基于YOLOv11n的无人机航拍小目标检测模型

李学威; 毛小坤; 孙滨; 黄宇达

doi:10.13756/j.gtxyj.2026.250012

您当前的位置：

首页 >

文章列表页 >

基于YOLOv11n的无人机航拍小目标检测模型

光传感与检测技术 | 更新时间：2026-10-21

- 基于YOLOv11n的无人机航拍小目标检测模型
- UAV Aerial Small Object Detection Algorithm based on YOLOv11n
- 光通信研究 2026年第5期页码：186-195
- 作者机构：
  
  1.周口职业技术学院　信息工程学院，河南　周口　466000
  2.中国人民解放军　山东省淄博军分区，山东　淄博　255000
  3.郑州工业应用技术学院　信息工程学院，郑州　451150
- 作者简介：
  
  李学威（1983-），男，河南周口人。副教授，硕士，主要研究方向为计算机应用。
  毛小坤，工程师。E-mail：1157488992@qq.com
- 基金信息：
  
  河南省科技厅科技攻关资助项目(232102210200;242102210101;242102211034);河南省高等学校重点科研资助项目(23B520036);河南省工程技术研究中心资助项目(豫科实[2024]1号)
- DOI：10.13756/j.gtxyj.2026.250012
  中图分类号： TN911.73-34;TP391.41
- 收稿：2025-01-13，
  
  修回：2025-01-23，
  
  纸质出版：2026-10-10
- 稿件说明：
移动端阅览
李学威，毛小坤，孙滨，等. 基于YOLOv11n的无人机航拍小目标检测模型[J]. 光通信研究，2026(5): 186-195.

Li X W, Mao X K, Sun B, et al. UAV Aerial Small Object Detection Algorithm based on YOLOv11n[J]. Study on Optical Communications, 2026(5): 186-195.
李学威，毛小坤，孙滨，等. 基于YOLOv11n的无人机航拍小目标检测模型[J]. 光通信研究，2026(5): 186-195. DOI： 10.13756/j.gtxyj.2026.250012.

Li X W, Mao X K, Sun B, et al. UAV Aerial Small Object Detection Algorithm based on YOLOv11n[J]. Study on Optical Communications, 2026(5): 186-195. DOI： 10.13756/j.gtxyj.2026.250012.

摘要

目的

针对无人机（UAV）航拍图像中小目标检测面临的背景复杂、尺度变化大导致的检测精度较低和推理速度较慢等问题，文章提出了一种基于YOLOv11n模型的改进模型YOLOv11n-DPD，旨在提升UAV航拍图像中小目标的检测精度，并实现模型轻量化和推理加速，便于其在资源受限的嵌入式设备上部署运行。

方法

首先，在跨阶段局部瓶颈双卷积（C3k2）模块中引入可变形卷积（DCN），增强模型在复杂背景下对多尺度目标的特征提取能力，特别是对于形变显著的小目标。其次，针对UAV航拍图像中小目标占比较多等情形，新增第2预测层（P2）即小目标检测层并去除了第5预测层（P5）即大目标检测层，有效提高了模型对小目标特征的捕获精度，同时减少了大目标检测层的冗余计算。然后，提出了一种结合聚焦损失（Focal Loss）与距离交并比损失（DIoU Loss）的Focal-DIoU损失函数，解决了小目标检测中各目标类别不平衡的问题，并优化了边界框的定位精度。最后，采用层自适应幅值剪枝（LAMP）方法对模型进行轻量化处理，通过适当调整剪枝加速比，自适应地按层评估并删除冗余连接，减少模型整体计算量并加速推理过程。

结果

基于VisDrone2019数据集的实验结果表明，YOLOv11n-DPD相较于原YOLOv11n模型，

mAP

50（即交并比阈值为0.5时的平均精度均值）指标提升了6.5%，达到40.7%，增强了对UAV航拍小目标的检测精度。同时，YOLOv11n-DPD在推理

速度方面，在Ubuntu和Jetson平台上分别提高了32和6 FPS，显著加快了推理过程。且该模型的模型大小和参数量分别减少了3.87和2.02 MB，证明了模型在轻量化方面的优势。此外，与现有主流的小目标检测方法相比，YOLOv11n-DPD在检测精度、推理速度和计算资源上综合表现更好，具有显著优势。

结论

文章所提YOLOv11n-DPD模型通过引入DCN、优化小目标检测层结构、改进损失函数和进行模型剪枝，提高了检测精度、加速了推理过程并优化了模型大小，与现有的小目标检测方法相比表现更优，尤其适用于UAV航拍小目标检测任务和嵌入式平台的部署推理。

Abstract

Objective

To address the problems of low detection accuracy and slow inference speed caused by complex backgrounds and significant scale variations in small object detection of Unmanned Aerial Vehicle (UAV) aerial images

this paper proposes an improved algorithm named YOLOv11n-DPD

based on the YOLOv11n model. The goal is to enhance the detection accuracy of small objects in UAV aerial images while achieving model lightweighting and inference acceleration

making it suitable for deployment on resource-constrained embedded devices.

