应用形态分量分析算法研究电成像测井空白条带填充

摘要

Borehole micro-resistivity images are an important source of high resolution information with broad application in geology and petrophysics.For better post-process and evaluation,gaps between pads in electrical logging images should be filled.Based on the analysis of advantages and disadvantages of current main filling-in method—Filtersim,an improved method is proposed which constraints filling-in sequence by linear structural strength in neighbors of bound point to replace sequential simulation in Filtersim.Structural information in images are effectively preserved by this method.Principle of Morphological Component Analysis (MCA) method based on sparse representation is discussed.We compare filling-in effects of two methods and find that MCA method is more feasible for decomposing and reconstructing the structural and texture information respectively from aspect of signal sparse representation.%分析了目前常用电成像图像空白条带填充方法Filtersim的优缺点,在此基础上,提出一种改进算法,以边缘图像点邻域内的线性结构强度参数为填充顺序约束,取代Filtersim中的序贯模拟,有效保持了图像中的结构信息.探讨了基于信号稀疏表示的形态分量分析(MCA)方法原理,通过比较分析Filtersim和MCA这2种方法的填充效果,认为MCA方法分别对图像结构信息和纹理信息进行分解和重构,填充效果比Filtersim更好.