电磁波CT数据预处理的研究

Research on the procession of electromagnetic wave CT data

电磁波CT测井技术利用CT接收机接收到的数据来反演重建图像,从而获取地质信息,但数据中含有的噪声将严重影响重建图像的质量,所以对采集的数据进行有效滤波就成为了电磁波CT成像的关键一环。针对传统采集方式获取的数据受噪声影响大的问题,对每一通道进行短时连续采样,将采集数据构造成三维矩阵,再把此三维矩阵按采样维和幅值维分解为二维矩阵,逐点进行加权滑动平均预处理,与限幅算数平均滤波方法相比,前者在恢复数据具有更好的平滑度且克服了限幅算数平均滤波对信号失真较大的弊端,能够为电磁波CT井间探测提供更准确的数据。

Electromagnetic wave CT logging technology using CT receiver receives the data to reconstruct image,getting geological information,but the data that contains noise will seriously affect the quality of the reconstructed image,so filtering data efficiently has become a key link of the electromagnetic wave CT imaging.Because of the traditional way to get data is deeply affected by noise,proposed taking short quick sampling method to obtain data for each channel,constructed three-dimensional matrix by the data,then decomposed it into two-dimensional matrix,using the weighted moving average procession point by point.According to the method of filtering by limiting arithmetic average,the former one has a better filtering performance in smoothing data in the aspect of recovering data,and overcomes the disadvantage that the method of filtering by limiting arithmetic average will have more serious distortions.It can support more exact data for the cross-hole electromagnetic wave detection.