【作者】 王利勇；

【导师】 王慧；

【作者基本信息】 解放军信息工程大学 ， 摄影测量与遥感， 2011， 硕士

【摘要】 当前,无人机低空遥感在很多领域有着越来越广泛的应用。本文利用固定翼无人机平台获取的遥感数据,研究低空遥感数字影像自动拼接与快速定位处理的技术方法。重点探讨了基于SFM(Structure From Motion)的数字影像自动拼接和低空遥感数字影像与移动道路测量系统相结合的快速空中三角测量。完成的主要工作和取得的成果如下:1、阐述了论文研究的背景和意义以及无人机低空遥感数据获取与处理的国内外研究现状。2、介绍了四种常用的兴趣点算子和两种局部不变性特征提取方法,针对无人机低空遥感数字影像的特点和实际应用需求,依据探测速度、适应性和提取效能三个评价准则,通过实验检验并评价了常用兴趣点算子和局部不变性特征在低空遥感数字影像中的应用性能。3、研究了低空遥感数字影像纹理增强预处理和小波去噪的方法及对影像特征点提取与匹配的影响,提出了一种基于SFM(Structure From Motion)的低空遥感数字影像自动生成“正射影像”与拼接的方法。实验证明该方法处理流程自动化,影像拼接效果好。4、研究了将无人机低空遥感系统和移动道路测量系统相结合的快速定位方法和作业流程。该方法缩短了摄影测量作业周期,提高了作业效率。5、简要介绍了无人机低空遥感影像数据的处理流程,研制了一套专门针对无人机低空遥感影像的数据处理系统。更多还原

【Abstract】 Currently, low altitude remote sensing with unmanned aerial vehicles (UAV) plays a more and more important role in many realms. In this article, we researched in the auto mosaic and fast positioning techniques of low altitude remote sensing digital images by using the remote sensing data from the fixed-wing unmanned aerial vehicle platform, focused on the discussion of full-automatic digital images mosaic on the basis of SFM (Structure From Motion) and the fast aerial triangulation which combines low altitude remote sensing digital images with mobile road survey. Our mainly work and achievement are as follows:1. Described the background and significance of this article and also the research status between domestic and aboard in the obtaining and processing of unmanned aerial vehicle remote sensing data.2. Introduced four types of common used interest points operators and two types of local invariant feature extraction methods .Based on three evaluate criteria in the aim of practical need of application and the characteristic of the low altitude unmanned aerial vehicle remote sensing: detection speed、adaptability、extraction efficiency, testified the application performance in the low altitude remote sensing images of these operators and local invariant features.3. Researched on the methods of texture enhancement pre-processing and wavelet denoising for low altitude remote sensing digital images and the influence of it for feature point extraction and matching. Proposed a full-automatic method to generate orthoimage and mosaic from low altitude remote sensing digital images on the basis of SFM. Experiment proved the method has automatic process and the effect of mosaic is well.4. Researched on the fast positioning method and working process by combining unmanned aerial vehicle low altitude remote sensing system with mobile road surveying system which shortened the period of photogrammetric work and improved efficiency.5. A brief introduction of the process of using low-altitude UAV remote sensing image data. Developed a data processing systems specialized in low-altitude UAV remote sensing.更多还原