Dual view fusion detection method for event camera detection of unmanned aerial vehicles

Li Miao; Chen Nuo; An Wei; Li Boyang; Ling Qiang; Li Weixing

doi:10.12086/oee.2024.240208

Abstract

With the widespread application of low-altitude drones, real-time detection of such slow and small targets is crucial for maintaining public safety. Traditional cameras capture image frames with a fixed exposure time, which makes it challenging to adapt to changes in lighting conditions, resulting in the detection of blind spots in intense light and other scenes. Event cameras, as a new type of neuromorphic sensor, sense differences in external brightness changes pixel by pixel. They can still generate high-frequency sparse event data under complex lighting conditions. In response to the difficulty of adapting image-based detection methods to sparse and irregular data from event cameras, this paper models the two-dimensional object detection task as a semantic segmentation task in a three-dimensional spatiotemporal point cloud and proposes a drone object segmentation model based on dual-view fusion. Based on the event camera collecting accurate drone detection datasets, the experimental results show that the proposed method has the optimal detection performance while ensuring real-time performance, achieving stable detection of drone targets.

Keywords

FullText(HTML)

1. 引　言

目前，基于事件相机的目标检测性能最优方法是将事件数据映射为类似图像帧的表示，然后进行后续处理。然而这种方法并不适合于远距离无人机等小目标的检测。由于远距离的无人机目标尺寸很小，在像平面中目标没有任何形态特征，和噪声十分相似，网络无法有效地学习到目标特征，所以很难直接将目标与噪声进行区分。为了实现运动小目标的高效检测，本文将事件数据视为三维空间中的点云。具体来说，事件相机输出的事件流可以看做是一种由二维空间位置信息和时间信息组成的点云，与激光雷达点云相比，这种由事件点组成的特殊点云不含目标的深度信息，但是时间信息组成了其第三维。本文将这种特殊的点云称为事件点云，而基于事件的无人机小目标检测可以视为从事件点云中分割具有线状拓扑结构的无人机轨迹。

传统的检测技术，如声波探测^[14]、无线电探测^[15]和雷达探测^[16]等，虽然具备较高的检测精度和范围，但往往伴随着高昂的设备成本和复杂的配置要求，限制了其在某些场景下的应用。相比之下，基于机器视觉的检测技术以其成本低廉、易于配置的优势，逐渐成为低空无人机检测领域的研究热点。

传统基于帧的相机在极端光照条件下(如过曝或黑夜场景)性能受限^[17]。这种限制直接影响了无人机目标的有效捕捉与识别。为了克服这一缺点，事件相机^[18]作为一种新型的神经形态传感器，凭借其独特的优势，为低空无人机检测技术的发展提供了新思路。动态视觉传感器(DVS, dynamic vision sensor)及其地址事件表示(AER, address-event representation)机制，为处理高速、高动态范围视觉信息提供了一种高效的方法。在AER中，每个事件点都包含了关键信息：事件发生的像素位置、时间戳以及事件极性，这些信息对于捕捉和分析快速移动或变化的目标非常有效。

低空空域的开放为无人机技术的广泛应用提供了广阔空间，无人机逐渐被应用于各种行业^[1]，如远程监控^[2]、精准农业^[3]、地质勘探^[4]、气象观测^[5]、电力巡检^[6]、应急救灾^[7]、快递物流^[8]及影视拍摄^[9]等。然而，无人机数量的激增及其使用行为的难以规范，不仅加剧了空域管理的难度，还引发了诸多安全隐患和社会问题。无人机的不当使用，如非法侵入禁飞区、侵犯个人隐私和干扰航空交通等，已成为制约无人机产业健康发展的关键因素。因此，探索并发展一种能够实时、准确地检测与跟踪低空空域内无人机活动的技术体系，对于维护公共安全、保障个人隐私和优化空域资源配置具有重要意义^[10-13]。

根据事件数据的不同表征方式，基于事件相机目标检测可以分为基于稀疏表示和基于稠密表示的检测方法。基于稀疏表示的方法直接处理事件数据，这些方法将事件数据视为非结构化数据，包括通过脉冲神经网络^[19-20](SNN)直接处理原始事件和通过图神经网络^[21-22](GNN)对时空动态进行建模。然而，这些方法受到对专用硬件的依赖或性能不足的限制。为了解决这些问题，基于稠密表示的方法将稀疏事件转换为类似图像的格式，如基于极性对事件积累生成事件帧，然后采用通用的目标检测器进行检测^[23]。时间表面(TS)将每个像素值存储为该像素位置最近事件的时间戳^[24]。然而，由于时间维度的压缩，这两种方法在目标检测任务中的性能有限。为了保留事件流的时间信息，体素网格通过插值将事件堆叠到多个时间窗口中，将事件流转换为4D张量^[25]。EventPillars方法^[26]和ASTMNet方法^[27]中设计了可学习的编码器，它们通过网络端到端的学习事件表示，在基于事件的对象检测任务中具有较好的性能。最近，GET方法^[28]采用了一种基于组的视觉变换器骨干网络，并基于Transformer设计了目标检测网络，在事件对象检测和分类任务中展现了最先进的性能。

2. 事件相机介绍

2.1 事件相机与传统相机的区别

原始事件数据在时空维度上表示为异步二进制离散点 $E(x,y,t,p)$ ，其中： $x\in [1，W]，y\in [1，H]$ 表示空间维度上的位置， $W，H$ 表示事件数据的宽度和高度范围，t表示时间戳， $p\in \left[\mathrm{0,1}\right]$ 表示亮度变化的极性值(即 $p=0$ 表示像素亮度下降， $p=1$ 表示增加)。事件相机与传统相机的输出差异如图1所示，传统相机等时间间隔的输出场景图像，每一张图像记录了某一时刻的场景状态；而事件相机并无帧的概念，它异步的输出稀疏事件点，当场景有运动时，它输出事件点，当场景静止时，事件相机无任何输出。相比于传统相机，事件相机具有以下优势：1)低冗余：只关注于运动的目标，它只输出场景中运动部分的信息，因此其数据量远低于传统相机；2)低功耗：由于只对感知变化区域，事件相机功耗极低，通常为毫瓦级，仅为传统相机的十分之一甚至是百分之一；3)高动态范围：可达120 dB，这为事件相机在强光和低光场景下捕获目标信号奠定了基础。

事件相机作为一种新型的动态视觉传感器，异步捕获像素级亮度变化。事件相机的高时间分辨率和异步性显示出捕捉具有快速运动和高动态范围场景的巨大潜力，其应用非常广泛^{[18, 29-30]}，如自动驾驶汽车、机器人技术和安全监控。与以固定帧率捕获整个场景详细纹理的传统静态相机(如RGB和红外相机)不同，事件相机以异步的方式对辐射强度变化的像素进行二元响应，专注于捕捉移动对象的运动和结构。

Figure 1. Full-Size Img PowerPoint

Comparison of output between event camera and traditional camera

2.2 事件相机无人机检测数据集

事件相机由于具有信息冗余低、功耗低和动态范围高的优点，特别适合用于低空安全监测，采用事件相机可以实现长时间、全天候的低空无人机检测。

由于至今尚无公开的基于事件相机的无人机目标检测数据集，因此本实验组自行采集与构建数据集进行模型的训练和评估。构建的低空无人机数据集Ev-UAV共包含60段事件相机记录的无人机飞行数据，每段数据持续时间10~20 s。该数据集包含各种飞行姿态(如盘旋、倾斜和高速起降等)以及各种场景(湖边、城市和树林等)。数据集逐事件点的对目标和背景进行了标注。部分数据可视化见图2。数据集按照4∶1∶1的比例划分为了训练集、验证集和测试集。

Figure 2. Event camera drone detection dataset. (a) Event point cloud, with blue background and red drone target; (b) Visible light image frame; (c) Event frame, with the target within the red box

