Font Size: a A A

Research On Filtering Method Based On Preserving Edge For Seismic Data

Posted on:2018-08-10Degree:MasterType:Thesis
Country:ChinaCandidate:S T ZhangFull Text:PDF
GTID:2370330596469762Subject:Electronic and communication engineering
Abstract/Summary:PDF Full Text Request
With the increasing refinement of seismic exploration,the quality request of seismic data is higher and higher to person.Because of the fault,channels,and crack contain significant geological information,we need to protect them through some methods.At this stage,the seismic data processing method is more concentrated in the aspect of noise suppression,preserving the fault feature is still in the stage of development.Therefore,this paper puts forward to the following three kinds of edge filtering methods based on the noise attenuation of seismic data and preserving the fault features,which can achieve noise attenuation and protect geological information of the fault.In this paper,main work is as follows.First of all,this paper introduced the trilateral filtering to the seismic data processing with the foundation of deeply understanding of the bilateral filtering,then proposed a trilateral filtering method based on pseudo median,a trilateral filtering method based on pseudo median and controllable kernel function.The algorithm can not only suppress random noise in seismic data to improve signal to noise ratio,but also protect event preferably so that preserving important information with geological explanation significance.Secondly,in terms of the phenomenon that seismic events are affected by static shifts,this thesis presented statics preserving projection filtering,which can estimate projection filters and static shifts simultaneously in order to perform noise attenuation.With the result of the synthetic seismic data and real seismic data,statics preserving projection filtering can suppress noise meanwhile protect the event,which verifies the validity and feasibility of the algorithm.At last,this thesis put forward to a strategy to suppress noise based on an enhanced sparse representation in transform domain.We realized the algorithm by the three successive steps:3D transformation of a group,shrinkage of the transform spectrum,and inverse 3D transformation.Applying the method to seismic data,the experimental results showed that the strategy achieved good effect on not only the objective peak signal to noise ratio but also subjective visual quality.It can better preserve seismic event,which improve the efficiency and the quality of seismic data processing.
Keywords/Search Tags:seismic denoising, preserving edge filter, trilateral filtering, controllable kernel function, projection filtering, collaborative filtering
PDF Full Text Request
Related items