利用光照模型的月球勘测轨道器窄角相机南极地区影像快速预筛选方法

A Quick Prefiltering Method for LRO NAC Images Based on Illumination Simulation with Application to Lunar South Pole

针对月球南极地区光照条件复杂、轨道器影像中多阴影、覆盖感兴趣区域(range of interest,RoI)的有效影像筛选困难的问题,提出了一种利用光照模型的月球勘测轨道器(lunar reconnaissance orbiter,LRO)窄角相机(narrow-angle camera,NAC)南极地区影像快速预筛选方法。使用月球南极地区的光照模型计算出LRO NAC影像拍摄时刻下RoI内的入射光照条件,进而通过约束RoI内被影像覆盖并且有光照的总面积,实现无需下载影像数据的LRO NAC影像快速预筛选。对于3个16 km×16 km的RoI,在该文实验环境下,平均处理一幅影像仅需0.005 s,约66.2%的LRO NAC影像被筛除,相比不考虑光照条件的传统预筛选方法多筛除19.6%的影像,证明了该方法的有效性。使用101幅LRO NAC影像对筛选准确率进行检验,仅出现两次错检,没有漏检,检验和筛选的准确率在98%左右,证明了该方法的可靠性。

Aiming at the problems of complex illumination conditions in the lunar polar regions,multiple shadows in the orbiter images,and difficulties in filtering effective images covering the range of interest(RoI),a quick prefiltering method for lunar reconnaissance orbiter(LRO)narrow-angle camera(NAC)images based on illumination simulation is proposed.The method uses the lunar illumination model to calculate the incident illumination conditions in the RoIs at the moment when the LRO NAC images are captured,and then realizes the quick prefiltering of LRO NAC images without downloading any image data by constraining the total area in the RoIs that is covered by the images and illuminated.For three 16 km×16 km RoI,the average processing of an image in our experimental environment takes only 0.005 s.About 66.2%of the LRO NAC images are filtered out,and about 19.6%more images are filtered out compared with the conventional prefiltering method that does not take into account the illumination conditions,which proves the effectiveness of the proposed method.Using 101 LRO NAC images to test the filtering accuracy,there are only two misdetections and no misses,and the accuracy of the filtering is around 98%,demonstrating the reliability of the proposed method.