基于电成像测井的图像形状特征提取与应用

Extraction and Application of the Shape Features of Electrical Image Logging Data

碳酸盐岩储集空间非均质性极强,种类多样,仅依靠常规测井很难判别其岩性和沉积相。电成像测井资料能够清晰地反映碳酸盐岩储层的结构组分和沉积构造,为岩性的判别和沉积相的划分提供可靠的依据。针对碳酸盐岩沉积相模式划分的问题,提出利用分水岭分割算法对图像进行初始分割,再依据OSTU算法(大津算法)进行二次分割,提取电成像测井图像形状特征,如面孔率、周长、面积等,最后利用最小欧式距离法对其沉积相模式进行判别。研究结果表明,该方法能准确地识别碳酸盐岩沉积相模式。

Because of the diversity of types of carbonate reservoir,and its heterogeneity was strong,it was very difficult to discriminate its sedimentary facies and lithology based on normal well logging.Electric imaging well logging data could clearly reflect the structural components and the sedimentary structure of carbonate reservoirs,and provide the reliable evidence to distinguish sedimentary facies and division of lithology.Aimed at the problem of discrimination of modes of carbonate sedimentary facies,watershed segmentation algorithm was used to segment the images initially and it was resegmented based on OSTU algrithm,and then the imaging shape features,like areal plane porosity,acreage and perimeter were extracted.Finally the method of minimum Euler-distance was applied to determine the sedimentary patterm.The result shows that the method can be used for accurately identifying the sedimentary pattern of carbonate sedimentary facies.