嫦娥一号IIM高光谱数据和月球轨道器LOLA DEM数据的配准与月表地形校正及评价
作者简介:马明亮(1988-),男,河南驻马店市人,硕士生,研究方向为行星遥感、高光谱遥感。E-mail:maminglianghdsf@163.com
收稿日期: 2014-07-09
要求修回日期: 2014-08-12
网络出版日期: 2015-01-05
基金资助
国家自然科学基金项目(41201358)
上海市科委重点支撑项目(13231203804)
地理信息科学教育部重点实验室主任基金
The Research and Assessment of Topographic Registration and Correction of Chang’E-1 IIM Data Based on LRO LOLA DEM Data
Received date: 2014-07-09
Request revised date: 2014-08-12
Online published: 2015-01-05
Copyright
在对月探测中,月表物质的可见光、近红外反射数据,是进行月表化学元素及矿物反演与制图的重要信息源。受月球表面地形起伏的影响,嫦娥一号干涉成像光谱仪(IIM)高光谱数据在开展月表参数定量反演前,须进行月表地形校正,还原月表真实的反射率信息。IIM高光谱数据和常用的美国LOLA月球DEM数据之间存在月球经纬度空间配准不精确的问题,对月表地形校正的精度产生了影响。以月表陨石坑为例,在两幅月球遥感影像上选取一定数量的同名点,使用多项式校正方法进行像元级配准,与直接使用经纬度开展空间配准作了对比分析,发现IIM高光谱数据与LOLA DEM数据之间在经度方向存在平均约3.5个像元的位置偏差,纬度方向存在约1.95个像元的偏差。在此基础上,尝试将地球地形校正中使用的C校正方法运用到月球,探究在微弱大气散射环境下,月球陨石坑地貌的月表地形校正可行性。研究发现,经过像元级空间配准的数据在月表地形校正的效果上,比直接使用月球经纬度进行匹配的校正效果有明显提高。经过匹配和C校正,月表反射率与太阳入射角的余弦值之间的线性相关方程的斜率降低了89.4%,很好地消除月表陨石坑阴影地区和阳坡高亮区域的月表地形效应,恢复月表阴影区域的光谱信息。验证结果表明,校正后的月表局部遥感影像更接近于月表真实反射率,为后续利用IIM数据开展月表理化要素定量反演研究提供了可靠的科学数据保障。
关键词: 嫦娥一号; 干涉成像光谱仪(IIM); 月球轨道器激光测高仪(LOLA); 地理配准; C校正; 地形校正
马明亮 , 王超 , 施润和 , 高炜 . 嫦娥一号IIM高光谱数据和月球轨道器LOLA DEM数据的配准与月表地形校正及评价[J]. 地球信息科学学报, 2015 , 17(1) : 118 -125 . DOI: 10.3724/SP.J.1047.2015.00118
The characteristics of the visible and near-infrared reflectance from lunar surface are key information for the inversion and mapping of chemical elements in lunar exploration. Due to the dimply topography, hyperspectral dataobtained from IIM should be introduced with topographic correction before the application of practical inversion to get the actual reflectivity information. There ported longitudes and latitudes of the IIM data generally did not match with the conventional data from the LOLA DEM. Consequently, the accuracy of the topographic correction might be affected and decreased. In this paper, a case study was done related to the lunar crater. More specifically, a certain number of ground control points in a crater with same characteristic were selected for both images to match the IIM data and LOLA DEM data. The results from polynomial correction method were different from the correction results calculated directly from lunar longitude and latitude. There was approximately 3.5 pixels’ offset in the direction of longitude and about 1.95 pixels’ offset in the direction of latitude between IIM data and LOLA DEM data. The conventional topographical correction used on earth (i.e. C correction method) was introduced for a possible topographical correction on the moon, where there is almost no atmosphere scattering. The results showed that the performance of the matched topographical correction data was much better than those from unmatched data. After the application of registration and C correction, the slope of the linear correlation equation between the incident angle cosine value and the reflectance was reduced by 89.4 percent, the terrain effects of the shadow area and the highlight area on the lunar surface was eliminated. We concluded that the actual reflectivity information from lunar surface could be closely assessed after applying registration and topographic correction to IIM data, which could provide a sound basis for the inversion of chemical elements and minerals based on IIM data.
