Research on Ship Target Detection in SAR Image Based on Improved YOLO v3 Algorithm

2021 
Synthetic aperture radar (SAR) has the characteristics of all-weather, all day and multi-application observation. In recent years, ship target detection based on SAR image has been widely concerned by relevant researchers. In this paper, based on the object detection method of deep learning algorithm, the detection performance of ship target in SAR image is studied by using YOLOv3 algorithm. In order to solve the problem of increasing error rate of ship target detection in complex background, YOLOv3 algorithm is improved. By adding a preprocessing layer in the front of the input layer, the accuracy of the ship detection is improved from 92.17% to 95.80%. The algorithm can be applied to other target detection in SAR image.
    • Correction
    • Source
    • Cite
    • Save
    • Machine Reading By IdeaReader
    4
    References
    0
    Citations
    NaN
    KQI
    []