Research on sea target detection method based on deep learning
Su Zhijin, Kang Lan, Zhao Zhanzhuang · 2023
Artificial intelligence technology has made rapid development in recent years, in which the deep learning field has shown strong applicability. In the field of national defense, the war mode has also changed from mechanized combat mode to intelligent combat mode. YOLOv5 network based on target classification and regression is widely used in the field of target detection. This paper proposes a ship detection method based on YOLOv5x video image. We use photographic data annotation and self-made data sets to obtain training data based on video images, and combine them with YOLOv5s model. We use the transfer learning method to train the model and modify the weight parameters. The experimental results show that the detection accuracy of the algorithm can reach 95% for the target class ships and ships in the video images.