本文为意大利特伦托大学（作者：Thomas Moranduzzo）的博士论文，共108页。
无人机（UAV），通常被称为无人驾驶飞机，是在遥感领域获得广泛应用的空中平台。无人机起源于军事技术，但在最近几年，已经逐渐用于民用任务的设备平台。这种采集系统的主要优点在于其使用简单。确实，无需太多的费用就可以使用无人机。由于无人机可以飞得非常接近被调查的物体，因此它们能够获取极高分辨率(EHR)图像，可以得到非常详细的细节特征。无人机图像所包含的海量信息为开发新的应用开辟了道路，但同时也迫使我们在方法学层面上面临新的挑战性问题。本文对于使无人机图像完全理解并易于处理和分析作出了微不足道的贡献，但希望是有用的。
本文提出的方法包括：
1）两种在城市场景中自动检测并对汽车数量进行计数的方法；
2）一个完整的处理链路，用于监控交通和估计移动车辆的速度；
3）一种利用非线性滤波器结合不同梯度的图像梯度特征和高斯过程（GP）建模来检测目标的方法；
4）一种利用多种表示和匹配策略“粗略”描述超高分辨率图像的新策略。
本文对真实的无人机图像进行了实验分析与讨论，分析结果证实了本文所提出的方法，并给出了未来提升算法性能的建议。尽管图像处理的过程很复杂，但是无人机图像的应用潜力是相当广阔的。
Unmanned Aerial Vehicles (UAVs), commonlyknown as drones, are aerial platforms that are gaining large popularity in theremote sensing field. UAVs derive from military technology, but in the last fewyears they are establishing as reference platforms also for civilian tasks. Themain advantage of these acquisition systems lies in their simplicity of use.Indeed, a UAV can be used when and where it is needed without excessive costs.Since UAVs can fly very close to the objects under investigation they allow theacquisition of extremely high resolution (EHR) images in which the items aredescribed with a very high level of details. The huge quantity of informationcontained in UAV images opens the way to develop novel applications but at thesame time force us to face new challenging problems at methodological level.This thesis represents a modest but hopefully useful contribution towardsmaking UAV images completely understood and easily processed and analyzed. Inparticular, the proposed methodological contributions include: i) two methodsdevoted to the automatic detection and counting of cars present in urbanscenarios; ii) a complete processing chain which monitors the traffic andestimate the speeds of moving vehicles; iii) a methodology which detectsclasses of objects by exploiting a nonlinear filter which combines imagegradient features at different orders and Gaussian process (GP) modeling; iv) anovel strategy to “coarsely” describe extremely high resolution images usingvarious representation and matching strategies. Experimental results conductedon real UAV images are presented and discussed. They show the validity of theproposed methods and suggest future possible improvements. Furthermore, theyconfirm that despite the complexity of the considered images, the potential ofUAV images is very wide.
1 引言及本文组织结构
2 无人机图像中的交通流量监控策略
3 用于无人机图像的基于滤波器的目标检测器
4 用于无人机图像的基于多级平铺的分析方法
5 致谢
6 结论
7 参考文献
8 发表文献列表
下载英文原文地址：
http://page5.dfpan.com/fs/1lecaj223271d2e9169/
更多精彩文章请关注微信号：