Figure 2. Full-Size Img PowerPoint

Event camera drone detection dataset. (a) Event point cloud, with blue background and red drone target; (b) Visible light image frame; (c) Event frame, with the target within the red box

3. 基于事件点云分割的无人机小目标检测算法

由于远距离的无人机目标尺寸很小，在像平面中目标没有任何形态特征，和噪声十分相似，所以很难直接将目标与噪声进行区分。本文将每个事件视为xyt空间中的一个三维点，这样小目标在空间中的运动轨迹表现为一条连续的曲线，而噪声由于是随机出现，不具有运动连续性，所以在三维空间中表现为一些杂乱无规则的点。所提出的方法基于二者在三维空间中的这种区别，通过分割三维空间中的连续的曲线，实现对目标的检测。所提方法的输入并不是转化为帧的事件图像，而是针对一段时间内的所有事件，基于目标的运动轨迹对目标和噪声进行区分。

图3展示了所提出的无人机小目标检测方法。它是一个双分支网络，上半部分为体素特征提取分支，下半部分为点特征提取分支，两个分支之间在不同的阶段进行交互融合。体素分支采用类似UNet^[31]的结构，首先使用一个输入卷积从原始输入中提取上下文信息，然后执行四个下采样阶段，最后执行四个上采样阶段以恢复原始输入状态。点云分支包含四层MLP，来提取不同维度的逐点特征。体素特征与点特征的融合依次发生在输入卷积、第四次下采样、第二次上采样和最后一次上采样阶段之后。

体素特征提取分支能够捕获到点云数据的全局上下文信息和空间结构，而点特征提取分支则通过多层感知机(MLP)提取每个点的局部和细节特征。两者的结合实现了全局与局部、结构与细节特征的互补，有助于更全面地提取目标特征。此外，多源特征的融合使得模型在面对复杂环境和小目标检测时具有更强的鲁棒性。即使部分特征受到噪声的影响，其他特征仍能提供有效的信息支持。

Figure 3. Full-Size Img PowerPoint

Point-voxel fusion segmentation network

3.1 点特征提取分支

点特征提取分支由四层MLP组成，每个MLP层由一个线性拟合和一个非线性激活函数组成，单层的MLP可以描述为： ${X_{{\mathrm{out}}}}{\text{ = }}f(W{X_{{\mathrm{in}}}} + b)$ ，其中 ${X_{{\mathrm{in}}}}$ 为该层的输入，W和b为神经元之间的权值和偏置，函数f为激活函数，通常为sigmoid函数或tanh函数。

分支中的四层MLP的维度分别为：32、256、128和32，即可以得到四个不同维度的点特征 ${X_i} \in {\mathbb{R}^{N \times d}}，d = \{ 32,256,128,32\}$ 。

为了从事件流中提取有效的点特征，本文设计了一个由多层MLP组成的特征提取分支。该分支首先将事件流划分为具有固定时间间隔的时间窗口，每个窗口内的事件点集合作为处理单元。给定事件流E，首先将其划分为具有固定时间间隔的时间窗口 $E = \left\{ {{\varepsilon _1},{\varepsilon _2}, \cdots ,{\varepsilon _n}} \right\}$ ，作为基本的处理单元。点特征提取分支的输入可以描述为： ${\varepsilon _i} = \{ {e_k} = \left( {{x_k},{y_k},{t_k},{p_k}} \right) |k = 1,2, \cdots ,N \}$ ，其中，每个事件点表示为空间坐标 $\left(x,y\right)$ 和时间坐标t，以及事件的极性p，N为点的数量。

3.2 体素特征提取分支

Figure 4. Full-Size Img PowerPoint

Event point cloud voxelization

体素特征提取分支以事件体素为输入，由原始事件点云进行离散化生成，每个体素单元的分辨率为2 pixel×2 pixel×10 ms，这种表示方式不仅减少了数据冗余，还便于捕捉局部时空特征，如所示。体素特征提取分支的输入可以描述为： $V \in {\mathbb{R}^{\tfrac{H}{2} \times \tfrac{W}{2} \times \tfrac{T}{{10}}}}$ 。该分支是一个类似于UNet的结构，通过逐步下采样以捕获高层次的语义信息，并通过逐步上采样以恢复空间分辨率并融合多尺度特征。具体而言，该分支包含一个输入层，四个下采样和四个上采样阶段，这9个阶段的维度分别为 32、64、 128、 256、 256、 128、 128、 64和32。下采样每个阶段的特征都通过跳跃连接与对应的上采样阶段特征连接。

3.3 双分支特征融合模块

其中： ${{\boldsymbol{X}}_{{\mathrm{con}}}}$ 和 ${{\boldsymbol{X}}_{{\mathrm{add}}}}$ 分别为级联和加法融合特征向量，由于点特征的多样性，不同视图分支的特征在重要性上并不一致，但是加法和级联特征融合策略忽略了每个特征向量的有用性，在融合过程中将大量无用特征与有用特征结合。门控机制可以通过测量每个特征的重要性来自适应地聚合信息，受其启发，本文设计了一种基于门控机制的多视图分支特征融合模块，该模块可以描述为

X_{door} = ψ \cdot X_{p} + (1 - ψ) \cdot X_{v} .

${X_{{\text{door}}}} = \psi \cdot {X_p} + (1 - \psi ) \cdot {X_v} \;.$

双分支特征融合的核心任务是在大量无用信息的干扰下将有用信息聚合在一起。加法和级联是聚合多个特征的常用操作，但它们都会受到许多非信息特征的影响。它们可以表述为

在点分支网络中，对于每一点，都会学习到一个“点特征”对应每个点的细节特征。在体素分支网络中，体素化将整个事件点云平均的划分为不同的子空间，因此一个子空间内包含多个事件点，这样提取的方式具有更大的感受野，可以更好地捕获全局特征。双分支特征融合模块将包含每个点细节特征的“点特征”和包含更多全局特征的“体素特征”在网络的不同阶段进行融合，使得网络可以有效地学习到由粗到细的特征，实现更好的分割效果。

ψ = s o f t m a x (c o n v (c o n c a t (X_{p}, X_{v}))),

$\psi = softmax (conv(concat({X_p},{X_v}))),$

对于某一事件点 ${e_i}$ ，来自点云特征提取分支的特征可表示为 $X_i^p \in {\mathbb{R}^d}$ ，来自体素的特征可表示为 $X_i^v \in {\mathbb{R}^d}$ 。 $X_i^p$ 通过不同的MLP层计算可得，不同的层具有不同的维度d。 $X_i^v$ 根据该点在体素的位置确定，即 $X_i^v = V(x,y,t)$ 。

Figure 5. Full-Size Img PowerPoint

Point-voxel feature interaction fusion module

该模块首先将两个视图的特征拼接，经过3×3卷积和softmax后转化为概率权重 $\psi$ 。概率权重表示两种在不同空间位置的特征的重要性。然后对相应通道的概率权重 $\psi$ 分离，并将分离后的概率权重与原始的点特征和体素特征分别进行加权，得到具有不同重要性的点特征和体素特征。最后，经过门控加权的两个视图特征向量相加融合得到最终的特征融合结果 ${X_{{\text{door}}}}$ 。在特征融合结果中，门控机制凸出了更重要的特征，抑制了不重要的特征。融合过程如图5所示。

{\boldsymbol X}_{add} = \sum_{i = 1}^{L} X_{i},

${{\boldsymbol{X}}_{{\text{add}}}} = \sum\limits_{i = 1}^L {{X_i}} ,$

{\boldsymbol X}_{con} = c o n c a t (X_{1}, \dots, X_{L}),

${{\boldsymbol{X}}_{{\text{con}}}} = concat({X_1}, \cdots ,{X_L}),$

4. 结果与分析

4.1 评价指标

P_{f} = \frac{F P}{T P + F N} .

