Search

Based on a substantial number of seabed sediment samples and corresponding acoustic characteristic and physical property test data collected in the continental shelf area of the northwestern South China Sea, this study employed a sound speed ratio correction model to analyze the correlation between the sound speed ratio of seabed sediments and parameters such as porosity, density, mean grain size, and water content. This study resulted in the development of both single-parameter and double-parameter empirical formulas for the sound speed ratio of seabed sediments in this marine region. The determination coefficients (r2) for the single-parameter empirical formulas based on physical properties, excluding the mean grain size (r2 ≤ 0.75), were all above 0.78, indicating a strong correlation. Additionally, the fitting curve at 100 kHz was slightly greater than that at 50 kHz, consistent with the acoustic dispersion characteristics of sediments. To assess the applicability of the empirical formulas to different maritime areas, the formulas derived from the South Yellow Sea, East China Sea, and South China Sea were compared with those of the study area. The results indicated that the empirical formulas for the sound speed ratio established in this study have general applicability, especially for sediments with low porosity, high density, and low mean grain size. In addition, the determination coefficient of the dual-parameter empirical formula for the sound speed ratio–porosity–mean grain size is superior to that of the single-parameter empirical formula for the sound speed ratio–mean grain size. Through comparative studies with other marine areas, it was found that there are certain differences in the relationships between porosity and mean grain size established in different marine areas. Therefore, based on the rough sphere model, this study transforms the GS model into a single-parameter model built on either porosity or mean grain size. The error analysis results indicate that the single-parameter GS model has a smaller error than the measured data. However, the GS model controlled by a single parameter performs poorly compared to the predicted curves of the sound speed ratio–porosity and the sound speed ratio–mean grain size fitted in this study. This discrepancy may be attributed to the fact that the input parameters of the GS model are reference values, emphasizing the need for further research to determine suitable input parameters for the GS model in this marine area. • The empirical equations for the sound speed ratio were developed in this study in different marine regions. • The established empirical equations for the study area were compared with those for the South Yellow Sea, East China Sea, and other marine regions. • This study compared the empirical equations with the grain shearing (GS) theoretical model. [ABSTRACT FROM AUTHOR]

When backpropagation neural network (BPNN) is often applied to supervised classification, problems arise, including a slow convergence rate, local extremum, and difficulty in determining the number of hidden layers and hidden nodes that affect the classification accuracy and efficiency. These problems can be overcome by using smarter network designs. Adaptive boosting (AdaBoost), which combines multiple weak classifiers to create a strong classifier, has a strong classification advantage. In this article, we propose an acoustic seabed classification method that combines AdaBoost with the particle swarm optimization (PSO). The PSO-BP-AdaBoost algorithm uses multibeam echosounder backscatter data to solve the multiclassification problem of diverse seafloor sediment types with small differences between types. We optimize a BPNN using the PSO algorithm to obtain the optimal initial weight and threshold and combine these to form an AdaBoost strong classifier. The input data is obtained from the sonar mosaic from multibeam echosounder backscatter data collected in Jiaozhou Bay using a series of fine processing techniques. These processing techniques result in 34-dimensional (34-D) features using ReliefF analysis. The most advantageous 8-D features are used as input into the AdaBoost algorithm based on one-level decision tree, PSO-BP algorithm, support vector machine (SVM), and PSO-BP-AdaBoost algorithm. The PSO-BP-AdaBoost classification model has better classification accuracy. The overall accuracy is improved by 12.68%, 6.78%, and 3.56%, respectively, which demonstrates that the PSO-BP-AdaBoost algorithm can be effectively applied to acoustic seabed classification and identification and achieves high precision. [ABSTRACT FROM AUTHOR]