基于时间演化的天然气水合物CT图像阈值分割

Natural Gas Hydrate CT Image Threshold Segmentation Based on Time Evolution

微米X射线计算机断层扫描作为一种数字岩心探测手段,被广泛应用于研究含天然气水合物沉积物赋存形态,但由于水合物与水对X射线的衰减系数相近,二者在CT图像中灰度区间存在交集,导致在从CT图像上对水合物与水进行分割时存在强非唯一性。为提高对CT图像中水合物与水阈值分割的准确性,本文通过分析天然气水合物生长过程中不同时刻CT图像直方图特征,提出归一化CT图像及其直方图的方法。首先选定甲烷气与石英砂的峰值灰度基准;然后用高斯函数分别对当前CT图像直方图中的甲烷气与石英砂曲线进行拟合,得到当前CT图像直方图中的甲烷气与石英砂峰值灰度;再将当前CT图像直方图中的甲烷气峰值灰度与石英砂峰值灰度归一化到选定的峰值灰度基准;进而用归一化的直方图对CT图像进行归一化;最后根据归一化灰度直方图的变化趋势,定量统计得到CT图像中水合物增加和气-水减少的灰度区间,完成图像中不同组分的阈值划分。实验结果表明,提出的阈值分割方法能够为天然气水合物CT图像中水合物与水边界的确定和水合物饱和度计算提供定量的依据,具有实际的工程应用价值。

Micro-scale X-ray computed tomography(CT) has been widely used to study the occurrence forms of gas hydratebearing sediments. However, the similarity between the X-ray attenuation coefficient of hydrate and that of water leads to a strong non-uniqueness in their phase differentiation in CT images. To improve threshold segmentation accuracy between hydrate and water in CT images, this study proposes a CT image and histogram normalized method by analyzing the histogram characteristics of CT images at different times during the growth process of natural gas hydrate. First, the peak gray value baseline of methane gas and quartz sand was selected. Then, a Gaussian function was used to fit the curves corresponding to methane gas and quartz sand in the current CT image histogram to obtain the peak gray values. In addition, the peak gray values of methane gas and quartz sand in the current CT image histogram were normalized to the chosen peak gray baseline.Subsequently, the normalized histogram was used to normalize the corresponding CT images. Finally, according to the changing trend of normalized gray histogram curves, the increasing gray ranges of hydrate and decreasing gray ranges of gaswater in CT images were obtained quantitatively, which guided threshold segmentation of CT images. Experimental results show that the proposed threshold segmentation method can provide a basis for phase differentiation between hydrate and water in CT images, improving the threshold segmentation accuracy.