High-precision Underwater Quality Classification Method Based On Multi-beam Water Depth And Backscatter Strength

Multi-beam Echosounder System is an important equipment for acquiring underwater topographic data.Therefore,it is of great significance to analyze and study multi-beam sounding technology so that find out the problems and then solve them.Based on the data of multi-beam Echosounder system and analysis of relevant research at domestic and foreign,the work on the quality control of multi-beam Bathymetric data,the data analysis of ALL original documents and the seabed classification technology in this paper are as follows:(1)The composition and working principle of the Multi-beam Echosounder System are introduced.The method of multi-beam beam steering and the multi-beam bottom detection method are discussed and analyzed.(2)Analyze and correct the sounding error of multi-beam,including the measurement error caused by ocean noise,sonar parameter deviation and sound speed profile errors,and corresponding correction method,at the same time,the comprehensive processing of sounding data is present.The multi-beam Backscatter Strength data is interpreted,and the Backscatter Strength gain correction and the seafloor incident angle correction method in multi-beam are introduced and analyzed.(3)The conventional filtering method of multi-beam sounding data is introduced.The traditional filtering is low automation,and the polynomial surface fitting is not consistent with the sea bottom.The mathematical model of the trend term usually cannot accurately represent the overall trend of the seabed topography,and how to wholely present the problems of systematic and randomness of multi-beam bathymetric data of non-stationary terrain on the seabed.For these probles,This study proposed the robust least square collocation method based on the improved multi-faceted function.The method uses the exponential function as the kernel function of the multi-quadric function,Trend terms were eliminated by the optimal unbiased linear estimation of the improved least square collocation.In the same time,parameter values of the empirical covariance function and the final results could be estimated accurately.The least squares configureuration eliminates gross or outliers present in the multi-beam data.The experimental results show that the method can more accurately characterize the overall trend change of seabed topography,and ensure the quality of Multi-beam Bathymetric data.(4)The data structure of ALL original file is analyzed in detail,and the original data is decoded by MATLAB,and the relevant water depth data and Backscatter Strength data are extracted.An Backscatter Strength image is generated by normalizing the Backscatter Strength data,geostatistical position reduction.(5)Introduce and analyze of methods for submarine sediment classification,Combined with the development of submarine sediment detection technology,This study proposed classification method based on the Bayes classification rule for water depth data fusion Backscatter Strength data.The method uses the bathymetry data,the bathymetry derivative and the like combined with the Backscatter Strength data to adopt the Bayesian classification rule for the rapid classification of the seafloor sediment,and compares it with the results of the classification using Backscatter Strength data alone.The experimental results show that the accuracy of the classification accuracy of the proposed method is higher,the overall classification accuracy reaches 91.7%,which is 8.4 percentage points higher than the classification accuracy of the Backscatter Strength data alone.The Kappa coefficient evaluation result is also higher than the single use Backscatter Strength data,The evaluation result reached 0.876,which further developed the development of the submarine sediment classification by exploring various data of the multi-beam Echsounder system.