${P_{\text{f}}} = \frac{{FP}}{{TP + FN}}\;.$

P_{d} = \frac{T P}{T P + F N} .

${P}_{\text{d}}=\frac{TP}{TP+FN}.$

AP 是根据 Precision-Recall 曲线计算的面积。它对不同召回率下的 Precision 进行插值，并计算插值曲线下的面积。AP 是评估目标检测算法在不同召回率下的综合性能指标。

I o U = \frac{T P}{T P + F P + F N} ，

$IoU=\frac{TP}{TP+FP+FN}\text{，}$

当分割的事件点和真值轨迹的交并比为80%以上时，认为对目标轨迹检测成功，标记为正确预测的轨迹TP。若将噪声错误检测为轨迹实例，认为对目标轨迹检测错误，标记为错误预测的轨迹FP。若将真实轨迹预测为背景噪声，则将该轨迹标记为错误预测的背景FN。轨迹检测率P_d指的是正确检测的轨迹实例占全部真值轨迹实例的比例，定义为

本文将事件目标检测任务视为事件点云分割任务，将点云IoU作为评价指标。为了方便与基于帧的目标检测方法进行对比，本文同时将AP作为评价指标。本文直接检测目标运动轨迹，因此将轨迹检测率和虚警率也作为评价指标。点云IoU表示预测的分割结果与真实标签的重叠程度。IoU的计算方法是将预测结果和真实标签的区域进行交集和并集运算，然后计算它们的比值。IoU值越大预测点云与真值点云之间的重叠度越高，分割效果越好，可表示为

其中：TP表示正确预测的目标，FP表示错误预测的目标，FN表示为错误预测的背景。

轨迹虚警率P_f指的是错误检测的轨迹实例占全部真值轨迹实例的比例，定义为

4.2 实验训练设置

对提出的网络进行训练，随机初始化模型参数，采用Adam优化器进行优化，每次输入0.5 s的事件数据，训练批大小为16，动量为0.9，衰减系数设置为0.0002，最大迭代次数为80000。初始学习率设置为0.0002，迭代次数到达60000时，学习率线性衰减到0.000002，以促进损失函数进一步收敛。实验在单张NVIDIA 3090 GPU上进行，完成一次完整的训练大约需要10 h。

4.3 实验结果与分析

4.3.1 对比实验

Experimental results of different algorithms on Ev-UAV dataset

不同算法在Ev-UAV上的实验结果

方法	框架	${P}_{{\mathrm{d}}}$ /%	${P}_{{\mathrm{f}}}$ /%	AP	IoU	推理时间/ms
SSD	帧	88.3(+8.2)	13.3(−10.1)	62.7(+6.8)	—	18.7
FastRCNN	帧	88.7(+7.8)	13.5(−10.3)	63.2(+6.3)	—	19.5
YOLOV7	帧	89.8(+6.7)	10.3(−7.1)	65.3(+4.2)	—	16.5
RVT	帧	90.3(+6.2)	10.1(−6.9)	65.7(+3.8)	—	10.5
SAST	帧	90.1(+6.4)	10.5(−7.3)	65.5(+4.0)	—	13.2
DNA-Net	帧	90.5(+6.0)	9.91(−6.7)	65.9(+3.6)	—	20.3
OSCAR	帧	90.3(+6.2)	10.2(−7.0)	66.3(+3.2)	—	25.3
Pointnet	点云	91.3(+5.2)	8.35(−5.1)	66.1(+3.4)	66.7(+3.6)	19.8
Pointnet++	点云	93.5(+3.0)	7.33(−4.1)	67.3(+2.2)	67.5(+2.8)	25.3
Ours	点云	96.5	3.21	69.5	70.3	15.2

CSV Show Table

选取三组具有代表性的场景，利用不同的检测算法对同一场景进行检测，可视化结果如图6所示。从图中可以看出，基于帧的算法YOLOv7出现了漏检，且出现了较多虚警，基于点云的算法PointNet++无漏检，但是存在少量的虚警，而所提算法可以在无虚警率情况下有效地检测出所有小目标无人机。

对比实验结果如所示，从中可以看到：1)相比于基于帧的方法，基于点的方法性能普遍更优。这主要是因为远距离无人机目标尺寸较小，没有明显的形状轮廓等特征，模型难以有效地检测到目标；而基于点云的方法可以直接分割目标运动轨迹特征，因此具有更好的性能。即使是专为小目标检测设计的基于帧的网络也具有较低的检测率和较高的虚警率。2)多视图融合方法优于单一视图的方法。尽管使用了多视图融合，但由于紧凑的网络设计，本文的PVNet仍然显示出非常有竞争力的运行延迟。相比于次优的方法Pointnet++，所提方法的检测率 ${P}_{\rm{d}}$ 提高了3%，虚警率 ${P}_{\rm{f}}$ 下降了4.1%，mAP提高了2.2%，IoU提高了2.8%，而运行时间缩短了10.1 ms。

为了验证所提算法对低空无人机目标的检测能力，分别对通用目标检测器 (YOLOv7^[32]、 SSD^[33]、FastRCNN^[34])、点云目标分割网络(Pointnet^[35]、Pointnet++^[36])、基于事件相机的目标检测网络(RVT^[37]、SAST^[38])以及小目标检测网络(DNA-Net^[39]、OSCAR^[40])四类网络在Ev-UAV数据集上进行了对比实验。点云目标分割网络的输入为时间 $\mathrm{\Delta }t$ 内的事件点云。为了适应网络的输入，通用目标检测器和小目标检测网络的输入为事件帧。事件帧L将时间 $\mathrm{\Delta }t$ 内的同一像素的事件进行积累，像素 $(i,j)$ 处的值为该时间段内产生的事件数量，可以表示为

为了比较的公平性，对于所有的对比方法， $\mathrm{\Delta }t$ 均与本文所提网络保持一致，设置为0.5 s。

L (i, j) = \sum_{t}^{t + Δ t} e (i, j, t, p) .

$L(i,j) = \sum\limits_t^{t + \Delta t} {e(i,j,t,p)} .$

Figure 6. Visual comparison of detection results. The red box represents the correct detection result, the yellow box represents false alarm, and the green box represents missed detection

Figure 6. Full-Size Img PowerPoint

Visual comparison of detection results. The red box represents the correct detection result, the yellow box represents false alarm, and the green box represents missed detection

4.3.2 消融实验

The impact of different views. V and V+ represent different voxel resolutions: V represents 4 pixel × 4 pixel × 20 ms, V+ represents 2 pixel × 2 pixel × 10 ms

不同视图的影响。V和V+表示不同的体素分辨率: V表示4 pixel×4 pixel×20 ms，V+表示2 pixel×2 pixel×10 ms

方法	${P}_{\rm{d}}$ /%	${P}_{\rm{f}}$ /%	AP	IoU/%	推理时间/ms
P	89.6 (+6.9)	9.86(−6.65)	66.3 (+3.2)	63.2 (+7.1)	7.2
V	90.6 (+5.9)	8.98 (−5.77)	65.7 (+3.8)	63.8 (+6.5)	10.2
V+	92.5 (+4.0)	5.33 (−2.12)	66.8 (+2.7)	64.7 (+5.6)	12.3
PV	94.3 (+2.2)	4.32 (−1.11)	67.2 (+2.3)	69.2 (+1.1)	19.2
PV+	96.5	3.21	69.5	70.3	20.3

