Hao, P. and He, M., 2020. Ship detection based on small sample learning. In: Yang, D.F. and Wang, H. (eds.), Recent Advances in Marine Geology and Environmental Oceanography. Journal of Coastal Research, Special Issue No. 108, pp. 135–139. Coconut Creek (Florida), ISSN 0749-0208.
Ship detection plays a vital role in the management of maritime transportation. So far, many ship detection methods have been proposed since the era from traditional to deep learning. However, most of them require complicated and time-consuming data annotation tasks, and the huge amount of data also brings challenges to real-time ship detection. In order to reduce the burden of data annotation and achieve high detection accuracy and real-time performance, this article proposes a method of real-time ship detection based on small sample learning. This method mainly consists of three modules: weight sharing, filtering-RPN, and matching detector. The experimental results show that the method proposed in this article not only can detect ships effectively but also has high detection accuracy. In addition, compared with other mainstream methods under the same experimental environment, the average precision of this method is over 96.7%, which is far better than other methods. The results prove the effectiveness of this method.