基于渗流模型的成像测井裂缝提取算法-太赫兹科学与电子信息学报.pdfVIP

第 12 卷 第 2 期 太赫兹科学与电子信息学报 Vo1．12，No．2 2014 年 4 月 Journal of Terahertz Science and Electronic Information Technology Apr．，2014 文章编号：2095-4980(2014)02-0272-04 基于渗流模型的成像测井裂缝提取算法 韦宇飞，吴晓红，袁传虎，刘如奎 (四川大学 电子信息学院，四川 成都 610064) 摘 要 ：针对成像测井图像成分复杂，噪声多导致裂缝提取效果不佳的问题，提出基于渗流 模型的成像测井裂缝提取算法。该算法把成像测井图像裂缝像素点之间的连续性作为成像测井裂 缝提取的主要依据，通过对初始像素点进行不断渗流，产生一个与初始像素点相关性极强的线性 渗流区域，再通过计算该渗流区域像素点的渗流参数来判断当前像素点是否属于裂缝上的目标。 实验表明，基于渗流模型的成像测井裂缝提取算法能有效抑制噪声和剔除孔洞等非裂缝物质，能 在复杂成像测井图像中准确地提取裂缝。 关键词 ：渗流模型；成像测井；裂缝；渗流参数 中图分类号 ：TN911.73 ；TP391.41 文献标识码 ：A doi ：10.11805/TKYDA201402.0272 Extraction algorithm of imaging well logging crack based on percolation model WEI Yu-fei，WU Xiao-hong，YUAN Chuan-hu，LIU Ru-kui (College of Electronic and Information，Sichuan University，Chengdu Sichuan 610064，China) Abstract： Most image segmentation algorithms for extraction of the imaging well logging crack are ineffective because of the complex composition and noises. A new extraction algorithm for imaging well logging crack based on percolation model is proposed. The algorithm uses the continuity of the crack pixels as the basis for extraction, and a linear seepage area that is closely related to the initial pixel is generated through the gradual percolation of the initial pixel. Then, by calculating the percolation parameters of the pixels in percolation area to determine whether the current pixel is the target of the crack. Experimental results show that the proposed algorithm can effectively suppress noises and remove holes and other substances, as