Extraction Of Formation Boundaries And Dips Using Edge Detection Techniques

In the process of seismic exploration,oil and gas are usually in the developed faults and fissures,then for the detection of faults and fractures is particularly important.If you can apply the edge detection technology in image processing into the interpretation of seismic profiles,this will greatly improve the efficiency and accuracy of seismic data interpretation.In fact,there are several problems in seismic exploration.1.Large faults and fissures can be identified by 3D exploration technology,but it is difficult to identify small faults and fractures.2.As the amount of 3D seismic data is huge,if we rely on the artificial interpretation,the efficiency will be very low.3.There are a lot of stratigraphic boundaries in the seismic profile,and some of them are relatively small differences in the impedance is not easy to distinguish.The geometrical properties of earthquake seismic data may gather different information,and geological characteristics of subtle will become amplified.The edge detection technology in computer image processing can not only filter and enhance the edges of different directions,but also suppress non-edge points.In order to study the problems mentioned above,this paper will introduce edge detection technology and seismic geometrical attributes,and considers the problems of pixel amplitude,rotation invariance and Gauss smoothing denoising of color images.The theoretical basis of formation boundary,seismic attribute and edge detection technology are introduced.The Sobel operator,Canny algorithm and GST algorithm are used to extract the edge information of the ordinary image,the geological structure of the formation boundary of the seismic image,the faults and folds.Finally,the Gauss filter is used to smooth the seismic image.The research work is as follows: 1.the basic principles of formation boundary and seismic attributes is introduced,and the basic principles and formulas of three dip attribute extraction algorithms are given.2.Introduces the basic principles and algorithm of three classic algorithms in edge detection,and compares their characteristics in three typical images.3.Apply three improved edge detection operators to two classic seismic images.Analyze the effect of the three operators in the formation boundary and fault extraction.When the continuity faults presence,the fault can be extracted by using the dip attribute extraction.4.Introduce and analyze the principle of Gauss smoothing filter and the influence of noise on edge detection.Compare the effect of edge detection before and after Gauss smoothing.Through the research above,we can find the directional gradient template of Sobel operator can extract geological structures in different directions.The GST algorithm has a good effect in the formation boundary extraction.It has good continuity and high simulation,which can enhance the main boundary and weaken the secondary boundary.It is better to use the dip attribute image for the continuity fault edge extraction.At the same time,non-maximum and double threshold suppression can be used to smooth the dip attribute image.The influence of noise on edge detection is relatively large,especially the dip attribute map.It is the key to preserve the effective edge to give a reasonable Gauss filter parameter.