近岸海域浅地层剖面数据处理技术及应用

Data Processing Technology of Shallow Strata Profile and Its Application in Coastal Area

浅地层剖面探测技术是应用声学原理高效获取海底浅部地质结构的探测技术,在海洋科学研究、海洋工程调查中发挥着重要作用。本文针对近岸海域浅地层剖面随机噪声和多次波发育的问题开展研究,首先分析了发育成因和影响因素,在此基础上,采用空间域滤波和频率域滤波相结合的方式对随机噪声衰减,通过中值滤波法对浅地层剖面影像的预处理,消除孤立、随机的噪声点;再采用二维矩形预测算子在频率域滤波,滤除噪声频带,衰减背景噪声,提高信噪比;然后,利用一次波与多次波之间的周期相关性,采用改进型的预测反褶积对多次波处理,并给出预测算子长度和预测步长的选取方式,提高浅地层剖面数据成像质量。在近岸海域实际工程应用中,通过本方法和常规方法的效果对比发现,本方法可有效衰减随机噪声,压制多次波效果明显,展示更多隐藏的海底管线和浅部地质结构,说明了本方法的有效性、可靠性和实用性。

The shallow strata profile as a detection technology to apply the acoustics principle to the efficient collection of the shallow geological structure of the seabed plays an important role in marine scientific research and marine engineering investigation.This paper fucuses on the development of random noise and multiple waves in shallow strata profiles in coastal areas.Firstly,after analyzing the causes and influencing factors of their development,it combines spatial domain filtering and frequency domain filtering to attenuate random noise.Through the preprocessing of shallow strata profile by median filtering method,the isolated and random noise points are eliminated,then the two-dimensional rectangular prediction operator in the frequency domain is used to filter out the noise frequency band,attenuate the background noise,and improve the signal-to-noise ratio.And then the periodic correlation between the primary wave and the multiples,the improved prediction deconvolution are used to process the multiple waves.Finally,the selection method of the prediction operator length and the prediction step is used to improve the imaging quality of shallow stratigraphic profile data.In practical engineering applications of offshore waters,after comparing the effect of this method and the conventional method,we find this method can effectively attenuate random noise,demonstrating an obvious advantegeous effect of suppressing multiple waves,displaying more hidden submarine pipelines and shallow geological structure information,which shows its effectiveness,reliability and practicality.