基于YOLO算法的船舶识别定位系统

Ship Identification and Positioning System Based on YOLO Algorithm

针对船舶AIS系统船舶定位的精确性以及可靠性的不足,不能很好的适用于港口码头的自动停泊,本文提出一种基于YOLO算法的船舶识别定位系统。YOLO算法模型是由卷积层、池化层以及全连接层组合而成,是一种端到端的快速识别监测算法,该算法计算速度很快并且拥有极强的抗干扰性。通过YOLO算法该系统能够快速的识别港区内船舶的类别、型号、货物以及位置信息,减小船舶AIS系统的误差,为船舶的自动停泊、靠泊提供数据支持。

According to the inaccuracy and unreliability of ship positioning in AIS system, it cannot be applied to the automatic parking of port terminals. This paper proposes a ship identification and positioning system based on YOLO algorithm. The YOLO algorithm model is a combination of convolutional layer, pooling layer and fully connected layer. It is an end-to-end fast identification monitoring algorithm. The algorithm is fast and has strong anti-interference. The system can quickly identify the type, model, cargo and position information of ships in the port area. reduce the error of the ship’s AIS system, and provide data support for the automatic ship mooring through the YOLO algorithm.