Seabed Mixed Sediment Classification with Multi-beam Echo Sounder Backscatter Data in Jiaozhou Bay

Abstract

The multi-beam echo sounder system can not only obtain high-precision seabed bathymetry data, but also obtain high-resolution seabed backscatter strength data. A number of studies have applied acoustic remote sensing method to classify seabed sediment type with multi-beam backscatter strength data, and obtained better classification results than the traditional sediment sampling method. However, these studies mainly focus on the single type sediment classification or seabed mixed sediment classification using single beam data, not multi-beam echo sounder data. Based on backscatter strength data by a high-frequency (300 kHz) Simrad EM3000 multi-beam echo sounder and seabed sediment sampling data of Jiaozhou Bay in Qingdao, China, we establish the relation model between seabed backscatter strength and sediment type characteristics after data processing and corrections. The purpose of data processing is to diminish or weaken the influence of local bottom slope and near nadir reflection on backscatter strength data. Processed backscatter strength data through corrections and compensations can better characterize the features of the seabed sediments. Applying the back propagation neural network method based on genetic algorithm, we present a fast and accurate seabed classification method in this paper which could identify not just a single type of sediment-like rock, sand but also mixed types of sediment like sand gravel, clayey silt and sand-silt-clay.

Deep-Sea Seabed Sediment Classification Using Finely Processed Multibeam Backscatter Intensity Data in the Southwest Indian Ridge

“…Because of the absorption of sound waves by the sea, the energy of sound decays with increasing distance, and the variation of the seabed topography will affect the size of the beam irradiation area, leading to the deviation of intensity calculation. The original multibeam backscatter intensity cannot directly reflect the real seabed sediment characteristics; it must be subjected to fine postprocessing [21]. The commonly used deep-sea multibeam backscatter intensity postprocessing algorithm does not consider the effects of deep-sea acoustic signal propagation loss, seabed topography fluctuation, and central beam specular reflection on backscatter intensity.…”

“…Researchers have employed spectrum analysis [3,4], texture analysis [5][6][7][8], statistical analysis [9][10][11], cluster analysis [12], geomorphometric analysis [13][14][15], neural networks [16][17][18][19][20][21][22][23][24], and other methods to classify and identify seabed sediments. However, most of these studies focus on the nearshore shallow water areas; due to the influence of the complex marine environment in the deep sea, it is more difficult to use multibeam to classify the seabed sediments there than in shallow water, and few studies have been reported on the classification of acoustic sediments in deep-sea areas [25,26].…”

Sediment Classification of Acoustic Backscatter Image Based on Stacked Denoising Autoencoder and Modified Extreme Learning Machine

“…The random selection of the input weights and deviations causes the ELM model to generate overfitting and instability problems [41], requiring a swarm intelligence algorithm to optimize its parameters. Common swarm intelligence algorithms include the ant colony algorithm [42], GA [27], and PSO [28]. In this paper, the PSO and RELM model are combined into the MELM classifier to optimize the weights and biases of the input layer (Figure 5).…”

“…Meanwhile, swarm intelligence algorithms could be used to optimize parameters of classifiers. For example, a NN with a genetic algorithm (GA) [27] and an SVM with particle swarm optimization (PSO) [28] were used to classify the sediments in offshore areas and islands, achieving better classification results. However, the classification sensitivity of these techniques needs to be improved in order to distinguish more types of sediments and the small differences in category characteristics.…”

Multifeature Extraction and Seafloor Classification Combining LiDAR and MBES Data around Yuanzhi Island in the South China Sea

“…The methods that use MBES data for seafloor classification are primarily based on SVM, learning vector quantization (LVQ), self-organizing feature map (SOM) classifiers, and cluster analysis methods [ 8 , 9 ]. Tang et al used a Simrad EM3000 multibeam sounder (Simrad Yachting, Oslo, Norway) to collect backscatter data and in situ sediment sampling data from Jiaozhou Bay in Qingdao, China, then used the back propagation neural network (BPNN) and genetic algorithm optimization of the BPNN (GA-BPNN) for classification, and the accuracy was up to 80.2% and 85.8%, respectively [ 10 ]. Giovanni et al studied the relationships between bathymetric data, backscatter data, angle response curve and sediment particle size, and seaweed distribution.…”