Electronic Chart Recognition Based on Ensemble Broad Learning System for Safety Analysis of Waterway Area

This paper aims to propose an electronic chart recognition method based on an ensemble broad learning system (EBLS) to improve the safety analysis ability of waterway areas. Firstly, it is constructed for a dataset containing the navigational difficulty levels of an electronic chart dataset. Secondly, the VGG16 network is employed as the preprocessing model. Finally, the preprocessed data is fed into the ensemble broad learning system to recognize the navigational difficulty levels of the electronic chart dataset. Compared with other algorithms, the experimental results show that the proposed algorithm obtains a great improvement in the accuracy of recognizing the navigational difficulty in the electronic chart dataset, thereby contributing to the enhancement of safety analysis in the waterway area.