Abstract: The accurate identification of seabed substrate types is essential for understanding the distribution of benthic marine communities and planning the sustainable development of marine resources, and machine learning algorithms are an effective means to obtain substrate types. In view of the limitations of substrate classification of single acoustic data of islands and reefs, fusion of multispectral remote sensing data provides a new idea to solve the limitation. In this study, a multi-source seafloor fusion of multispectral remote sensing data and multibeam data based on feature selection and Genetic Algorithm Extreme Gradient Boosting (GA-XGBoost) is proposed as a seafloor classification method. Firstly, the WorldView-2 multispectral data and multibeam data are preprocessed to unify the geographic coordinate system and align the spatial resolution; then the spectral features of the multispectral image, the topographic features of the bathymetry data, and the texture features of the backscatter intensity are extracted to form an 18-dimensional feature parameters, and the 12-dimensional optimal feature subsets are selected from the 18-dimensional features based on the XGBoost (Extreme Gradient Boosting) algorithm combined with forward stepwise feature selection. Then, the GA-XGBoost classification model was established, and the model was trained and tested using the data from multiple sources that provide a new idea to solve the limitation, respectively, and is compared with the BP (Back Propagation neural network (BP), GA-BP (Genetic Algorithm- Back Propagation neural network, GA-BP) and XGBoost classification algorithms. Finally, the optimal GA-XGBoost classification model was applied to classify and visualize the substrate in the whole study area. The experimental results show that the overall accuracy of the method in seabed substrate classification reaches 91.23%, with a Kappa coefficient of 0.87 and an F1 score of 91.18%, which is significantly better than the single data source input and the comparison algorithm. Indicating that GA-XGBoost can provide a new way for the rapid and accurate classification of seafloor sediment.