Methods

First

Deformable Convolution (DCN) is introduced into the CSP Bottleneck with 2 Convolutions (C3k2) module to enhance the model's ability to extract features of multi-scale objects in complex backgrounds

especially for small objects with significant deformation. Second

according to the high proportion of small objects in UAV aerial images

a new Prediction Layer 2(P2) is added as the small object detection layer

while the Prediction Layer 5(P5) for large object detection is removed. This improves the capture accuracy of small object features and reduces redundant computation in large object detection. Then

a Focal-DIoU loss function combining Focal Loss and Distance Intersection over Union (DIoU) Loss is proposed to alleviate class imbalance in small object detection and optimize the localization accuracy of bounding boxes. Finally

Layer-Adaptive Magnitude-based Pruning (LAMP) is employed to lighten the model by adaptively evaluating each layer and removing redundant connections

thereby reducing the overall computational load and accelerating the inference process.

Results

Experimental results on the VisDrone2019 dataset show that YOLOv11n-DPD outperforms the original YOLOv11n model

with a 6.5% improvement in

mAP

namely the mean average precision at an intersection over union threshold of 0.5

reaching 40.7%. This significantly improves the detection accuracy of small objects in UAV aerial images. In addition

YOLOv11n-DPD increases the inference speed by 32 and 6 FPS on the Ubuntu and Jetson platforms

respectively. The model size and parameter count are reduced by 3.87 and 2.02 MB

respectively

demonstrating the advantage of the proposed algorithm in lightweighting. Furthermore

compared with existing mainstream small object detection methods

YOLOv11n-DPD achieves better overall performance in detection accuracy

inference speed

and computational resource consumption.

Conclusion

The proposed YOLOv11n-DPD algorithm improves detection accuracy

accelerates inference

and optimizes model size by introducing DCN and model pruning

optimizing the small object detection layer structure

and improving the loss function. It is especially suitable for UAV aerial small object detection tasks and deployment inference on embedded platforms.

关键词

Keywords

references

张明淳 , 牛春晖 , 刘力双 , 等 . 用于无人机探测系统的红外小目标检测模型 [J ] . 激光技术 , 2024 , 48 ( 1 ): 114 - 120 .

Zhang M C , Niu C H , Liu L S , et al . Infrared Small Target Detection Algorithm for UAV Detection System [J ] . Laser Technology , 2024 , 48 ( 1 ): 114 - 120 .

王晓燕 , 王禧钰 , 李杰 , 等 . 无人机视角下的道路损伤检测模型MAS-YOLOv8n [J ] . 光电工程 , 2024 , 51 ( 10 ): 240170 .

Wang X Y , Wang X Y , Li J , et al . MAS-YOLOv8n Road Damage Detection Algorithm from the Perspective of Drones [J ] . Opto-Electronic Engineering , 2024 , 51 ( 10 ): 240170 .

肖恒树 , 李军营 , 梁虹 , 等 . 基于改进YOLOv8和无人机遥感影像的大田烟株数量检测 [J ] . 电子测量技术 , 2024 , 47 ( 9 ): 163 - 171 .

Xiao H S , Li J Y , Liang H , et al . Detection of Tobacco Plant Numbers in Large Fields based on Improved YOLOv8 and UAV Remote Sensing Imagery [J ] . Electronic Measurement Technology , 2024 , 47 ( 9 ): 163 - 171 .

张润梅 , 肖钰霏 , 贾振楠 , 等 . 改进YOLOv7的无人机视角下复杂环境目标检测模型 [J ] . 光电工程 , 2024 , 51 ( 5 ): 240051 .

Zhang R M , Xiao Y F , Jia Z N , et al . Improved YOLOv7 Algorithm for Target Detection in Complex Environments from UAV Perspective [J ] . Opto-Electronic Engineering , 2024 , 51 ( 5 ): 240051 .

宋晔 , 吴一全 . 基于无人机航拍的风力发电机叶片表面缺陷检测综述 [J ] . 仪器仪表学报 , 2024 , 45 ( 10 ): 1 - 25 .

Song Y , Wu Y Q . Review of Wind Turbine Blade Surface Defect Detection based on UAV Aerial Photography [J ] . Chinese Journal of Scientific Instrument , 2024 , 45 ( 10 ): 1 - 25 .

肖振久 , 吴正伟 , 张杰浩 , 等 . 自适应前景聚焦无人机航拍图像目标检测 [J ] . 光电工程 , 2024 , 51 ( 9 ): 240149 .

