[1]程期浩,陈东方,王晓峰.基于NDM-YOLOv8的无人机图像小目标检测[J].计算机技术与发展,2024,34(09):63-69.[doi:10.20165/j.cnki.ISSN1673-629X.2024.0177]
　CHENG Qi-hao,CHEN Dong-fang,WANG Xiao-feng.Small Target Detection in UAV Images Based on NDM-YOLOv8[J].,2024,34(09):63-69.[doi:10.20165/j.cnki.ISSN1673-629X.2024.0177]
点击复制

基于NDM-YOLOv8的无人机图像小目标检测()

《计算机技术与发展》[ISSN:1006-6977/CN:61-1281/TN]

卷:
34
期数:
2024年09期
页码:
63-69
栏目:
媒体计算
出版日期:
2024-09-10

文章信息/Info

Title:
Small Target Detection in UAV Images Based on NDM-YOLOv8
文章编号:
1673-629X(2024)09-0063-07
作者:
程期浩1; 2; 陈东方1; 2; 王晓峰1; 2
1. 武汉科技大学 计算机科学与技术学院,湖北 武汉 430065;2. 智能信息处理与实时工业系统湖北省重点实验室,湖北 武汉 430065
Author(s):
CHENG Qi-hao1; 2; CHEN Dong-fang1; 2; WANG Xiao-feng1; 2
1. School of Computer Science and Technology,Wuhan University of Science and Technology,Wuhan 430065,China;2. Hubei Provincial Key Laboratory of Intelligent Information Processing and Real-time Industrial Systems,Wuhan 430065,China
关键词:
无人机; 小目标检测; YOLOv8; 感受野; 特征提取
Keywords:
unmanned aerial vehicle; small target detection; YOLOv8; receptive field; feature extraction
分类号:
TP391.4
DOI:
10.20165/j.cnki.ISSN1673-629X.2024.0177
摘要:
针对无人机航拍图像中小目标实例多、目标之间存在遮挡的现象,容易造成漏检、误检等问题,提出一种新的基于非跨步动态多头结构的小目标检测算法(Non-strided Dynamic Multihead YOLOv8,NDM-YOLOv8)。 首先,设计了 SConv 模块,融合了非跨步卷积,尽可能地保留输入数据的判别特征信息,以降低小目标特征的细粒度信息的丢失;其次,设计了 C2f-LSK 模块,通过采用选择机制对空间特征进行有效加权,动态地调整感受野,灵活地捕捉不同尺度的特征和上下文信息,提高模型对小目标的关注度;最后,设计了 P2 小目标检测头,并和高层网络进行残差连接,减少小目标特征丢失,以强化算法对小目标特征的提取能力。 实验表明,NDM-YOLOv8 有效提高了对无人机图像中小目标检测精度。 在公开数据集 VisDrone2019 上,NDM-YOLOv8 比 YOLOv8n 在 mAP0. 5提高了 5. 3 百分点,mAP0. 5:0. 95上提高了 3. 3 百分点,对比其他模型,也取得了较优的检测效果,能够有效地完成无人机航拍图像中小目标检测任务。
Abstract:
In view of the phenomenon that there are many instances of small targets in UAV aerial images and there are occlusions between targets,which can easily cause problems such as missed detection and false detection,a new small target detection algorithm based on a non-strided dynamic multihead structure,NDM-YOLOv8(Non-strided Dynamic Multihead YOLOv8),is proposed. Firstly,the SConv module was designed,which integrated non-strided convolution and preserved the discriminative feature information of the input data as much as possible to reduce the loss of fine-grained information of small target features. Secondly,the C2f-LSK module was designed to effectively weight spatial features through a selection mechanism,dynamically adjust receptive fields,flexibly capture features and contextual information of different scales,and improve the model’s attention to small targets. Finally,a P2 small target detection head was designed and residual connected to the high-level network to reduce the loss of small object features and enhance the algorithm’s ability to extract small target features. Experiments show that the NDM - YOLOv8 effectively improves the detection accuracy of small targets in drone images. On the public dataset VisDrone2019,the NDM-YOLOv8 improved by 5. 3 percentage points on mAP0. 5 and 3. 3 percentage points on mAP0. 5:0. 95 compared with YOLOv8n. Compared with other models,it also achieved better detection performance and can more effectively complete the task of small target detection in UAV aerial images.

