摘要
为了提高SINS/GPS组合导航中景象匹配辅助的精确性和可靠性,必须在景象匹配区的适配性因素不同时选择到合适的匹配算法。系统对SURF和SIFT2种匹配算法的适用条件进行研究,比较各适配性因素对匹配区选择以及匹配算法性能的影响。实验结果表明,2种算法都更容易在地物稳定的区域内找到稳定特征点,匹配成功率取决于基准图抗噪性和实时图旋转角度等适配性因素。
To improve the accuracy and the reli-ability of SINS/GPS integrated navigation aided by scene matching,comparison studies were conducted to figure out the applicability of the two state of art matching algorithms SURF and SIFT and the influence of the suitability factors.The results of simulation experiments indicate that the image re-gion with high gradient is suitable to be as marc-hing region.
出处
《机械与电子》
2014年第3期10-13,共4页
Machinery & Electronics
基金
航空科学基金资助项目(20130153002)
关键词
组合导航
景象匹配
适配性
匹配算法
integrated navigation
scene matc-hing
suitability
matching algorithms