Xiao Z J , Wu Z W , Zhang J H , et al . Adaptive Foreground Focusing for Target Detection in UAV Aerial Images [J ] . Opto-Electronic Engineering , 2024 , 51 ( 9 ): 240149 .

李茂森 , 梁四幺 , 郭东海 , 等 . 基于Mask-RCNN模型的无人机巡检影像船只目标检测方法研究 [J ] . 测绘与空间地理信息 , 2023 , 46 ( 11 ): 99 - 102 .

Li M S , Liang S Y , Guo D H , et al . Research on Ship Target Recognition Algorithm based on Mask-RCNN Algorithm in UAV Inspection Images [J ] . Geomatics & Spatial Information Technology , 2023 , 46 ( 11 ): 99 - 102 .

许子阳 . 基于改进Faster R-CNN的航拍小目标检测方法 [J ] . 船舶标准化与质量 , 2024 ( 4 ): 30 - 37 .

Xu Z Y . Aerial Small Target Detection Method based on Improved Faster R-CNN [J ] . Shipbuilding Standardization & Quality , 2024 ( 4 ): 30 - 37 .

张波 , 刘隽 . 改进YOLOv5s的轻量化无人机小目标检测 [J ] . 电光与控制 , 2024 , 31 ( 11 ): 83 - 89 .

Zhang B , Liu J . Improved YOLOv5s for Lightweight Unmanned Aerial Vehicle Small Target Detection [J ] . Electronics Optics & Control , 2024 , 31 ( 11 ): 83 - 89 .

黄毅 , 周纯 , 刘欣军 , 等 . 基于YOLOv10的无人机复杂背景下多尺度检测模型 [J ] . 光通信研究 , 2025 ( 5 ): 240158 .

Huang Y , Zhou C , Liu X J , et al . Multi-Scale Detection Model for Complex Backgrounds in UAV Images based on YOLOv10 [J ] . Study on Optical Communications , 2025 ( 5 ): 240158 .

杨海涛 , 赵俊羽 , 王瑞 , 等 . 基于DBB-YOLOv10s的无人机小目标检测 [J/OL ] . 红外技术 . ( 2024-11-27 ) [ 2025-01-13 ] . https://link.cnki.net/urlid/53.1053.TN.20241126.1129.002 https://link.cnki.net/urlid/53.1053.TN.20241126.1129.002 .

Yang H T , Zhao J Y , Wang R , et al . UAV Small Object Detection based on DBB-YOLOv10s [J/OL ] . Infrared Technology . ( 2024-11-27 ) [ 2025-01-13 ] . https://link.cnki.net/urlid/53.1053.TN.20241126.1129.002 https://link.cnki.net/urlid/53.1053.TN.20241126.1129.002 .

Khanam R , Hussain M . YOLOv11: An Overview of the Key Architectural Enhancements [EB/OL ] . ( 2024-10-23 ) [ 2025-01-13 ] . https://arxiv.org/abs/2410.17725 https://arxiv.org/abs/2410.17725 .

Zhu X , Hu H , Lin S , et al . Deformable ConvNets V2: More Deformable, Better Results [C ] // 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR) . Long Beach, CA, USA : IEEE , 2019 : 9300 - 9308 .

Xu K , Wang Z , Geng X , et al . Efficient Joint Optimization of Layer-Adaptive Weight Pruning in Deep Neural Networks [C ] // 2023 IEEE/CVF International Conference on Computer Vision (ICCV) . Paris, France : IEEE , 2023 : 17401 - 17411 .

彭晏飞 , 陈炎康 , 赵涛 , 等 . 基于重参数化注意力的无人机航拍目标检测方法 [J ] . 电光与控制 , 2024 , 31 ( 9 ): 81 - 86 .

Peng Y F , Chen Y K , Zhao T , et al . Detection of Object in UAV Aerial Photography based on Reparameterized Attention [J ] . Electronics Optics & Control , 2024 , 31 ( 9 ): 81 - 86 .

李云文 , 冯宇平 , 安文志 , 等 . 基于改进YOLOv8的无人机图像小目标检测模型 [J ] . 高技术通讯 , 2024 , 34 ( 7 ): 765 - 775 .

Li Y W , Feng Y P , An W Z , et al . Small Object Detection Algorithm for Unmanned Aerial Vehicle Image based on Improved YOLOv8 [J ] . Chinese High Technology Letters , 2024 , 34 ( 7 ): 765 - 775 .

浏览量

下载量

CSCD

CNKI被引量

文章被引用时，请邮件提醒。

提交

工具集

关联资源

EGW-YOLOv8n：一种检测卫星缺陷的轻量级算法模型

卫星互联网的高效频谱共享研究综述

光纤传感技术在电网安全监测领域的应用

超表面光场调控技术及应用（特邀）

基于混沌加密的无人机可见光安全传输系统