With the development of globalization, ship detection is of great significance for water traffic. The monitoring data of the camera are used to detect the ship. This paper introduces and compares three target detection algorithms: Yolov3, Centernet and Yolov5, and adds SE (Squeeze Excitation) module to Yolov5. Firstly, we preprocess the data set, and then carry out experiments to compare the above algorithms. Finally, four evaluation parameters: Precision, Recall, mAP_0.5 (mean average precision) and mAP_05:0.95 are used to evaluate the performance of the model. At the same time, the processing time and parameters of the algorithm are compared. The results show that Yolov3 has the highest accuracy, but the parameter size is larger and SE-Yolov5 can not only consider the speed, but also maintain the above four performance indicators.
Research on Real-Time Ship Detection Using Deep Learning
2022-11-11
4254950 byte
Conference paper
Electronic Resource
English
Real-Time Video Processing for Ship Detection Using Transfer Learning
Springer Verlag | 2022
|Accurate Real-time Ship Target detection Using Yolov4
IEEE | 2021
|Near Real Time Ship Detection Experiments
British Library Conference Proceedings | 2010
|