阅读论文DOTA和RoITransformer_DOTA

通过阅读这两篇论文，可以大致了解遥感图像目标检测领域所存在的问题，以及目前所采用的解决方法有哪些。
由于遥感图像天生的特点：

1. the huge variation in the scale, orientation and shape of the object instances：目标尺寸、方向、形状的变化大
2. 小而稠密的物体

4

DOTA
介绍
由于重力原因，传统目标检测数据集中的目标通常方向是竖直的，但是遥感图像中目标方向是任意的。

Recently, driven by the successes of deep learning-based algorithms for object detection, Earth Vision researchers have pursued approaches based on fine-tuning networks pre-trained on large-scale image datasets (e.g., ImageNet [6] and MSCOCO [13]) for detection in the aerial domain, see e.g. [2,3,18,29]. 运用深度学习的方法做遥感领域目标检测
[2] G. Cheng, P. Zhou, and J. Han. Learning rotation-invariant convolutional neural networks for object detection in VHR optical remote sensing images. IEEE Trans. Geosci. Remote Sens., 54(12):7405–7415, 2016.
[3] G. Cheng, P. Zhou, and J. Han. Rifd-cnn: Rotation-invariant and fisher discriminative convolutional neural networks for object detection. In CVPR, pages 2884–2893, 2016.
[18] Y. Long, Y. Gong, Z. Xiao, and Q. Liu. Accurate object localization in remote sensing images based on convolutional neural networks. IEEE Trans. Geosci. Remote Sens., 55(5):2486–2498, 2017.
[29] M. Vakalopoulou, K. Karantzalos, N. Komodakis, and N. Paragios. Building detection in very high resolution multispectral data with deep learning features. In IGARSS, pages 1873–1876, 2015.

不同于传统的目标检测，遥感图像目标检测主要有以下特点：
1.遥感图像中目标尺度变化大，不仅仅是由于传感器的空间分辨率，而且由于相同目标类别尺寸的变化性。
2.遥感图像中许多小目标聚集在一起，而且，遥感图像中目标的个数也不平衡，比如，一些小尺寸（1k1k）的图像中包含1900个目标，然而一些大尺寸（4k4k）图像中可能只包含很少目标。
3.遥感图像中目标是任意方向的。

However, existing annotated datasets for object detection in aerial images, such as UCAS-AOD [39] and NWPU VHR-10 [2], tend to use images in ideal conditions (clear backgrounds and without densely distributed instances), which cannot adequately reflect the problem complexity.