【摘要】 该文研究了一种无人机航拍高清影像快速拼接方法,主要包括路线规划、视频特定帧抽取、帧图像拼接三步。其中,为得到较精确的对地分辨率,对实验场景进行路线规划;为解决无人机影像冗余帧信息量大的问题,提出了一种基于目标场景和无人机飞行状态参数的抽取算法;为解决配准阶段的耗时问题,用SURF算子仅在重叠区域内进行图像配准,减少配准用时。实验表明,该方法基本达到实时性要求,且拼接图像的视觉效果较好。更多还原

【Abstract】 This paper developed a mosaic method for unmanned aerial vehicle( UAV) photograph video with high definition( HD),it mainly includes three steps: route planning,a specific frame extraction of video and image mosaic. First,in order to get more accurate to the resolution,the scene of route should be planned; Second,in order to solve the problem of large amount of redundant frames of UAV image,a new algorithm based on the target scene and the flight state parameters of UAV; Third,the SURF operator is used to make the image registration in the overlapping area,then the time consuming for registration is reduced. Finally,the simulation results show that the method basically reached UAV aerial image mosaic real-time requirements,and the visual effect of mosaic image is fine.更多还原