1. College of Artificial Intelligence,Shenyang Aerospace University,Shenyang 110136,China
2. College of Computer Science,Shenyang Aerospace University,Shenyang 110136,China
Show less
文章历史+
Received
Accepted
Published
2022-06-07
Issue Date
2025-08-05
PDF (1291K)
摘要
在利用视觉检测算法进行检测与定位的过程中,当空域中目标无人机相对较小时,现有的检测算法容易受到空中其他飞行物、复杂背景和光照强度变化影响导致检测精度较低。为了解决这一问题,提出了一种基于YOLOv3改进的目标检测算法。当空域中目标无人机体积相对较小、视觉特征较弱、存在其他干扰时,通过增加主干特征提取网络对图像特征提取的层级数,提取多个不同尺度特征层进行跨层连接融合,使多个不同层级的特征层之间的语义信息联系得更加紧密,让网络模型可以学习到不同尺度目标的特征信息,以此增强检测算法对小目标无人机检测的精度。最后,利用Drone vs Birds数据集进行实验测试,所提出的算法可以有效地提高小型无人机目标的检测精度,检测速度基本满足实际要求。
Abstract
When the target UAV(Unmanned Aerial Vehicle) in the air domain is relatively small, the existing detection algorithm is prone to be affected by other flying objects in the air, complex background and light intensity changes, resulting in low detection accuracy in the process of using visual detection algorithm for detection and positioning. To solve this problem, an improved target detection algorithm based on YOLOv3 was proposed. When the target UAV in the space domain was relatively small in size, with weak visual features and other disturbances, the main feature extraction network was added to the image feature extraction level, and multiple feature layers of different scales were extracted for cross-layer connection and fusion, so that the semantic information between multiple feature layers of different levels was more closely related. The network model could learn the characteristic information of targets of different scales, so as to enhance the accuracy of detection algorithm for small target UAV detection. Finally, Drone vs Birds dataset was used for experimental testing. The algorithm can effectively improve the detection accuracy of small UAV target, and the detection speed basically meets the actual requirements.
小目标无人机的检测过程中存在无人机尺度变化比较大的问题,导致算法无法准确捕获无人机位置。本文通过延伸主干特征网络,扩大感受野,可以使算法模型学习到更多的无人机形态,从而提高检测无人机目标的精度。在检测过程中,当发生光照变化、存在复杂的背景和其他的飞行物(例如鸟类)的干扰时,会影响检测的精度。本文通过改进算法模型的特征融合方式,加强算法模型对无人机目标特征的学习,提高算法的检测精度。最后,本文介绍了算法生成无人机目标边界框的方式和算法模型的损失函数。利用Drone vs Birds数据集[18]进行训练验证,通过实验证明,相比于一般算法,本文所提算法在多种复杂场景下对小型无人机目标的检测具有更高的精度。
AnwarM Z, KaleemZ, JamalipourA.Machine learning inspired sound-based amateur drone detection for public safety applications[J].IEEE Transactions on Vehicular Technology,2019,68(3):2526-2534.
[3]
SeidaliyevaU, AkhmetovD, IlipbayevaL,et al.Real-time and accurate drone detection in a video with a static background[J].Sensors,2020,20(14):3856.
[4]
KangH G, JoungJ, KimJ,et al.Protect your sky:a survey of counter unmanned aerial vehicle systems[J].IEEE Access,2020,8: 671-710.
[5]
YangS B, LuoY, MiaoW,et al.RF signal-based UAV detection and mode classification:a joint feature engineering generator and multi-channel deep neural network approach[J].Entropy,2021,23(12):1678.
[6]
SvanströmF, Alonso-FernandezF, EnglundC.A dataset for multi-sensor drone detection[J].Data in Brief,2021,39:107521.
[7]
TahaB, ShoufanA.Machine learning-based drone detection and classification:state-of-the-art in research[J].IEEE Access,2019,7: 669-682.
[8]
MasitaK L, HasanA, ShongweT.Deep learning in object detection:a review[C]//2020 International Conference on Artificial Intelligence,Big Data,Computing and Data Communication Systems (icABCD).Durban,South Africa:IEEE,2020:1-11.
LiuS, LiG T, ZhanY F,et al.MUSAK:a multi-scale space kinematic method for drone detection[J].Remote Sensing,2022,14(6):1434.
[13]
HuY Y, WuX J, ZhengG D,et al.Object detection of UAV for anti-UAV based on improved YOLO v3[C]//2019 Chinese Control Conference (CCC).Guangzhou,China:IEEE,2019:8386-8390.
[14]
MaQ, ZhuB, ChengZ D,et al.Fast detection and recognition method of UAV in sky background[C]//Proc SPIE 11209,Eleventh International Conference on Information Optics and Photonics (CIOP 2019),Xi'an,China:Chinese laser Press,2019:36-41.
[15]
FanJ X, LiD W, WangH L,et al.UAV low altitude flight threat perception based on improved SSD and KCF[C]//2019 IEEE 15th International Conference on Control and Automation (ICCA).Edinburgh,UK:IEEE,2019:356-361.
[16]
LuoK, ZhuG B, LiY,et al.Low altitude and low speed uav identification based on hybrid model[C]//Proceedings of the 3rd International Conference on Computer Engineering,Information Science & Application Technology (ICCIA 2019. Paris,France:Atlantis Press,2019:76-87.
[17]
QueJ F, PengH F, XiongJ Y.Low altitude,slow speed and small size object detection improvement in noise conditions based on mixed training[J].Journal of Physics:Conference Series,2019,1169:012029.
[18]
ColucciaA, FascistaA, SchumannA,et al.Drone-vs-bird detection challenge at IEEE AVSS2021[C]//2021 17th IEEE International Conference on Advanced Video and Signal Based Surveillance (AVSS).Washington,DC,USA:IEEE,2022:1-8.