科技创新与应用2025，Vol.15Issue(28)：23-32,10.DOI:10.19981/j.CN23-1581/G3.2025.28.005
基于YOLOv11-DeepSORT改进的遥感图像舰船多目标跟踪算法
摘要
Abstract
Multi-object tracking of ships in remote sensing images plays a crucial role in marine monitoring,environmental protection and other related fields.Currently,detection-based multi-object tracking(MOT)methods dominate the field,where tracking performance heavily depends on the accuracy of the detection network.To address the issue of low tracking performance caused by insufficient detection accuracy in existing algorithms,this paper proposes an improved YOLOv11 object detection algorithm based on the RMS-FPN feature enhancement network.Additionally,LDConv is integrated into the backbone network,and two specialized modules are designed for ship detection in remote sensing images:the SSFT and the SDFT.Compared to the original YOLOv11,the improved method achieves an accuracy increase of 11.0%,a recall improvement of 5.0%,an mAP50 increase of 10.4%and an mAP50-95 increase of 5.2%.Furthermore,the feature extraction network of DeepSORT is optimized by replacing the original ResNet with EfficientNet,leading to a 29.92%improvement in MOTA,a 4.14%increase in IDF1,and a reduction of 53 IDs compared to the baseline.关键词
舰船检测/舰船跟踪/YOLOv11/DeepSORT/特征增强网络Key words
ship detection/ship tracking/YOLOv11/DeepSORT/feature enhanced network分类
信息技术与安全科学引用本文复制引用
苏小璇,齐向阳,范怀涛..基于YOLOv11-DeepSORT改进的遥感图像舰船多目标跟踪算法[J].科技创新与应用,2025,15(28):23-32,10.基金项目
国家重点研发计划(E2BD160204) （E2BD160204）
