论文部分内容阅读
利用无人机进行低空摄影获取地面高分辨率影像,具有成本低、方便、快捷等优点,但由于无人机飞行姿态不稳定,会导致影像自动匹配效率及准确程度降低.SIFT算子具有良好的尺度、旋转、光照等不变特性,但传统SIFT算法复杂度较高,处理影像时间较长,使数据处理工作效率降低.介绍了一种简化的SIFT算法,并与原SIFT算法做出了对比.说明了影像重叠度分析的主要步骤,将简化的SIFT算法RANSAC算法相结合应用到序列影像重叠度分析中,通过实验证明了可行性.
It is low cost, convenient and fast to take advantage of low altitude aerial photography by drone.However, due to the unstable flight attitude of the UAV, the automatic matching efficiency and accuracy of the image will be reduced.SIFT operator has good performance However, the traditional SIFT algorithm is more complex and takes longer time to process images, which reduces the data processing efficiency.A simplified SIFT algorithm is introduced and compared with the original SIFT algorithm The main steps of image overlap analysis are illustrated, the simplified SIFT algorithm RANSAC algorithm is applied to the overlap analysis of sequence images, and the feasibility is proved through experiments.