CSV Show Table

为了验证多视图融合的有效性，对单个视图和双视图融合进行了消融实验，结果如所示。从表中可以看出，相比于仅使用单视图，所提方法的检测率 ${P}_{\rm{d}}$ 最高可以提高6.9%，虚警率 ${P}_{\rm{f}}$ 可以降低6.6%，IoU最高可以提高7.1%，而推理时间依旧满足实时检测要求。以上结果说明视图融合的方法在可接受的复杂性开销下可以有效地提高性能。

1) 视图的有效性

为了验证门控机制对多视图融合的有效性，采用不同的融合方法进行了消融实验，结果如所示。可以看到，门控融合方法与加法、乘法以及拼接相比，检测率 ${P}_{\rm{d}}$ 分别提高了6.3%，4.7%，2.7%，虚警率 ${P}_{\rm{f}}$ 分别降低了5.4%，4.4%，3.1%，IoU分别提升了3.7%，6.1%和5%，这说明了门控融合机制对多视图融合的有效性。

The effectiveness of different fusion methods

不同融合方法的有效性

方法	${P}_{\rm{d}}$ /%	${P}_{\rm{f}}$ /%	AP	IoU
Add	90.2 (+6.3)	8.63 (−5.42)	67.3 (+2.2)	66.8 (+3.7)
Multi	91.8 (+4.7)	7.68 (−4.47)	66.9 (+2.6)	64.2 (+6.1)
Concat	93.8 (+2.7)	6.38 (3.17)	67.6 (+1.9)	65.3 (+5.0)
Ours	96.5	3.21	69.5	70.3

CSV Show Table

2) 融合方法的有效性

5. 结　论

本研究针对神经形态传感器(事件相机)在低空无人机探测中面临的目标尺寸小、特征少、事件稀疏不规则等问题，突破传统的图像帧处理思路，将无人机目标检测任务建模为三维点云语义分割任务，设计了一种基于双视图融合的事件无人机目标分割模型和一种用于融合双视图特征的门控融合机制。实验中对自建的事件相机无人机目标检测数据集的定量和定性评估证明，与当前最先进的目标检测和点云分割网络相比，所提方法在保持低虚警的情况下具有最高的检测性能。

利益冲突

所有作者声明无利益冲突

References (40)

