摘要
在常规的地震勘探中,由于受到激发、接收环境或仪器等因素影响,地震数据中会不可避免地混杂随机噪声,导致有效信号被淹没,从而无法被清晰识别,因此随机噪声的压制至关重要。本次研究提出一种改进曲波变换阈值去噪结合快速非局部均值去除地震随机噪声的方法:首先将循环平移和块状复数域阈值方法引入到曲波变换阈值去噪中,改进传统的曲波阈值去噪方法,得到最佳去噪结果;然后将去除的噪声利用快速非局部均值方法进行滤波,得到有效信号;最后将上述两步得到的数据相加得到最终的去噪结果。将该方法与曲波变换阈值方法、小波阈值方法及快速非局部均值算法的去噪结果进行比较,模型试验和实际地震数据去噪效果表明,该方法去噪结果的信噪比更高,信号保真度更高。因此本文方法压制随机噪声的效果更佳。
In the conventional seismic exploration,due to the influence of excitation,receiving environment or instrument and other factors,the random noise is inevitably mixed in the seismic data,and the effective signal is often submerged,which makes it impossible to identify clearly.Therefore,the suppression of random noise is very important.This paper proposes a method to remove random noise by using modified curvelet transform threshold denoising combined with fast non-local mean(FNLM).Firstly,cycle spinning and complex block thresholding are introduced into curvelet transform threshold denoising,which improves the traditional curvelet threshold denoising method and obtains the best denoising result.Secondly,the denoising noise is filtered by FNLM to obtain an effective signal.Finally,the data obtained in the above two steps are added to obtain the final denoising result.This method is compared with the denoising results of curvelet transform threshold method,wavelet threshold method and FNLM.The model test and actual seismic data denoising effect show that the denoising result of this method has a higher signal-to-noise ratio and fidelity.Therefore,this method is more effective in suppressing random noise.
作者
孙思亮
刘怀山
Sun Siliang;Liu Huaishan(Key Lab of Submarine Geosciences and Prospecting Techniques, Ministry of Education, Ocean University of China,Qingdao Shandong 266100,China;Qingdao National Laboratory for Marine Science and Technology, Laboratory for Marine Mineral Resources, Qingdao Shandong,266071, China)
出处
《工程地球物理学报》
2021年第2期153-161,共9页
Chinese Journal of Engineering Geophysics
基金
国家自然科学基金资助项目(编号:91958206,41230318)
国家高技术研究发展计划(863计划)资助项目(编号:2013AA092501)。
