Page 70 - 《应用声学》2022年第4期

第 41 卷 第 4 期 Vol. 41, No. 4
2022 年 7 月 Journal of Applied Acoustics July, 2022

⋄ 研究报告 ⋄

各向异性扩散滤波远探测声波测井图像降噪方法 ∗

傅艳莉 1,2,3 李 超 1,2,3† 陈 浩 1,2,3 曹雪砷 2,3

(1 中国科学院大学 北京 100049)
(2 中国科学院声学研究所 声场声信息国家重点实验室 北京 100190)
(3 北京市海洋深部钻探测量工程技术研究中心 北京 100190)
摘要：声波远探测测井技术近年来在复杂油气藏的构造识别和储层评价中发挥着重要作用。该技术利用来自
井外的反射波对井旁地质构造进行准确成像，但由于反射波具有幅度低、受幅度强的井孔直达波干扰等特点，
实际数据提取到的反射波信噪比往往比较低，需要对反射波进行降噪处理。非线性各向异性扩散滤波能够在
滤除图像噪声的同时保留图像边缘及细节等信息，在地震数据处理和医学图像去噪中都有广泛应用。该文从
各向异性扩散滤波的基本原理入手，将提取到的井外反射波信号当作图像，采用不同扩散张量进行处理，通过
含噪声的模拟数据处理验证了该方法的处理效果并建立起适合于远探测测井的数据处理流程，实际远探测数
据处理结果进一步表明其具有较好的应用前景。
关键词：远探测技术；反射波测井；各向异性扩散滤波；图像降噪；相干增强
中图法分类号: P631.5 文献标识码: A 文章编号: 1000-310X(2022)04-0568-10
DOI: 10.11684/j.issn.1000-310X.2022.04.008

Anisotropic diffusion filtering based image noise reduction method for
remote detection acoustic logging

FU Yanli 1,2,3 LI Chao 1,2,3 CHEN Hao 1,2,3 CAO Xueshen 2,3
(1 University of Chinese Academy of Sciences, Beijing 100049, China)
(2 State Key Laboratory of Acoustics, Institute of Acoustics, Chinese Academy of Sciences, Beijing 100190, China)

(3 Beijing Engineering Research Center for Deep Drilling Exploration and Measurement, Beijing 100190, China)
Abstract: In recent years, remote detection acoustic logging technology has played an important role in
structural identification and reservoir evaluation of complex oil and gas reservoir. This technology uses the
reflected wave outside the well to accurately image the geological structures near the borehole. However, because
of the characteristics of low amplitude and influenced by the borehole direct wave with high amplitude, the
signal to noise ratio of the reflected wave extracted from the field data is generally low. Therefore, it is necessary
to denoise the reflected wave. The nonlinear anisotropic diffusion filtering can eliminate the noise of the image
while preserving the edge and detail information. It has been widely used in seismic data processing and
medical image denoising. This research starts with the basic principle of the anisotropic diffusion filter. The
filtered reflections are regarded as an image and processed with different diffusion tensor. The effectiveness
of the method is verified by simulated data with noise. Then a processing procedure is established suitable
for remote detection acoustic logging data. The field data processing results further show a good application
prospect of this method.
Keywords: Remote detection technology; Reflected wave logging; Anisotropic diffusion filtering; Image
denoising; Coherence enhancement

2021-07-02 收稿; 2021-12-17 定稿
国家自然科学基金项目 (42004117), 中国科学院科研仪器设备研制项目 (YJKYYQ20200072)
∗
作者简介: 傅艳莉 (1997– ), 女, 山东青岛人, 硕士研究生, 研究方向: 信号与信息处理。
通信作者 E-mail: chaoli@mail.ioa.ac.cn
†