Research On Wave Elevation Measurement Based On Binocular Vision And 3D Visualization Technology

With the development of science and technology,mankind's demand for marine research,development and utilization is increasing,and it is necessary to monitor the marine surface environment.Wave elevation is one of the important parameters of the ocean surface environment.At present,there are many methods for measuring wave elevation,but each method has its application scope.It is difficult to deploy wave buoys in offshore operations.Radar and satellite technology is limited by resolution,and monitoring of small-scale waves on the sea surface needs to be improved.At the same time,in the study of the ocean,the 3D visualization technology can accurately reflect the distribution and contour characteristics of the sea surface waves,and provide an intuitive and reliable real situation for monitoring the surface environment of the ocean.Aiming at the above problems,this paper proposes a method based on binocular vision photogrammetry to obtain the wave elevation,and performs three-dimensional visualization of the measured water surface.First,the overall scheme of binocular vision for ocean wave measurement is proposed,the main steps are introduced,and the overall scheme block diagram is given.At the same time,the camera is calibrated with parameters,the conversion relationship between different coordinate systems in binocular vision measurement is studied,and common binocular camera calibration methods are classified.The steps and principles of Zhang Zhengyou's camera calibration method are mainly studied.Calibration experiments were performed on the camera according to the calibration principle,and internal and external parameters of the camera were obtained.Secondly,in order to solve the problem that it is difficult to obtain camera external parameters with a calibration board in outdoor experiments,a camera self-calibration method based on SURF feature matching is used to solve it.The principle of feature point extraction related to SURF and the method of fast matching feature points are introduced.For the false matching points in fast matching,the RANSAC method is proposed to eliminate the false matching points,and the corresponding program flow is designed,and the validity of the method is experimentally verified.Then,the average water surface equation and elevation are solved.The stereo matching algorithms are classified,and the advantages and disadvantages of different algorithms are listed.Through comparative analysis,a semi-global stereo matching algorithm with both speed and accuracy is selected to obtain parallax images.An analysis of the two cases in the space point reconstruction problem is given later.An RANSAC method is proposed to obtain the average water surface equation,and a corresponding program flow is designed to solve the elevation information.Finally,field experiments verify.Firstly,the realization steps of 3D visualization of water surface are introduced,and the important Delaunay triangulation in the steps is studied and analyzed.Then,the parallax image was obtained by on-site observation,and the threedimensional visualization of the water surface was completed.Then carry out the elevation measurement experiment,put forward the elevation measurement standard in this article,and introduce the experimental software for elevation measurement;carry out on-site observations in Songhua River and Majiagou River,respectively,and sample the same set of video images for 3 different frequencies.Compare elevation measurements at different sampling frequencies.The experimental results show that the three-dimensional visualization of the water surface in the two observation sites is good.The relative standard deviations of the elevations measured by the method in this paper are about 5.97% and 4.30%,respectively,which verifies the validity of the method in this paper.