References

[1]	Bouguettaya A, Zarzour H, Kechida A, et al. Vehicle detection from UAV imagery with deep learning: a review[J]. IEEE Trans Neural Netw Learn Syst, 2022, 33(11): 6047−6067. DOI: 10.1109/TNNLS.2021.3080276 CrossRef Google Scholar
[2]	陈海永, 刘登斌, 晏行伟. 基于IDOU-YOLO的红外图像无人机目标检测算法[J]. 应用光学, 2024, 45(4): 723−731. Chen H Y, Liu D B, Yan X W. Infrared image UAV target detection algorithm based on IDOU-YOLO[J]. Journal of Applied Optics, 2024, 45(4): 723−731. Google Scholar
[3]	韩江涛, 谭凯, 张卫国, 等. 协同随机森林方法和无人机LiDAR空谱数据的盐沼植被“精灵圈”识别[J]. 光电工程, 2024, 51(3): 230188. DOI: 10.12086/oee.2024.230188 Han J T, Tan K, Zhang W G, et al. Identification of salt marsh vegetation “fairy circles” using random forest method and spatial-spectral data of unmanned aerial vehicle LiDAR[J]. Opto-Electron Eng, 2024, 51(3): 230188. DOI: 10.12086/oee.2024.230188 CrossRef Google Scholar
[4]	Park S, Choi Y. Applications of unmanned aerial vehicles in mining from exploration to reclamation: a review[J]. Minerals, 2020, 10(8): 663. DOI: 10.3390/min10080663 CrossRef Google Scholar
[5]	Sziroczak D, Rohacs D, Rohacs J. Review of using small UAV based meteorological measurements for road weather management[J]. Prog Aerosp Sci, 2022, 134: 100859. DOI: 10.1016/j.paerosci.2022.100859 CrossRef Google Scholar
[6]	Wang Z Y, Gao Q, Xu J B, et al. A review of UAV power line inspection[C]//Proceedings of 2020 International Conference on Guidance, Navigation and Control, Tianjin, 2022: 3147–3159. https://doi.org/10.1007/978-981-15-8155-7_263. Google Scholar
[7]	Khan A, Gupta S, Gupta S K. Emerging UAV technology for disaster detection, mitigation, response, and preparedness[J]. J Field Robot, 2022, 39(6): 905−955. DOI: 10.1002/rob.22075 CrossRef Google Scholar
[8]	Li Y, Liu M, Jiang D D. Application of unmanned aerial vehicles in logistics: a literature review[J]. Sustainability, 2022, 14(21): 14473. DOI: 10.3390/su142114473 CrossRef Google Scholar
[9]	Mademlis I, Mygdalis V, Nikolaidis N, et al. High-level multiple-UAV cinematography tools for covering outdoor events[J]. IEEE Trans Broadcast, 2019, 65(3): 627−635. DOI: 10.1109/TBC.2019.2892585 CrossRef Google Scholar
[10]	奚玉鼎, 于涌, 丁媛媛, 等. 一种快速搜索空中低慢小目标的光电系统[J]. 光电工程, 2018, 45(4): 170654. DOI: 10.12086/oee.2018.170654 Xi Y D, Yu Y, Ding Y Y, et al. An optoelectronic system for fast search of low slow small target in the air[J]. Opto-Electron Eng, 2018, 45(4): 170654. DOI: 10.12086/oee.2018.170654 CrossRef Google Scholar
[11]	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算法[J]. 光电工程, 2024, 51(5): 240051. DOI: 10.12086/oee.2024.240051 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-Electron Eng, 2024, 51(5): 240051. DOI: 10.12086/oee.2024.240051 CrossRef Google Scholar
[12]	陈旭, 彭冬亮, 谷雨. 基于改进YOLOv5s的无人机图像实时目标检测[J]. 光电工程, 2022, 49(3): 210372 DOI: 10.12086/oee.2022.210372 Chen X, Peng D L, Gu Y. Real-time object detection for UAV images based on improved YOLOv5s[J]. Opto-Electron Eng, 2022, 49(3): 210372 DOI: 10.12086/oee.2022.210372 CrossRef Google Scholar
[13]	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算法[J]. 激光技术, 2024, 48(1): 114−120. DOI: 10.7510/jgjs.issn.1001-3806.2024.01.018 Zhang M C, Niu C H, Liu L S, et al. Infrared small target detection algorithm for UAV detection system[J]. Laser Technol, 2024, 48(1): 114−120. DOI: 10.7510/jgjs.issn.1001-3806.2024.01.018 CrossRef Google Scholar
[14]	Sedunov A, Haddad D, Salloum H, et al. Stevens drone detection acoustic system and experiments in acoustics UAV tracking[C]//Proceedings of 2019 IEEE International Symposium on Technologies for Homeland Security, Woburn, 2019: 1–7. https://doi.org/10.1109/HST47167.2019.9032916. Google Scholar
[15]	Chiper F L, Martian A, Vladeanu C, et al. Drone detection and defense systems: Survey and a software-defined radio-based solution[J]. Sensors, 2022, 22(4): 1453. DOI: 10.3390/s22041453 CrossRef Google Scholar
[16]	de Quevedo Á D, Urzaiz F I, Menoyo J G, et al. Drone detection and radar-cross-section measurements by RAD-DAR[J]. IET Radar Sonar Navig, 2019, 13(9): 1437−1447. DOI: 10.1049/iet-rsn.2018.5646 CrossRef Google Scholar
[17]	Gallego G, Delbrück T, Orchard G, et al. Event-based vision: a survey[J]. IEEE Trans Pattern Anal Mach Intell, 2020, 44(1): 154−180. DOI: 10.1109/TPAMI.2020.3008413 CrossRef Google Scholar
[18]	Shariff W, Dilmaghani M S, Kielty P, et al. Event cameras in automotive sensing: a review[J]. IEEE Access, 2024, 12: 51275−51306. DOI: 10.1109/ACCESS.2024.3386032 CrossRef Google Scholar
[19]	Paredes-Vallés F, Scheper K Y W, De Croon G C H E. Unsupervised learning of a hierarchical spiking neural network for optical flow estimation: from events to global motion perception[J]. IEEE Trans Pattern Anal Mach Intell, 2020, 42(8): 2051−2064. DOI: 10.1109/TPAMI.2019.2903179 CrossRef Google Scholar
[20]	Cordone L, Miramond B, Thierion P. Object detection with spiking neural networks on automotive event data[C]// Proceedings of 2022 International Joint Conference on Neural Networks, Padua, 2022: 1–8. https://doi.org/10.1109/IJCNN55064.2022.9892618. Google Scholar
[21]	Li Y J, Zhou H, Yang B B, et al. Graph-based asynchronous event processing for rapid object recognition[C]//Proceedings of 2021 IEEE/CVF International Conference on Computer Vision, Montreal, 2021: 914–923. https://doi.org/10.1109/ICCV48922.2021.00097. Google Scholar
[22]	Schaefer S, Gehrig D, Scaramuzza D. AEGNN: asynchronous event-based graph neural networks[C]//Proceedings of 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition, New Orleans, 2022: 12361–12371. https://doi.org/10.1109/CVPR52688.2022.01205. Google Scholar
[23]	Jiang Z Y, Xia P F, Huang K, et al. Mixed frame-/event-driven fast pedestrian detection[C]// Proceedings of 2019 International Conference on Robotics and Automation, Montreal, 2019: 8332–8338. https://doi.org/10.1109/ICRA.2019.8793924. Google Scholar
[24]	Lagorce X, Orchard G, Galluppi F, et al. HOTS: a hierarchy of event-based time-surfaces for pattern recognition[J]. IEEE Trans Pattern Anal Mach Intell, 2017, 39(7): 1346−1359. DOI: 10.1109/TPAMI.2016.2574707 CrossRef Google Scholar
[25]	Zhu A, Yuan L Z, Chaney K, et al. Unsupervised event-based learning of optical flow, depth, and egomotion[C]//Proceedings of 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition, Long Beach, 2019: 989–997. https://doi.org/10.1109/CVPR.2019.00108. Google Scholar
[26]	Wang D S, Jia X, Zhang Y, et al. Dual memory aggregation network for event-based object detection with learnable representation[C]//Proceedings of the 37th AAAI Conference on Artificial Intelligence, Washington, 2023: 2492–2500. https://doi.org/10.1609/aaai.v37i2.25346. Google Scholar
[27]	Li J N, Li J, Zhu L, et al. Asynchronous spatio-temporal memory network for continuous event-based object detection[J]. IEEE Trans Image Process, 2022, 31: 2975−2987. DOI: 10.1109/TIP.2022.3162962 CrossRef Google Scholar
[28]	Peng Y S, Zhang Y Y, Xiong Z W, et al. GET: group event transformer for event-based vision[C]//Proceedings of 2023 IEEE/CVF International Conference on Computer Vision, Paris, 2023: 6015–6025. https://doi.org/10.1109/ICCV51070.2023.00555. Google Scholar
[29]	Chen N F Y. Pseudo-labels for supervised learning on dynamic vision sensor data, applied to object detection under ego-motion[C]//Proceedings of 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, Salt Lake City, 2018: 644–653. https://doi.org/10.1109/CVPRW.2018.00107. Google Scholar
[30]	Afshar S, Nicholson A P, Van S A, et al. Event-based object detection and tracking for space situational awareness[J]. IEEE Sensors Journal, 2020, 20(24): 15117−15132 DOI: 10.1109/JSEN.2020.3009687 CrossRef Google Scholar
[31]	Huang H M, Lin L F, Tong R F, et al. UNet 3+: a full-scale connected UNet for medical image segmentation[C]// Proceedings of ICASSP 2020–2020 IEEE International Conference on Acoustics, Speech and Signal Processing, Barcelona, 2020: 1055–1059. https://doi.org/10.1109/ICASSP40776.2020.9053405. Google Scholar
[32]	Wang C Y, Bochkovskiy A, Liao H Y M. YOLOv7: trainable bag-of-freebies sets new state-of-the-art for real-time object detectors[C]//Proceedings of 2023 IEEE/CVF Conference on Computer Vision and Pattern Recognition, Vancouver, 2023: 7464–7475. https://doi.org/10.1109/CVPR52729.2023.00721. Google Scholar
[33]	Liu W, Anguelov D, Erhan D, et al. SSD: single shot multibox detector[C]//Proceedings of the 14th European Conference on Computer Vision, Amsterdam, 2016: 21–37. https://doi.org/10.1007/978-3-319-46448-0_2. Google Scholar
[34]	Girshick R. Fast R-CNN[C]//Proceedings of 2015 IEEE International Conference on Computer Vision, Santiago, 2015: 1440–1448. https://doi.org/10.1109/ICCV.2015.169. Google Scholar
[35]	Charles R Q, Su H, Kaichun M. PointNet: deep learning on point sets for 3D classification and segmentation[C]// Proceedings of 2017 IEEE Conference on Computer Vision and Pattern Recognition, Honolulu, 2017: 77–85. https://doi.org/10.1109/CVPR.2017.16. Google Scholar
[36]	Charles R Q, Yi L, Su H, et al. PointNet++: deep hierarchical feature learning on point sets in a metric space[C]//Proceedings of the 31st International Conference on Neural Information Processing Systems, Long Beach, 2017: 5105–5114. https://dl.acm.org/doi/abs/10.5555/3295222.3295263. Google Scholar
[37]	Gehrig M, Scaramuzza D. Recurrent vision transformers for object detection with event cameras[C]//Proceedings of 2023 IEEE/CVF Conference on Computer Vision and Pattern Recognition, Vancouver, 2023: 13884–13893. https://doi.org/10.1109/CVPR52729.2023.01334. Google Scholar
[38]	Peng Y S, Li H B, Zhang Y Y, et al. Scene adaptive sparse transformer for event-based object detection[C]//Proceedings of 2024 IEEE/CVF Conference on Computer Vision and Pattern Recognition, Seattle, 2024: 16794–16804. https://doi.org/10.1109/CVPR52733.2024.01589. Google Scholar
[39]	Li B Y, Xiao C, Wang L G, et al. Dense nested attention network for infrared small target detection[J]. IEEE Trans Image Process, 2023, 32: 1745−1758. DOI: 10.1109/TIP.2022.3199107 CrossRef Google Scholar
[40]	Dai Y M, Li X, Zhou F, et al. One-stage cascade refinement networks for infrared small target detection[J]. IEEE Trans Geosci Remote Sens, 2023, 61: 5000917. DOI: 10.1109/TGRS.2023.3243062 CrossRef Google Scholar

View full references list

Cited By

Cited by

Periodical cited type(1)

陈俊. 基于YOLOv8的校园候车点人群计数系统设计与实现. 现代信息科技. 2025(05): 72-77 .

Other cited types(0)

Author Information

Author Information
- Li Miao, lm8866@nudt.edu.cn On this Site On Google Scholar
- Chen Nuo, chennuo97@nudt.edu.cn On this Site On Google Scholar
- Corresponding author: An Wei, anwei@nudt.edu.cn On this Site On Google Scholar
  anwei@nudt.edu.cn
- Li Boyang On this Site On Google Scholar
- Ling Qiang On this Site On Google Scholar
- Li Weixing On this Site On Google Scholar

Copyright

The copyright belongs to the Institute of Optics and Electronics, Chinese Academy of Sciences, but the article content can be freely downloaded from this website and used for free in academic and research work.

