基于匹配技术的砂箱砂体CT图像处理方法研究.docVIP

基于匹配技术的砂箱砂体CT图像处理方法研究 任旭虎, 李志超， 崔倩，王宗亮，姜翔飞，禹继法 （中国石油大学 信控与控制工程学院 山东省东营市 257061） 摘要：在对构造物理模拟实验砂箱砂体CT断层图像特征进行分析和研究的基础上，提出了砂箱砂体CT断层序列图像处理的技术方案和方法流程。通过比较直方图均衡化、分段线性变换两种图像增强方法以及邻域平均、自适应维纳滤波、中值滤波、小波去噪法四种图像滤波算法对砂箱砂体CT图像的处理结果，分析了这些不同方法的处理效果；并在这些方法的基础上，针对砂箱砂体CT断层序列图像模糊、质量差、砂体层与层之间的界限不明显等特征，提出了一种改进的模板匹配算法——二次模板匹配，二次匹配是在第一次模板匹配后的误差最小点的邻域内，对目标图像进行搜索匹配，找寻到匹配点。二次匹配有效地减少了噪声干扰，加强了图像特征，能较完整地提取出砂层标志线。 关键词：直方图均衡化；分段线性变换；邻域平均；自适应维纳滤波；中值滤波；中值结合小波去噪；模板匹配 Research Based on Template Matching Technology’s Sandbox Sandbody CT Image Processing Methods REN Xu-hu, Li Zhi-chao,Cui Qian, Wang Zong-liang,JIANG Xiang-fei，(College of Information and Control Engineering,China University of Petroleum,Dongying,257061,China) Abstract: Based on analyzing and discussing the sand body CT fault image features in tectonic physical simulation experimental sand box, some technical schemes and methods of sand body CT image processing are proposed. By comparing the results of different image enhancement algorithms棗hfistogram equalization and piecewise linear transformation , and image filtering algorithms for sand body CT image processing棗maverage neighborhood、adaptive wiener filtering、median filtering、 wavelet denoising， the processing accuracy and efficiency of these methods are analyzed. On this basis, aiming at the characteristics of sand body CT sequence images, such as blurs, poor quality and obscure boundary between the layers, an improved template matching algorithm is proposed棗tthe second template matching. The second template matching is in the error minimal point neighborhood after the first template matching, searching and matching for the target image, find the matching points at last.The new method can effectively reduce the noise, strengthen the image characteristics, and extract effective sand layer mark lines more integrally. Keywords: histogram equalization；piecewise linear transformation；average neighborhood；adaptive wiener filtering；median filtering；median filtering combined with wavel