一种改进的降落影像序列特征匹配方法

发布时间：2018-06-27 09:33

本文选题：SIFT + 嫦娥三号着陆器　；参考：《测绘科学》2017年05期

【摘要】：针对传统SIFT匹配方法用于嫦娥三号降落影像匹配效率低的问题,提出了一种改进的降落影像序列特征匹配方法。重构了着陆器垂直降落阶段拍摄过程的几何模型,并得到了不同降落影像中的同名匹配点之间的几何约束关系,然后通过SIFT方法进行影像中极值点的提取和特征描述,最终将几何模型约束用于极值点的匹配点搜索过程中,优化了匹配过程。采用嫦娥三号着陆器真实降落影像进行了试验,结果表明,改进方法能够取得较为稳定的匹配结果,且平均可以减少约为18%的匹配耗时,研究结果对后续深空探测任务的开展具有重要的参考价值。
[Abstract]:Aiming at the low efficiency of traditional sift matching method for Chang 'e-3 landing image, an improved feature matching method for landing image sequence is proposed. The geometric model of the shooting process of the lander in the vertical landing stage is reconstructed, and the geometric constraint relations between the matching points of the same name in different landing images are obtained. Then, the extraction and feature description of the extreme points in the images are carried out by sift method. Finally, geometric model constraints are applied to the matching point search process of extremum points, and the matching process is optimized. The real landing image of Chang 'e-3 lander is used to test. The results show that the improved method can obtain stable matching results and reduce the matching time by about 18% on average. The research results have important reference value for the following deep space exploration mission.
【作者单位】：辽宁工程技术大学测绘与地理科学学院;中国科学院遥感与数字地球研究所;
【基金】：国家重点研发计划(2016-YFC0803102) 国家自然科学基金项目(41401535)
【分类号】：P184.5

【相似文献】