碳酸盐岩复杂背景孔洞自动识别及参数计算

Automatic identification and parameter calculation of carbonate cavities under complex background

 李志虎¹ 张翔¹ 罗利² 毛英雄² 李猛³

长江大学油气资源与勘探技术教育部重点实验室，湖北 武汉 430100 中国石油集团川庆钻探工程有限公司测井公司，重庆 400021 中国石化中原油田分公司采油五厂，河南 濮阳 457001）

MOE Key Laboratory of Exploration Technologies for Oil and Gas Resources, Yangtze University, Wuhan 430100, China CCDC Well Logging Company, CNPC, Chongqing 400021, China No.5 Oil Production Plant, Zhongyuan Oilfield Company, SINOPEC, Puyang 457001, China)

 碳酸盐岩储层解释与评价的核心之一就是孔洞识别及参数定量计算。目前孔洞识别及参数定量计算存在两方面问题。一方面，主要采用基于OTSU图像分割算法提取孔洞目标，在灰度直方图存在多峰、图像存在暗点干扰等复杂背景时，孔洞难以精确提取；另一方面，主要利用孔洞轮廓坐标计算孔洞参数，存在速度慢、计算量大的问题。针对此类问题，文中提出了OTSU结合分水岭的图像分割算法，有效地从复杂背景中提取孔洞目标；同时，研究了孔洞编号标记算法、轮廓追踪算法，利用Freeman八方向链码储存轮廓信息，根据链码快速计算孔洞面积、周长、面孔率等参数，有效地节省内存空间，提高运算速度。实测资料处理结果表明，该方法能在复杂背景下有效地提取孔洞，较快地计算孔洞参数。

 The key of interpretation and evaluation of carbonate reservoir is cavity recognition and parameter calculation. At present, there are two problems. On one hand, it mainly uses OTSU to extract cavities. But this way is not effective when there are many peaks in histogram and dark spots in image. On the other hand, it mainly uses cavities outline coordinate to calculate its parameters. This way has the problem of slow speed and much calculation. To solve this problem, watershed combined with OTSU image segmentation algorithm, which can effectively extract cavities under complex background, is studied in this paper. We also study labeling algorithm based on equivalence pair and contour tracing algorithm. In order to facilitate the cavities parameter calculation, this paper uses Freeman-8 codes to store cavities outline information which can gain cavities area, perimeter, and plane porosity. In this way, we can increase speed and improve accuracy. Measured data show that the proposed algorithm in this paper indicates a good effect under complex background.

10.6056/dkyqt201603009