Research on Unmanned Vessel Surface Object Detection Based on Fusion of SSD and Faster-RCNN

摘要

This paper presents a real-time obstacle detection algorithm for obstacle detection on the sea surface for Unmanned Surface Vehicle (USV). The algorithm uses Kalman filtering to fuse SSD and Faster-RCNN model. Simulation tests have been taken and shown that compared with the traditional algorithm, this algorithm can significantly improve the speed, stability and accuracy.