相似文献/References:

[1]张晓军,陆兴华.加入不确定扰动的无人机飞行轨迹跟踪控制[J].计算机技术与发展,2018,28(01):182.[doi:10.3969/ j. issn.1673-629X.2018.01.039]
　ZHANG Xiao-jun,LU Xing-hua.Flight Trajectory Tracking Control Algorithm for Unmanned AerialVehicle with Uncertain Disturbance[J].,2018,28(09):182.[doi:10.3969/ j. issn.1673-629X.2018.01.039]
[2]王俊明,曹科才,肖斌,等. 基于Paparazzi Lisa/M2.0的无人机自主飞行设计[J].计算机技术与发展,2016,26(05):197.
　WANG Jun-ming,CAO Ke-cai,XIAO Bin,et al. Design of Autonomous Flight for UAV Based on Paparazzi Lisa/M2.0[J].,2016,26(09):197.
[3]陆兴华. 采用波束域加权的大迎角飞行姿态控制算法[J].计算机技术与发展,2017,27(01):117.
　LU Xing-hua. High Angle of Attack Flight Attitude Control Algorithm Based on Beam Domain Weighting[J].,2017,27(09):117.
[4]陆兴华.基于波束空间二维谱峰搜索的无人机控制算法[J].计算机技术与发展,2017,27(12):52.[doi:10.3969/ j. issn.1673-629X.2017.12.012]
　LU Xing-hua.Unmanned Aerial Vehicle Control Algorithm Based on Two Dimensional Spectrum Peak Searching for Beam Space[J].,2017,27(09):52.[doi:10.3969/ j. issn.1673-629X.2017.12.012]
[5]陶晓力,武 建,杨 坤.基于无人机视觉的桥梁裂缝检测[J].计算机技术与发展,2018,28(03):174.[doi:10.3969/ j. issn.1673-629X.2018.03.037]
　TAO Xiao-li,WU Jian,YANG Kun.Bridge Crack Detection Based on Unmanned Aerial Vehicle Vision[J].,2018,28(09):174.[doi:10.3969/ j. issn.1673-629X.2018.03.037]
[6]张晨,刘宁钟.基于无人机视觉的道路违法搭建检测[J].计算机技术与发展,2018,28(07):140.[doi:10.3969/ j. issn.1673-629X.2018.07.030]
　ZHANG Chen,LIU Ning-zhong.llegal Construction Detection of Road Based on Unmanned Aerial Vehicle Vision[J].,2018,28(09):140.[doi:10.3969/ j. issn.1673-629X.2018.07.030]
[7]王立春,刘宁钟,李强懿.基于无人机航拍图像的公路标线检测算法[J].计算机技术与发展,2018,28(09):138.[doi:10.3969/ j. issn.1673-629X.2018.09.028]
　WANG Li-chun,LIU Ning-zhong,LI Qiang-yi.A Road Markings Detection Algorithm Based on Aerial Image of UAV[J].,2018,28(09):138.[doi:10.3969/ j. issn.1673-629X.2018.09.028]
[8]肖 建,梁定康,徐 威,等.一种基于无人机的违章违停自主巡检系统[J].计算机技术与发展,2019,29(12):153.[doi:10. 3969 / j. issn. 1673-629X. 2019. 12. 027]
　XIAO Jian,LIANG Ding-kang,XU Wei,et al.A Self-inspection Violation System Based on UAV[J].,2019,29(09):153.[doi:10. 3969 / j. issn. 1673-629X. 2019. 12. 027]
[9]陆兴华,明 仲,邱子琪,等.无人机低空滑翔抗攻击突防控制律优化设计[J].计算机技术与发展,2020,30(02):183.[doi:10. 3969 / j. issn. 1673-629X. 2020. 02. 035]
　LU Xing-hua,MING Zhong,QIU Zi-qi,et al.Optimal Design of Anti-attack Penetration Control Law for UAV Gliding at Low Altitude[J].,2020,30(09):183.[doi:10. 3969 / j. issn. 1673-629X. 2020. 02. 035]
[10]产世兵,刘宁钟,沈家全.基于轻量级网络的 PCB 元器件检测[J].计算机技术与发展,2020,30(10):14.[doi:10. 3969 / j. issn. 1673-629X. 2020. 10. 003]
　CHAN Shi-bing,LIU Ning-zhong,SHEN Jia-quan.PCB Component Detection Based on Lightweight Network[J].,2020,30(09):14.[doi:10. 3969 / j. issn. 1673-629X. 2020. 10. 003]
[11]陶晓力,刘宁钟,沈家全.基于深度信息融合的航拍车辆检测[J].计算机技术与发展,2019,29(09):117.[doi:10. 3969 / j. issn. 1673-629X. 2019. 09. 023]
　TAO Xiao-li,LIU Ning-zhong,SHEN Jia-quan.Aerial Vehicle Detection Based on Depth Information Fusion[J].,2019,29(09):117.[doi:10. 3969 / j. issn. 1673-629X. 2019. 09. 023]
[12]赵 侃,汪慧兰,郭娇娇,等.基于 DTA-FSAF 的无人机小目标检测研究[J].计算机技术与发展,2024,34(04):101.[doi:10. 3969 / j. issn. 1673-629X. 2024. 04. 016]
　ZHAO Kan,WANG Hui-lan,GUO Jiao-jiao,et al.Research on Small Object Detection of UAV Based on DTA-FSAF[J].,2024,34(09):101.[doi:10. 3969 / j. issn. 1673-629X. 2024. 04. 016]

更新日期/Last Update: 2024-09-10