Research on Seabed Sediment Classification Based on the MSC-Transformer and Sub-Bottom Profiler

This paper proposed an MSC-Transformer model based on the Transformer’s neural network, which was applied to seabed sediment classification. The data came from about 2900 km2 of area northern slope South China Sea. Using submarine backscattering intensity and depth obtained by sub-bottom profiler, combined with latitude longitude information, a dataset Sea constructed. Moreover, using MSC-Transformer, accurate identification judgment types such as calcareous bio-silt, bio-clay silt, silty sand, medium sand gravel were realized. Compared conventional deep network CNN, RNN, etc., shows advantages when shallow sea region confirms feasibility validity provides reliable tool for classification in field marine science. completeness accuracy good performance provide solid foundation scientificalness practicability study.