Deep-Sea Multibeam Sediment Classification Using Echo Waveform Features
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing seabed sediment classification methods in deep-sea environments face challenges due to the scarcity of sampling, limited characterization of acoustic wave penetration, and the inability to accurately represent mixed sediment categories, especially in low-resolution data sets.
Innovation Solution
A sediment classification method and system that extracts multidimensional features from multibeam water column data, utilizes cosine similarity for sample expansion, and employs a random forest model for decision fusion to classify discrete point clouds, constructing a classification category abundance map for complex mixed sediments.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Measurement precision
If traditional underwater in-situ sampling technology is used for sediment classification, then classification accuracy can be achieved, but the process becomes heavy, time-consuming, labor-consuming and difficult in deep water areas
Solution Approach 1:
The patent replaces traditional mechanical underwater sampling systems with an acoustic detection system (multibeam echo-sounding). The system uses acoustic waves to penetrate the water column and seabed, automatically acquiring sediment information without physical contact. This substitution eliminates the need for heavy sampling equipment and manual operations in deep water, dramatically reducing time and labor while maintaining classification accuracy through acoustic property analysis of different sediment types.
2Productivity
If multibeam backscattering intensity features are used for sediment classification, then the process becomes efficient, but the method is limited to surface layer characterization and cannot penetrate deeper seabed layers
Solution Approach 1:
The patent transitions from analyzing only the intensity magnitude of backscattering signals to analyzing the temporal dimension of the entire echo waveform. By examining the time series data from beam emission to seabed penetration, the system captures acoustic wave behavior at different depths and penetration stages. This temporal dimension analysis enables characterization of both surface and deep seabed layers simultaneously, overcoming the depth limitation of traditional intensity-only methods while maintaining high classification efficiency.
3Device complexity
If hard classification method is used where each observation unit is assigned to one pure sediment category, then the classification process is simple, but mixed sediment categories and intermediate states cannot be accurately characterized
Solution Approach 1:
The patent changes the classification output parameter from discrete categorical labels to continuous abundance proportions. Instead of assigning each observation unit to a single sediment category, the system calculates the proportion (abundance) of each sediment type within the unit. This parameter transformation enables accurate representation of mixed sediments and intermediate states, as the continuous proportions reflect the actual compositional complexity while the underlying classification algorithm remains relatively simple.
Applied Scientific Principles
This section explains which scientific principles are used to turn an abstract innovation direction into a practical engineering solution.
Function Achieved in This Case
Enhances the reliability of sediment classification results by transforming hard classification to soft classification, effectively interpreting mixed sediments in deep-sea environments, overcoming limitations of traditional sampling methods and improving data resolution.
Implementation Method 1
The internal structure and physical properties of seabed sediments are closely related to the acoustic properties such as acoustic attenuation and acoustic waves scattering of the sediment layer
Implementation Method 2
a multibeam echo-sounding system (MBES) has become the mainstream method for evaluating seabed roughness, sediment types, and benthic habitats
Data Source
AI summary
Disclosed is a sediment classification method and system based on bottom echo information of a deep-sea multibeam water column, falling within the technical field of seabed sediment classification. Multidimensional features of seabed multibeam time series are extracted by using multibeam water column data, and multidimensional features of seabed spots are constructed by fusing an angle response curve; double-waveform spectrum comparison of original sample point waveform with seabed measured point waveform is performed, and two groups of corresponding waveforms are matched by using Cosine Similarity to expand original sample points; and the reliability of classification results is improved. In the present disclosure, the transformation from hard to soft classification is realized, and the seabed habitat inversion of complex mixed sediment in a deep sea is further realized.