了良好的效果。文末,对系统各部分功能进行总结,指出目前工作中存在的不足,并提出了工作展望。
[Abstract]:Seismic exploration is one of the most commonly used means of oil and gas reservoir exploration in the world. In recent years, with the rapid development of seismic exploration technology, the depth of seismic exploration is becoming larger and larger, the geological structure is becoming more and more complex, and the size of seismic data volume is becoming larger and larger. Underground information, these huge, messy, abstract data in the form of graphics, images to display, for the next step of seismic data processing and interpretation work is a powerful help, this topic is the study of seismic data visualization related content. In the field of seismic data visualization, two-dimensional section display is also one of the most commonly used visualization methods. However, with the continuous progress of three-dimensional seismic exploration technology, compared with the problem of information loss in two-dimensional visualization process, three-dimensional visualization technology can make full use of seismic data to intuitive three-dimensional images of various complex structures. In this case, how to make use of the high-speed computing ability of CPU and GPU of the computer central processing unit to realize the three-dimensional visualization algorithm, improve the efficiency of the algorithm and enrich the software interaction mode is an urgent problem to be solved. After investigating a large number of literatures, this paper summarizes the techniques involved in the background of seismic data visualization, including visualization of scientific computing, seismic data acquisition and storage formats, and cross-platform image database Qt and opening. In view of the visualization of seismic data, starting from the visualization technology of scientific computing, this paper analyzes the general visualization processing flow and common visualization algorithm of scientific computing. On this basis, aiming at the massive seismic data, this paper analyzes the three-dimensional data gridding technology. This paper summarizes the existing methods of two-dimensional section visualization and three-dimensional seismic data field visualization, and analyzes the advantages and disadvantages of these methods, which provides a reasonable reference for the final realization of the system visualization method. The trend-virtual reality technology is also discussed briefly and prospectively in this paper.In view of the interactive technology of seismic data field,the system scaling function,color mark setting function,section drawing function and layer picking function are analyzed emphatically.The realization principle of these four interactive functions is introduced in detail. Finally, the system is developed according to the idea of software engineering. After the steps of system overview, system requirement analysis, system overall design and system flow design, the lightweight interactive seismic data field visualization system based on Linux platform is realized by object-oriented programming. The system uses the cross-platform image library Qt as the interface design tool and C++ as the main development language to realize the interactive and visualization functions. Secondly, the system functions and performances are tested with five different sizes of seismic data volume: I-710KB, II-3.6MB, III-91MB, IV-363MB, V-3.0GB. The test results show that the system has good performance. The system basically satisfies the requirements of geological personnel for visualization function, but compared with the mature commercial software of the same kind, the system has a single function, and there is still much room for improvement in performance. Then, aiming at the problem of slow rendering image when the system loads large seismic data volume, the GPU high-performance computing CUDA parallel computing technology is adopted. The rendering algorithm is optimized and accelerated. The results show that the technique can greatly improve the rendering speed of data volume, save the waiting time of geologists and achieve good results.
【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】：P631.44;TP391.41
[Abstract]:Seismic exploration is one of the most commonly used means of oil and gas reservoir exploration in the world. In recent years, with the rapid development of seismic exploration technology, the depth of seismic exploration is becoming larger and larger, the geological structure is becoming more and more complex, and the size of seismic data volume is becoming larger and larger. Underground information, these huge, messy, abstract data in the form of graphics, images to display, for the next step of seismic data processing and interpretation work is a powerful help, this topic is the study of seismic data visualization related content. In the field of seismic data visualization, two-dimensional section display is also one of the most commonly used visualization methods. However, with the continuous progress of three-dimensional seismic exploration technology, compared with the problem of information loss in two-dimensional visualization process, three-dimensional visualization technology can make full use of seismic data to intuitive three-dimensional images of various complex structures. In this case, how to make use of the high-speed computing ability of CPU and GPU of the computer central processing unit to realize the three-dimensional visualization algorithm, improve the efficiency of the algorithm and enrich the software interaction mode is an urgent problem to be solved. After investigating a large number of literatures, this paper summarizes the techniques involved in the background of seismic data visualization, including visualization of scientific computing, seismic data acquisition and storage formats, and cross-platform image database Qt and opening. In view of the visualization of seismic data, starting from the visualization technology of scientific computing, this paper analyzes the general visualization processing flow and common visualization algorithm of scientific computing. On this basis, aiming at the massive seismic data, this paper analyzes the three-dimensional data gridding technology. This paper summarizes the existing methods of two-dimensional section visualization and three-dimensional seismic data field visualization, and analyzes the advantages and disadvantages of these methods, which provides a reasonable reference for the final realization of the system visualization method. The trend-virtual reality technology is also discussed briefly and prospectively in this paper.In view of the interactive technology of seismic data field,the system scaling function,color mark setting function,section drawing function and layer picking function are analyzed emphatically.The realization principle of these four interactive functions is introduced in detail. Finally, the system is developed according to the idea of software engineering. After the steps of system overview, system requirement analysis, system overall design and system flow design, the lightweight interactive seismic data field visualization system based on Linux platform is realized by object-oriented programming. The system uses the cross-platform image library Qt as the interface design tool and C++ as the main development language to realize the interactive and visualization functions. Secondly, the system functions and performances are tested with five different sizes of seismic data volume: I-710KB, II-3.6MB, III-91MB, IV-363MB, V-3.0GB. The test results show that the system has good performance. The system basically satisfies the requirements of geological personnel for visualization function, but compared with the mature commercial software of the same kind, the system has a single function, and there is still much room for improvement in performance. Then, aiming at the problem of slow rendering image when the system loads large seismic data volume, the GPU high-performance computing CUDA parallel computing technology is adopted. The rendering algorithm is optimized and accelerated. The results show that the technique can greatly improve the rendering speed of data volume, save the waiting time of geologists and achieve good results.
【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】：P631.44;TP391.41
【相似文献】
相关期刊论文 前10条
1 史俊文;冯炜;邵佳妮;申俊飞;;数据场在可视化上的应用[J];测绘科学;2010年S1期
2 董峰,叶旭Q，
本文编号：2248658
本文链接：https://www.wllwen.com/kejilunwen/diqiudizhi/2248658.html
