Yantang river tidal bore visual identification early warning device and method

The invention discloses a Yantang river tidal bore visual identification early warning device and method. According to the method, the deep learning method is adopted to train the images acquired and marked by the Yantang river eagre so as to obtain the eagre feature model, visual recognition of the eagre is realized through close-range recognition of eagre features, and the arrival position recognition precision of the eagre is improved. The method uses the tidal bore visual identification result and the tidal bore forecast data, obtains the precise forecast time of the tidal bore reaching each downstream point location through calculation, and improves the precision of the tidal bore forecast time. The mode that the eagle identification edge device and the sound-light-electricity early warning device are separated is adopted, Lora is used for networking communication, early warning is carried out on the eagle arrival time, and the implementation cost is remarkably reduced. 本发明公开了一种钱塘江涌潮视觉识别预警装置及方法。本发明采用深度学习