About this Article

About this Article
DOI: 10.12086/oee.2024.240208

Cite this Article

Li Miao, Chen Nuo, An Wei, Li Boyang, Ling Qiang, Li Weixing. Dual view fusion detection method for event camera detection of unmanned aerial vehicles. Opto-Electronic Engineering 51, 240208 (2024). DOI: 10.12086/oee.2024.240208

Li Miao, Chen Nuo, An Wei, Li Boyang, Ling Qiang, Li Weixing. Dual view fusion detection method for event camera detection of unmanned aerial vehicles. Opto-Electronic Engineering 51, 240208 (2024). DOI: 10.12086/oee.2024.240208

Download Citation
Article History
- Received Date September 01, 2024
- Revised Date October 11, 2024
- Accepted Date October 11, 2024
- Published Date November 24, 2024
Article Metrics

Article Views(2525) PDF Downloads(90)
Share:

Related Articles
- Hypergraph computed efficient transmission multi-scale feature small target detection algorithm
  
  Sun Yemei, Sang Xueting, Zhang Yan, Liu Guorui, Chen Shuaiyu
  
  Opto-Electronic Engineering, DOI: 10.12086/oee.2025.250061
- Multi-level refined UAV image target detection
  
  Xiao Zhenjiu, Lai Siyu, Qu Haicheng
  
  Opto-Electronic Engineering, DOI: 10.12086/oee.2025.240287
- Small target detection in sonar images with multilevel feature screening and task dynamic alignment
  
  Wang Yan, Wang Honghui, Liu Shudong, Zhang Yan, Hao Zeyu
  
  Opto-Electronic Engineering, DOI: 10.12086/oee.2024.240196
- A 3D reconstruction method based on multi-view fusion and hand-eye coordination for objects beyond the visual field
  
  Zhang Botao, Li Zhengqiang, Hua Chaohao, Xie Jialong, Lu Qiang
  
  Opto-Electronic Engineering, DOI: 10.12086/oee.2024.240180
- Feature coordination and fine-grained perception of small targets in remote sensing images
  
  Xiao Zhenjiu, Zhang Jiehao, Lin Bohan
  
  Opto-Electronic Engineering, DOI: 10.12086/oee.2024.240066
- A multi-target semantic segmentation method for millimetre wave SAR images based on a dual-branch multi-scale fusion network
  
  Ding Junhua, Yuan Minghui
  
  Opto-Electronic Engineering, DOI: 10.12086/oee.2023.230242
- Small object detection based on multi-scale feature fusion using remote sensing images
  
  Ma Liang, Gou Yutao, Lei Tao, Jin Lei, Song Yixuan
  
  Opto-Electronic Engineering, DOI: 10.12086/oee.2022.210363
- An object detection and tracking algorithm based on LiDAR and camera information fusion
  
  Chang Xin, Chen Xiaodong, Zhang Jiachen, Wang Yi, Cai Huaiyu
  
  Opto-Electronic Engineering, DOI: 10.12086/oee.2019.180420
- Multi-exposure image fusion based on tensor decomposition and convolution sparse representation
  
  Qi Yubin, Yu Mei, Jiang Hao, Shao Hua, Jiang Gangyi
  
  Opto-Electronic Engineering, DOI: 10.12086/oee.2019.180084
- Background modeling method based on multi-feature fusion
  
  Guo Zhicheng, Dang Jianwu, Wang Yangping, Jin Jing
  
  Opto-Electronic Engineering, DOI: 10.12086/oee.2018.180206

Article Contents
Figures/Tables
References

3. 基于事件点云分割的无人机小目标检测算法

View in article Downloads

Full-Size Img PowerPoint
View in article Downloads

Full-Size Img PowerPoint
View in article Downloads

Full-Size Img PowerPoint
View in article Downloads

Full-Size Img PowerPoint
View in article Downloads

Full-Size Img PowerPoint
View in article Downloads

Full-Size Img PowerPoint

方法	框架	${P}_{{\mathrm{d}}}$ /%	${P}_{{\mathrm{f}}}$ /%	AP	IoU	推理时间/ms
SSD	帧	88.3(+8.2)	13.3(−10.1)	62.7(+6.8)	—	18.7
FastRCNN	帧	88.7(+7.8)	13.5(−10.3)	63.2(+6.3)	—	19.5
YOLOV7	帧	89.8(+6.7)	10.3(−7.1)	65.3(+4.2)	—	16.5
RVT	帧	90.3(+6.2)	10.1(−6.9)	65.7(+3.8)	—	10.5
SAST	帧	90.1(+6.4)	10.5(−7.3)	65.5(+4.0)	—	13.2
DNA-Net	帧	90.5(+6.0)	9.91(−6.7)	65.9(+3.6)	—	20.3
OSCAR	帧	90.3(+6.2)	10.2(−7.0)	66.3(+3.2)	—	25.3
Pointnet	点云	91.3(+5.2)	8.35(−5.1)	66.1(+3.4)	66.7(+3.6)	19.8
Pointnet++	点云	93.5(+3.0)	7.33(−4.1)	67.3(+2.2)	67.5(+2.8)	25.3
Ours	点云	96.5	3.21	69.5	70.3	15.2

View in article Downloads

方法	${P}_{\rm{d}}$ /%	${P}_{\rm{f}}$ /%	AP	IoU/%	推理时间/ms
P	89.6 (+6.9)	9.86(−6.65)	66.3 (+3.2)	63.2 (+7.1)	7.2
V	90.6 (+5.9)	8.98 (−5.77)	65.7 (+3.8)	63.8 (+6.5)	10.2
V+	92.5 (+4.0)	5.33 (−2.12)	66.8 (+2.7)	64.7 (+5.6)	12.3
PV	94.3 (+2.2)	4.32 (−1.11)	67.2 (+2.3)	69.2 (+1.1)	19.2
PV+	96.5	3.21	69.5	70.3	20.3

View in article Downloads

方法	${P}_{\rm{d}}$ /%	${P}_{\rm{f}}$ /%	AP	IoU
Add	90.2 (+6.3)	8.63 (−5.42)	67.3 (+2.2)	66.8 (+3.7)
Multi	91.8 (+4.7)	7.68 (−4.47)	66.9 (+2.6)	64.2 (+6.1)
Concat	93.8 (+2.7)	6.38 (3.17)	67.6 (+1.9)	65.3 (+5.0)
Ours	96.5	3.21	69.5	70.3

