| The multibeam echosounder(MBES)system can collect both bathymetry information and beam backscatter strength information to obtain underwater topography and sonar images.However,there are still shortcomings in the processing method of multibeam backscatter strength,which leads to errors in the obtained sonar images.In order to make the acquired multibeam backscatter strength accurately reflect the underwater substrate information,this paper studies the multibeam system calibration,acoustic transmission loss correction,nadir striped artifacts correction and backscatter strength angular response correction,and finally obtains high-quality underwater sonar image.The main work and innovation of this paper are as follows:1.A MBES system calibration is carried out by using experimental pool with known bottom substrate.MBES system is a complex system composed of multiple sensors.Both installation parameters and system parameters of multibeam system will affect the accuracy of backscatter strength.In addition,due to the different signal processing methods of different MBES systems,there is often a systematic deviation between the obtained backscatter strength of multibeam and the real substrate.Therefore,a MBES system calibration method is proposed,which uses geometric structure of the experimental pool to check the system installation parameters and a known bottom substrate to check the system parameters.There total 10 inclination planes of experimental pool and artificial designed planes are chosen as target planes used for system installation parameters calibration.Firstly,a high precision total station measures landmarks of target planes before water injection for fitting plane parameters.Secondly,the MBES system is used to acquire beam coordinates by measuring target planes after water injection.Lastly,a least squares adjustment model is established for obtaining multibeam system installation parameter error.Results show that residual standard deviation drop down to 4.8cm from8.9cm.Estimation accuracy of angle systematic deviation is about 54 ",and the estimation accuracy of displacement systematic deviation is 1.2mm.The influence of system parameters on the backscatter strength results is analyzed by measuring a known bottom substrate with different system parameters.The system deviation of the backscatter strength of the multibeam system was 122.48dB by comparing the backscatter strength collected by MBES system with optimal parameter combination with the theoretical value.2.An improved acoustic transmission loss model was proposed to solve the absorption effect of suspended particles on sound waves in turbid waters.In turbid waters,in addition to the absorption of chemical substances such as boric acid and magnesium salt in the water and the loss of sound energy caused by the expansion of sound wave itself,the viscous absorption of sound wave caused by a large number of suspended particles in the water will also cause sound energy loss.Therefore,an improved model of transmission loss considering the viscous absorption of suspended particles is proposed.In addition,the acoustic wave transmission loss is calculated layer by layer through temperature profile,and the transmission loss accumulation of each layer is taken as the overall transmission loss of the beam.The experiment verified that the correlation coefficient of backscatter strength and propagation distance obtained by this method was 0.04,that is,the backscatter strength after treatment was basically eliminated by the influence of the propagation distance.3.A weighted fitting method to correct the nadir striped artifacts is proposed for the anomaly of backscatter strength in the center of the multibeam strip.In the vicinity of the central beam of the multibeam survey lines,the echo signal is mainly reflected by specular reflection,so the backscatter strength value of this part is usually large,which is shown as the nadir artifacts in the sonar image of the measuring line.So,it cannot accurately reflect the information of underwater bottom.Therefore,a weighted fitting method to correct the nadir artifacts is proposed.Quadratic curve of RMSE for R2Sonic series and Seabat series MBES system and quadratic curve of average strength for EM2040 are fitted and the correction of the backscatter strength of this part of the beam can be obtained by comparing the strength of the nadir artifacts area with the fitting curve.4.A correction method for angular response using the three-dimensional(3D)probability density of backscatter strength is proposed in view of the imperfection of the current Angular response correction method for complex distribution of underwater substrates.The acoustic wave emitted by the MBES system is affected by the acoustic wave scattering mechanism on the seabed.As there may be many types of substrates on the bottom and their distribution is complex,the current angular response correction method is not adaptable enough to deal with the backscatter strength on the bottom of a variety of substrates with complex distribution.As a result,some errors still exist in the acquired sonar image.Therefore,an Angular response correction method which is not affected by the substrate distribution is proposed.This method firstly using the gaussian weighted average method to get average angular response curve of 50 consecutive ping pulses.A 3D probability density analysis is then carried out using the average angular response curves of single line or a region.A few angular response curves only corresponding to different bottom substrate types can be obtained from the 3D probability density results.Finally,each beam backscatter strength gets a correction based on its substrate type and corresponding angular response curve.The experimental results verify that the average mismatch value of echo strength in the overlapping area of the survey line is drop down to-0.O6dB from-0.38 and standard deviation drop down to 3.36dB from 4.95 compared to clustering method,which achieves high-quality underwater sonar image. |
PDF Full Text Request