Research And Implementation Of Submarine Pipeline Detection And 3D Visualization Technology

Submarine pipeline is an important transportation mode of ocean oil and gas resources.If it is suspended or even ruptured,it will bring huge economic losses and serious environmental pollution.In view of the difficulties like low visibility and high detection cost in complex seabed environment,this paper studies and realizes the detection and 3D visualization technology of submarine pipeline.Firstly,a method of automatic detection,recognition and extraction of underwater gas from multibeam sonar image is presented.The best threshold is selected by the maximum variance between classes method,and the underwater targets and water background are separated.Then,the feature selection and improved local threshold segmentation method is used,the threshold value is set for the difference image of original image and reference image,so that the threshold value is related to the position of pixel point and the local light and shade degree,and the gas is segmented from the underwater targets.This method has a good segmentation effect when the gray value difference between gas target and water background is small,which solves the problem of low efficiency and poor accuracy of artificial interpretation sonar image.The contour is extracted by mathematical morphology to realize the quantitative estimation of the location and scale of gas leakage.Secondly,the information is extracted from the side scan sonar file and a twodimensional side scan sonar image is generated.Combined with the physical and geometric relationship during detection,a method for calculating the suspended height of submarine pipeline is given,and its effectiveness and practicability are verified by the measured data.Combined with the characteristics of submarine pipeline,3D point cloud reconstruction is completed in the form of fitting circle.Then,in view of the holes in the seabed terrain point cloud where the submarine pipeline is located,combined with the characteristics of equal interval sampling of side scan sonar,the point cloud is uniformly sliced and projected,and the holes are repaired by cubic B-spline curve fitting.On this basis,an improved method is proposed,that is,before fitting,every three adjacent type value points in two-dimensional projection point column of each slice are supplemented to solve the problems of sparse point column and uneven density after fitting.In the process of 3D mesh reconstruction of seafloor terrain,the insertion point location and local optimization procedure of point by point insertion algorithm for constructing Delaunay triangular grid are improved respectively,which not only improves the construction efficiency,but also improves the mesh model quality,especially in steep terrain area.Finally,three-dimensional models of submarine pipeline and terrain are generated respectively,which are matched by geographical location to realize three-dimensional fusion and simulation,which intuitively shows the overall trend,location,abnormal suspended area information and visualization of surrounding environment of submarine pipeline within the detection sea area,and realizes the scientific monitoring of submarine pipeline.