View in article Downloads

[1]	Bouguettaya A, Zarzour H, Kechida A, et al. Vehicle detection from UAV imagery with deep learning: a review[J]. IEEE Trans Neural Netw Learn Syst, 2022, 33(11): 6047−6067. DOI: 10.1109/TNNLS.2021.3080276 CrossRef Google Scholar
[2]	陈海永, 刘登斌, 晏行伟. 基于IDOU-YOLO的红外图像无人机目标检测算法[J]. 应用光学, 2024, 45(4): 723−731. Chen H Y, Liu D B, Yan X W. Infrared image UAV target detection algorithm based on IDOU-YOLO[J]. Journal of Applied Optics, 2024, 45(4): 723−731. Google Scholar
[3]	韩江涛, 谭凯, 张卫国, 等. 协同随机森林方法和无人机LiDAR空谱数据的盐沼植被“精灵圈”识别[J]. 光电工程, 2024, 51(3): 230188. DOI: 10.12086/oee.2024.230188 Han J T, Tan K, Zhang W G, et al. Identification of salt marsh vegetation “fairy circles” using random forest method and spatial-spectral data of unmanned aerial vehicle LiDAR[J]. Opto-Electron Eng, 2024, 51(3): 230188. DOI: 10.12086/oee.2024.230188 CrossRef Google Scholar
[4]	Park S, Choi Y. Applications of unmanned aerial vehicles in mining from exploration to reclamation: a review[J]. Minerals, 2020, 10(8): 663. DOI: 10.3390/min10080663 CrossRef Google Scholar
[5]	Sziroczak D, Rohacs D, Rohacs J. Review of using small UAV based meteorological measurements for road weather management[J]. Prog Aerosp Sci, 2022, 134: 100859. DOI: 10.1016/j.paerosci.2022.100859 CrossRef Google Scholar
[6]	Wang Z Y, Gao Q, Xu J B, et al. A review of UAV power line inspection[C]//Proceedings of 2020 International Conference on Guidance, Navigation and Control, Tianjin, 2022: 3147–3159. https://doi.org/10.1007/978-981-15-8155-7_263. Google Scholar
[7]	Khan A, Gupta S, Gupta S K. Emerging UAV technology for disaster detection, mitigation, response, and preparedness[J]. J Field Robot, 2022, 39(6): 905−955. DOI: 10.1002/rob.22075 CrossRef Google Scholar
[8]	Li Y, Liu M, Jiang D D. Application of unmanned aerial vehicles in logistics: a literature review[J]. Sustainability, 2022, 14(21): 14473. DOI: 10.3390/su142114473 CrossRef Google Scholar
[9]	Mademlis I, Mygdalis V, Nikolaidis N, et al. High-level multiple-UAV cinematography tools for covering outdoor events[J]. IEEE Trans Broadcast, 2019, 65(3): 627−635. DOI: 10.1109/TBC.2019.2892585 CrossRef Google Scholar
[10]	奚玉鼎, 于涌, 丁媛媛, 等. 一种快速搜索空中低慢小目标的光电系统[J]. 光电工程, 2018, 45(4): 170654. DOI: 10.12086/oee.2018.170654 Xi Y D, Yu Y, Ding Y Y, et al. An optoelectronic system for fast search of low slow small target in the air[J]. Opto-Electron Eng, 2018, 45(4): 170654. DOI: 10.12086/oee.2018.170654 CrossRef Google Scholar
[11]	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算法[J]. 光电工程, 2024, 51(5): 240051. DOI: 10.12086/oee.2024.240051 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-Electron Eng, 2024, 51(5): 240051. DOI: 10.12086/oee.2024.240051 CrossRef Google Scholar
[12]	陈旭, 彭冬亮, 谷雨. 基于改进YOLOv5s的无人机图像实时目标检测[J]. 光电工程, 2022, 49(3): 210372 DOI: 10.12086/oee.2022.210372 Chen X, Peng D L, Gu Y. Real-time object detection for UAV images based on improved YOLOv5s[J]. Opto-Electron Eng, 2022, 49(3): 210372 DOI: 10.12086/oee.2022.210372 CrossRef Google Scholar
[13]	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算法[J]. 激光技术, 2024, 48(1): 114−120. DOI: 10.7510/jgjs.issn.1001-3806.2024.01.018 Zhang M C, Niu C H, Liu L S, et al. Infrared small target detection algorithm for UAV detection system[J]. Laser Technol, 2024, 48(1): 114−120. DOI: 10.7510/jgjs.issn.1001-3806.2024.01.018 CrossRef Google Scholar
[14]	Sedunov A, Haddad D, Salloum H, et al. Stevens drone detection acoustic system and experiments in acoustics UAV tracking[C]//Proceedings of 2019 IEEE International Symposium on Technologies for Homeland Security, Woburn, 2019: 1–7. https://doi.org/10.1109/HST47167.2019.9032916. Google Scholar
[15]	Chiper F L, Martian A, Vladeanu C, et al. Drone detection and defense systems: Survey and a software-defined radio-based solution[J]. Sensors, 2022, 22(4): 1453. DOI: 10.3390/s22041453 CrossRef Google Scholar
[16]	de Quevedo Á D, Urzaiz F I, Menoyo J G, et al. Drone detection and radar-cross-section measurements by RAD-DAR[J]. IET Radar Sonar Navig, 2019, 13(9): 1437−1447. DOI: 10.1049/iet-rsn.2018.5646 CrossRef Google Scholar
[17]	Gallego G, Delbrück T, Orchard G, et al. Event-based vision: a survey[J]. IEEE Trans Pattern Anal Mach Intell, 2020, 44(1): 154−180. DOI: 10.1109/TPAMI.2020.3008413 CrossRef Google Scholar
[18]	Shariff W, Dilmaghani M S, Kielty P, et al. Event cameras in automotive sensing: a review[J]. IEEE Access, 2024, 12: 51275−51306. DOI: 10.1109/ACCESS.2024.3386032 CrossRef Google Scholar
[19]	Paredes-Vallés F, Scheper K Y W, De Croon G C H E. Unsupervised learning of a hierarchical spiking neural network for optical flow estimation: from events to global motion perception[J]. IEEE Trans Pattern Anal Mach Intell, 2020, 42(8): 2051−2064. DOI: 10.1109/TPAMI.2019.2903179 CrossRef Google Scholar
[20]	Cordone L, Miramond B, Thierion P. Object detection with spiking neural networks on automotive event data[C]// Proceedings of 2022 International Joint Conference on Neural Networks, Padua, 2022: 1–8. https://doi.org/10.1109/IJCNN55064.2022.9892618. Google Scholar
[21]	Li Y J, Zhou H, Yang B B, et al. Graph-based asynchronous event processing for rapid object recognition[C]//Proceedings of 2021 IEEE/CVF International Conference on Computer Vision, Montreal, 2021: 914–923. https://doi.org/10.1109/ICCV48922.2021.00097. Google Scholar
[22]	Schaefer S, Gehrig D, Scaramuzza D. AEGNN: asynchronous event-based graph neural networks[C]//Proceedings of 2022 IEEE/CVF Conference on Computer Vision and Pattern Recognition, New Orleans, 2022: 12361–12371. https://doi.org/10.1109/CVPR52688.2022.01205. Google Scholar
[23]	Jiang Z Y, Xia P F, Huang K, et al. Mixed frame-/event-driven fast pedestrian detection[C]// Proceedings of 2019 International Conference on Robotics and Automation, Montreal, 2019: 8332–8338. https://doi.org/10.1109/ICRA.2019.8793924. Google Scholar
[24]	Lagorce X, Orchard G, Galluppi F, et al. HOTS: a hierarchy of event-based time-surfaces for pattern recognition[J]. IEEE Trans Pattern Anal Mach Intell, 2017, 39(7): 1346−1359. DOI: 10.1109/TPAMI.2016.2574707 CrossRef Google Scholar
[25]	Zhu A, Yuan L Z, Chaney K, et al. Unsupervised event-based learning of optical flow, depth, and egomotion[C]//Proceedings of 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition, Long Beach, 2019: 989–997. https://doi.org/10.1109/CVPR.2019.00108. Google Scholar
[26]	Wang D S, Jia X, Zhang Y, et al. Dual memory aggregation network for event-based object detection with learnable representation[C]//Proceedings of the 37th AAAI Conference on Artificial Intelligence, Washington, 2023: 2492–2500. https://doi.org/10.1609/aaai.v37i2.25346. Google Scholar
[27]	Li J N, Li J, Zhu L, et al. Asynchronous spatio-temporal memory network for continuous event-based object detection[J]. IEEE Trans Image Process, 2022, 31: 2975−2987. DOI: 10.1109/TIP.2022.3162962 CrossRef Google Scholar
[28]	Peng Y S, Zhang Y Y, Xiong Z W, et al. GET: group event transformer for event-based vision[C]//Proceedings of 2023 IEEE/CVF International Conference on Computer Vision, Paris, 2023: 6015–6025. https://doi.org/10.1109/ICCV51070.2023.00555. Google Scholar
[29]	Chen N F Y. Pseudo-labels for supervised learning on dynamic vision sensor data, applied to object detection under ego-motion[C]//Proceedings of 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops, Salt Lake City, 2018: 644–653. https://doi.org/10.1109/CVPRW.2018.00107. Google Scholar
[30]	Afshar S, Nicholson A P, Van S A, et al. Event-based object detection and tracking for space situational awareness[J]. IEEE Sensors Journal, 2020, 20(24): 15117−15132 DOI: 10.1109/JSEN.2020.3009687 CrossRef Google Scholar
[31]	Huang H M, Lin L F, Tong R F, et al. UNet 3+: a full-scale connected UNet for medical image segmentation[C]// Proceedings of ICASSP 2020–2020 IEEE International Conference on Acoustics, Speech and Signal Processing, Barcelona, 2020: 1055–1059. https://doi.org/10.1109/ICASSP40776.2020.9053405. Google Scholar
[32]	Wang C Y, Bochkovskiy A, Liao H Y M. YOLOv7: trainable bag-of-freebies sets new state-of-the-art for real-time object detectors[C]//Proceedings of 2023 IEEE/CVF Conference on Computer Vision and Pattern Recognition, Vancouver, 2023: 7464–7475. https://doi.org/10.1109/CVPR52729.2023.00721. Google Scholar
[33]	Liu W, Anguelov D, Erhan D, et al. SSD: single shot multibox detector[C]//Proceedings of the 14th European Conference on Computer Vision, Amsterdam, 2016: 21–37. https://doi.org/10.1007/978-3-319-46448-0_2. Google Scholar
[34]	Girshick R. Fast R-CNN[C]//Proceedings of 2015 IEEE International Conference on Computer Vision, Santiago, 2015: 1440–1448. https://doi.org/10.1109/ICCV.2015.169. Google Scholar
[35]	Charles R Q, Su H, Kaichun M. PointNet: deep learning on point sets for 3D classification and segmentation[C]// Proceedings of 2017 IEEE Conference on Computer Vision and Pattern Recognition, Honolulu, 2017: 77–85. https://doi.org/10.1109/CVPR.2017.16. Google Scholar
[36]	Charles R Q, Yi L, Su H, et al. PointNet++: deep hierarchical feature learning on point sets in a metric space[C]//Proceedings of the 31st International Conference on Neural Information Processing Systems, Long Beach, 2017: 5105–5114. https://dl.acm.org/doi/abs/10.5555/3295222.3295263. Google Scholar
[37]	Gehrig M, Scaramuzza D. Recurrent vision transformers for object detection with event cameras[C]//Proceedings of 2023 IEEE/CVF Conference on Computer Vision and Pattern Recognition, Vancouver, 2023: 13884–13893. https://doi.org/10.1109/CVPR52729.2023.01334. Google Scholar
[38]	Peng Y S, Li H B, Zhang Y Y, et al. Scene adaptive sparse transformer for event-based object detection[C]//Proceedings of 2024 IEEE/CVF Conference on Computer Vision and Pattern Recognition, Seattle, 2024: 16794–16804. https://doi.org/10.1109/CVPR52733.2024.01589. Google Scholar
[39]	Li B Y, Xiao C, Wang L G, et al. Dense nested attention network for infrared small target detection[J]. IEEE Trans Image Process, 2023, 32: 1745−1758. DOI: 10.1109/TIP.2022.3199107 CrossRef Google Scholar
[40]	Dai Y M, Li X, Zhou F, et al. One-stage cascade refinement networks for infrared small target detection[J]. IEEE Trans Geosci Remote Sens, 2023, 61: 5000917. DOI: 10.1109/TGRS.2023.3243062 CrossRef Google Scholar