Key words: Chang’E-1; IIM; LOLA; registration; C Correction; topographic correction
Fig. 1 Thesketch map of the study area.图1 研究区域示意图 |
Fig. 2 The flow chart of data processing图2 数据处理流程图 |
Fig. 3 The overlay chart of the DEM and IIM reflectance image before (a) and after (b)registration.图3 配准前后IIM影像与DEM影像叠置图比较 |
Tab. 1 Comparison of the sample numbers and line numbers of the ground control points in the IIM image before and after registration表1 配准前、后IIM影像特征点行列号比较 |
特征点号 | 配准前行号 | 配准后行号 | 行偏移量 | 配准前列号 | 配准后列号 | 列偏移量 |
---|---|---|---|---|---|---|
1 | 115 | 111 | 4 | 26 | 27 | 1 |
2 | 124 | 120 | 4 | 332 | 334 | 2 |
2 | 69 | 65 | 4 | 163 | 165 | 2 |
4 | 81 | 77 | 4 | 142 | 144 | 2 |
5 | 98 | 94 | 4 | 123 | 125 | 2 |
6 | 12 | 9 | 3 | 151 | 153 | 2 |
7 | 162 | 158 | 4 | 245 | 247 | 2 |
8 | 28 | 25 | 3 | 267 | 269 | 2 |
9 | 50 | 46 | 4 | 77 | 79 | 2 |
10 | 35 | 32 | 3 | 298 | 300 | 2 |
11 | 63 | 60 | 3 | 352 | 354 | 2 |
12 | 148 | 145 | 3 | 324 | 326 | 2 |
13 | 47 | 44 | 3 | 103 | 105 | 2 |
14 | 122 | 118 | 4 | 57 | 59 | 2 |
15 | 101 | 98 | 3 | 296 | 298 | 2 |
16 | 160 | 157 | 3 | 311 | 313 | 2 |
17 | 149 | 145 | 4 | 324 | 326 | 2 |
18 | 90 | 87 | 3 | 353 | 355 | 2 |
19 | 95 | 92 | 3 | 88 | 90 | 2 |
20 | 150 | 146 | 4 | 207 | 209 | 2 |
平均偏移量 | 3.5 | 1.95 |
Fig. 4 The reflectance image of band 20 of IIM: the original image (a), spatial unmatched image (b), and spatial matched image (c)图4 校正前后IIM第20波段的影像对比 |
Fig. 5 (a) is the original average reflectance of IIM 26 bands in the shadow area (the top red polygon in Fig. 4(a))and the highlight area (the bottom red polygon in Fig. 4 (a)). (b) is the correction average reflectance of IIM 26 bands in the shadow area and the highlight area.图5 阴影区域(图4(a)的右上红色方框)和阳坡高亮区域(图4(a)的左下红色方框)校正前、后26个波段波长的平均反射率 |
Fig. 6 The scatter plots of the reflectance from the original image (a)and topographic corrected image (b) of the band 20 IIM image.图6 IIM第20波段校正前、校正后影像的反射率散点图 |
Fig. 7 The histograms of the reflectance from the original image (a) and corrected image (b)of the band 20 IIM image图7 校正前、校正后IIM第20波段反射率值分布直方图 |
Tab.2 Comparison between the statistical parameters of the original and topographic corrected bands of the IIM image表2 月表地形校正前后各个波段IIM影像两种统计参数比较 |
IIM波长(nm) | 校正前平均值 | 校正后平均值 | 校正前标准差 | 校正后标准差 |
---|---|---|---|---|
522 | 0.1733 | 0.1755 | 0.0464 | 0.0375 |
531 | 0.1738 | 0.1760 | 0.0468 | 0.0376 |
541 | 0.1796 | 0.1819 | 0.0477 | 0.0382 |
550 | 0.1815 | 0.1838 | 0.0473 | 0.0378 |
561 | 0.1829 | 0.1851 | 0.0484 | 0.0386 |
571 | 0.1864 | 0.1888 | 0.0488 | 0.0388 |
582 | 0.1885 | 0.1909 | 0.0495 | 0.0393 |
594 | 0.1924 | 0.1949 | 0.0504 | 0.0400 |
605 | 0.1955 | 0.1980 | 0.0507 | 0.0400 |
618 | 0.1993 | 0.2018 | 0.0515 | 0.0406 |
631 | 0.2017 | 0.2042 | 0.0520 | 0.0409 |
644 | 0.2049 | 0.2075 | 0.0526 | 0.0413 |
658 | 0.2048 | 0.2074 | 0.0516 | 0.0403 |
673 | 0.2095 | 0.2122 | 0.0531 | 0.0415 |
688 | 0.2095 | 0.2152 | 0.0531 | 0.0418 |
704 | 0.2167 | 0.2195 | 0.0549 | 0.0429 |
721 | 0.2206 | 0.2234 | 0.0557 | 0.0432 |
738 | 0.2238 | 0.2267 | 0.0560 | 0.0436 |
757 | 0.2266 | 0.2296 | 0.0563 | 0.0435 |
776 | 0.2289 | 0.2320 | 0.0564 | 0.0435 |
797 | 0.2303 | 0.2334 | 0.0559 | 0.0431 |
818 | 0.2312 | 0.2344 | 0.0554 | 0.0426 |
841 | 0.2305 | 0.2337 | 0.0542 | 0.0413 |
865 | 0.2306 | 0.2339 | 0.0536 | 0.0407 |
891 | 0.2321 | 0.2355 | 0.0532 | 0.0403 |
918 | 0.2342 | 0.2376 | 0.0535 | 0.0406 |
The authors have declared that no competing interests exist.
[1] |
|
[2] |
|
[3] |
|
[4] |
|
[5] |
|
[6] |
|
[7] |
|
[8] |
|
[9] |
|
[10] |
|
[11] |
|
[12] |
|
[13] |
|
[14] |
|
[15] |
|
[16] |
|
[17] |
|
[18] |
|
[19] |
|
[20] |
|
[21] |
|
[22] |
|
[23] |
|
[24] |
|
[25] |
|
[26] |
|
[27] |
|
[28] |
|
[29] |
|
[30] |
|
[31] |
|
[32] |
|
/
〈 |
|
〉 |