面向航天光学遥感复杂场景图像的舰船检测

Ship detection for complex scene images of space optical remote sensing

基于深度学习的目标检测算法直接应用于航天光学遥感(Space Optical Remote Sensing,SORS)复杂场景图像中会出现舰船目标检测效果不佳的问题。针对该问题,本文以近海复杂背景的密集排布舰船和远海多干扰中小目标舰船为检测对象,提出一种改进的YOLOX-s(Improved You Only Look Once-s,IM-YOLO-s)算法。在特征提取阶段,引入CA位置注意力模块,分别从高度与宽度两个方向对目标信息的位置进行权重分配,提高了模型的检测精度;在特征融合阶段,将BiFPN加权特征融合算法应用到IM-YOLO-s的颈部结构,进一步提升了小目标船只检测精度;在模型优化训练阶段,以CIoU损失替代IoU损失、以变焦损失替代置信度损失、调整类别损失权重,增大了正样本分布密集区域的训练权重,减少了密集分布船只的漏检率。另外,在HRSC2016数据集的基础上增加额外的离岸中小舰船图像,自建了HRSC2016-Gg数据集,HRSC2016-Gg数据集增强了海上船只及中小像素船只检测时的鲁棒性。通过数据集HRSC2016-Gg评测算法性能,实验结果表明:IM-YOLO-s对于SORS场景舰船检测的召回率为97.18%,AP@0.5为96.77%,F1值为0.95,较原YOLOX-s算法分别提高了2.23%,2.40%和0.01。这充分表明该算法可以对SORS复杂背景图像进行高质量舰船检测。

When deep-learning-based target detection algorithms are directly applied to the complex scene images generated by space optical remote sensing(SORS),the ship target detection effect is often poor.To address this problem,this paper proposes an improved YOLOX-S(IM-YOLO-s)algorithm,which uses densely arranged offshore ships with complex backgrounds and ships with multi-interference and small targets in the open sea as detection objects.In the feature extraction stage,the CA location attention mod⁃ule is introduced to distribute the weight of the target information along the height and width directions,and this improves the detection accuracy of the model.In the feature fusion stage,the BiFPN weighted feature fusion algorithm is applied to the neck structure of IM-YOLO-s,which further improves the detec⁃tion accuracy of small target ships.In the training stage of model optimization,the CIoU loss is used to re⁃place the IoU loss,zoom loss is used to replace the confidence loss,and weight of the category loss is ad⁃justed,which increases the training weight in the densely distributed areas of positive samples and reduces the missed detection rate of densely distributed ships.In addition,based on the HRSC2016 dataset,addi⁃tional images of small and medium-sized offshore ships are added,and the HRSC2016-Gg dataset is con⁃structed.The HRSC2016-Gg dataset enhances the robustness of marine ship and small and medium-sized pixel ship detection.The performance of the algorithm is evaluated based on the dataset HRSC2016-Gg.The experimental results indicate that the recall rate of IM-YOLO-s for ship detection in the SORS scene is 97.18%,AP@0.5 is 96.77%,and the F1 value is 0.95.These values are 2.23%,2.40%,and 0.01 higher than those of the original YOLOX-s algorithm,respectively.This indicates that the algorithm can achieve high quality ship detection from SORS complex background images.