基于改进YOLOv7的低空飞行物目标检测方法

被引:2
|
作者
甄然 [1 ]
刘雨涵 [1 ]
孟凡华 [1 ]
刘颖 [1 ]
王文林 [1 ]
李素康 [1 ]
赵昊天 [1 ]
机构
[1] 河北科技大学电气工程学院
关键词
YOLOv7; 目标检测; 无人机; 低空飞行物; 注意力机制;
D O I
暂无
中图分类号
TP391.41 [];
学科分类号
080203 ;
摘要
目前无人机检测技术应用广泛,但无人机在执行目标检测任务中可能遭遇多种空中障碍物,这些目标具有成像小、像素特征少和相对速度变化快等检测难点,针对此类目标引起的漏检误检问题,提出了一种基于改进YOLOv7算法的低空飞行物目标检测算法。在传统YOLOv7算法的基础上,在Head网络引入SimAM注意力机制,该机制与现有的通道和空间注意力模块相比,同时考虑空间和通道维度信息,且不在原始网络中添加额外参数;在主干网络中结合ConvNeXt网络,提出CvNX模块,降低网络计算量,并保留目标特征;用SIoU-Loss代替原有的CIoU-Loss,提高算法收敛速度;在图像后处理阶段使用SIoU-NMS,减少遮挡导致的目标漏检。在自有低空飞行物数据集上实验结果表明,改进的YOLOv7算法平均精度(Average Precision, AP)达到97.1%,相比YOLOv7算法,平均精度均值(mean Average Precision, mAP)提高了1.7%,且误检、漏检率低,达到了在复杂背景下检测低空飞行物目标的要求。
引用
收藏
页码:633 / 643
页数:11
相关论文
共 8 条
  • [1] 基于优化YOLOv4算法的低空无人机检测与跟踪
    赵玥萌
    刘会刚
    [J]. 激光与光电子学进展, 2022, (12) : 397 - 406
  • [2] 基于A-YOLOv5s的机场小目标检测方法
    刘闪亮
    吴仁彪
    屈景怡
    李云龙
    [J]. 安全与环境学报, 2023, 23 (08) : 2742 - 2749
  • [3] Real-time vehicle detection algorithm based on a lightweight You-Only-Look-Once (YOLOv5n-L) approach[J] Bie Minglin;Liu Yanyan;Li Guoning;Hong Jintao;Li Jin Expert Systems With Applications 2023,
  • [4] A deep learning approach incorporating YOLO v5 and attention mechanisms for field real-time detection of the invasive weed Solanum rostratum Dunal seedlings[J] Wang Qifan;Cheng Man;Huang Shuo;Cai Zhenjiang;Zhang Jinlin;Yuan Hongbo Computers and Electronics in Agriculture 2022,
  • [5] ETL-YOLO v4: A face mask detection algorithm in era of COVID-19 pandemic[J] Kumar Akhil;Kalia Arvind;Kalia Aayushi Optik 2022,
  • [6] YOLOv4_Drone: UAV image target detection based on an improved YOLOv4 algorithm[J] Tan Li;Lv Xinyue;Lian Xiaofeng;Wang Ge Computers and Electrical Engineering 2021,
  • [7] Distance-IoU Loss: Faster and Better Learning for Bounding Box Regression[J] Zhaohui Zheng;Ping Wang;Wei Liu;Jinze Li;Rongguang Ye;Dongwei Ren Proceedings of the AAAI Conference on Artificial Intelligence 2020,
  • [8] Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks.[J] Ren Shaoqing;He Kaiming;Girshick Ross;Sun Jian IEEE transactions on pattern analysis and machine intelligence 2017,