航空影像数据集

1. UC Merced Land-Use：用于土地利用研究的可见光遥感图像数据集，图片取自USGS National Map Urban Area Imagery。

YANG Y,NEWSAM S.Bag-of-visual-words and spatial extensions for land-use classification[C]∥Proceedings of the 18th SIGSPATIAL International Conference on Ad- vances in Geographic Information Systems.New York: Association for Computing Machinery,2010:270-2

2. NWOU VHR-10：该数据集有10个类别的对象，从Google Earth和Vaihingen数据集裁剪而来，由专家进行了类别标注。

CHENG G,ZHOU P,HAN J,et al.Learning rotation-invariant convolutional neural networks for object detec- tion in VHR optical remote sensing images[J].IEEE Transactions on Geoscience and Remote Sensing,2016, 54(12):7405-7415.

3. VE-DAI：该数据集用于多种类车辆检测任务，数据集的航空影像取自美国犹他州AGRC。

RAZAKARIVONY S,JURIE F.Vehicle detection in aerial imagery[J].Journal of Visual Communication and Im- age Representation,2016,34(C):187-203.

4. UCAS-AOD：用于航空图像下车辆和飞机的目标检测，图像采集于Google Earth，车辆数据集共210张图片，飞机数据集共600张图片。

ZHU H,CHEN X,DAI W,et al.Orientation robust object detection in aerial images using deep convolutional neural network[C]∥2015IEEE International Conference on Image Processing.Piscataway,NJ:IEEE Press, 2015:3735-3739.

5. DOTA：DOTA数据集是一个用于航空影像中用于目标检测的大型数据集，图像采集来自于不同的传感器和平台，包含了不同比例、方向和形状的目标对象。

 XIA G S,BAI X,DING J,et al.DOTA:A large-scale dataset for object detection in aerial images[C]∥CVPR 2018:Proceedings of the 2018IEEE Conference on Com- puter Vision and Pattern Recognition.Washington, D.C.:IEEE Computer Society,2018:3974-3983.

5. Stanford Drone Dataset：由无人机拍摄而制作的图像/视频数据集，影像采集于校园中目标较为拥挤的场景。

 ROBICQUET A,SADEGHIAN A,ALAHI A,et al.Learning social etiquette:Human trajectory understanding in crowded scenes[C]∥ECCV 2016:2016European Con- ference on Computer Vision.Berlin:Springer,2016:549- 565.

6. Okutama Action Dataset：用无人机拍摄，用于检测人体动作的视频数据集。

BAREKATAIN M,MARTI M,SHIH H,et al.Okuta-ma-Action:an aerial view video dataset for concurrent hu- man action detection[C]∥CVPR 2017:Proceedings of the 2017IEEE Conference on Computer Vision and Pattern Recognition.Washington,D.C.:IEEE Computer Socie- ty,2017:2153-2160.

7.CARPK：停车场数据集，包含90000辆车，用于无人机对车辆的检测和计数任务。

HSIEH M,LIN Y,HSU W H,et al.Drone-based object counting by spatially regularized regional proposal network [C]∥ICCV 2017:Proceedings of the 2017IEEE Interna-tional Conference on Computer Vision.Piscataway: IEEE,2017:4165-4173.

8.VisDrone：包含不同天气和光照条件下的288个视频和10000余张图像，用于无人机图像目标检测、视频目标检测、单目标跟踪和多目标跟踪。

ZHU P,WEN L,BIAN X,et al.Vision meets drones:a challenge[DB/OL].ArXiv Preprint:1804.07437,2018. 

9. DroneVehicle：面向车辆检测和车辆计数任务的数据集，包含了RGB图像和红外图像，采集涵盖了昼夜时段以及目标的遮挡和尺度变化。

 ZHU P,SUN Y,WEN L,et al.Drone based RGBT ve-

hicle detection and counting:a challenge[DB/OL].ArXiv Preprint:2003.02437,2020.