Hypergraph computed efficient transmission multi-scale feature small target detection algorithm

Sun Yemei, Sang Xueting, Zhang Yan, Liu Guorui, Chen Shuaiyu

Opto-Electronic Engineering, DOI: 10.12086/oee.2025.250061
Multi-level refined UAV image target detection

Xiao Zhenjiu, Lai Siyu, Qu Haicheng

Opto-Electronic Engineering, DOI: 10.12086/oee.2025.240287
Small target detection in sonar images with multilevel feature screening and task dynamic alignment

Wang Yan, Wang Honghui, Liu Shudong, Zhang Yan, Hao Zeyu

Opto-Electronic Engineering, DOI: 10.12086/oee.2024.240196
A 3D reconstruction method based on multi-view fusion and hand-eye coordination for objects beyond the visual field

Zhang Botao, Li Zhengqiang, Hua Chaohao, Xie Jialong, Lu Qiang

Opto-Electronic Engineering, DOI: 10.12086/oee.2024.240180
Feature coordination and fine-grained perception of small targets in remote sensing images

Xiao Zhenjiu, Zhang Jiehao, Lin Bohan

Opto-Electronic Engineering, DOI: 10.12086/oee.2024.240066
A multi-target semantic segmentation method for millimetre wave SAR images based on a dual-branch multi-scale fusion network

Ding Junhua, Yuan Minghui

Opto-Electronic Engineering, DOI: 10.12086/oee.2023.230242
Small object detection based on multi-scale feature fusion using remote sensing images

Ma Liang, Gou Yutao, Lei Tao, Jin Lei, Song Yixuan

Opto-Electronic Engineering, DOI: 10.12086/oee.2022.210363
An object detection and tracking algorithm based on LiDAR and camera information fusion

Chang Xin, Chen Xiaodong, Zhang Jiachen, Wang Yi, Cai Huaiyu

Opto-Electronic Engineering, DOI: 10.12086/oee.2019.180420
Multi-exposure image fusion based on tensor decomposition and convolution sparse representation

Qi Yubin, Yu Mei, Jiang Hao, Shao Hua, Jiang Gangyi

Opto-Electronic Engineering, DOI: 10.12086/oee.2019.180084
Background modeling method based on multi-feature fusion

Guo Zhicheng, Dang Jianwu, Wang Yangping, Jin Jing

Opto-Electronic Engineering, DOI: 10.12086/oee.2018.180206

Dual view fusion detection method for event camera detection of unmanned aerial vehicles

Abstract

Keywords

1. 引 言

2. 事件相机介绍

2.1 事件相机与传统相机的区别

2.2 事件相机无人机检测数据集

3. 基于事件点云分割的无人机小目标检测算法

3.1 点特征提取分支

3.2 体素特征提取分支

3.3 双分支特征融合模块

4. 结果与分析

4.1 评价指标

4.2 实验训练设置

4.3 实验结果与分析

4.3.1 对比实验

4.3.2 消融实验

5. 结 论

利益冲突

References

Cited by

Periodical cited type(1)

Other cited types(0)

Author Information

Li Miao, lm8866@nudt.edu.cn On this SiteOn Google Scholar

Chen Nuo, chennuo97@nudt.edu.cn On this SiteOn Google Scholar

Corresponding author: An Wei, anwei@nudt.edu.cn On this SiteOn Google Scholar

Li Boyang On this SiteOn Google Scholar

Ling Qiang On this SiteOn Google Scholar

Li Weixing On this SiteOn Google Scholar

Copyright

About this Article

Cite this Article

Article History

Article Metrics

Related Articles

Links

Related Articles

Catalog

Li Weixing

Manuscript Submission

More Content

Legal & Privacy

Export File

Citation

Format

Content

WeChat Qrcode

1. 引　言

5. 结　论

Li Miao, lm8866@nudt.edu.cn On this Site On Google Scholar

Chen Nuo, chennuo97@nudt.edu.cn On this Site On Google Scholar

Corresponding author: An Wei, anwei@nudt.edu.cn On this Site On Google Scholar

Li Boyang On this Site On Google Scholar

Ling Qiang On this Site On Google Scholar

Li Weixing On this Site On Google Scholar