地理研究 ›› 2012, Vol. 31 ›› Issue (11): 2007-2016.doi: 10.11821/yj2012110009

• 地球信息科学 • 上一篇    下一篇

海岸带地区SPOT卫星影像大气校正方法比较及精度验证

韩晓庆1, 苏艺2, 李静3, 张芸4, 刘佳2, 高伟明2   

  1. 1. 河北省国土资源利用规划院, 石家庄050051;
    2. 河北师范大学资源与环境科学学院河北省环境演变与生态建设实验室, 石家庄050024;
    3. 河北交通职业技术学院交通工程系, 石家庄050035;
    4. 河北省地理科学研究所, 石家庄050011
  • 收稿日期:2011-11-05 修回日期:2012-04-15 出版日期:2012-11-10 发布日期:2012-11-10
  • 通讯作者: 高伟明(1962-),男,河北滦县人,教授,主要从事海洋资源环境评价与可持续利用研究。E-mail:gaowmd@263.Net E-mail:gaowmd@263.Net
  • 作者简介:韩晓庆(1981-),男,山东莱州人,硕士,工程师,主要从事基于“3S”的环境演变研究。E-mail:xqhan312@126.com
  • 基金资助:

    河北省应用基础研究计划重点基础研究项目(09966719D);国家海洋局“908”专项河北省海岸带及海岛调查(HE908-01-03-04)

Atmospheric correction and verification of the SPOT remote sensing image in coastal zones

HAN Xiao-qing1, SU Yi2, LI Jing3, ZHANG Yun4, LIU Jia2, GAO Wei-ming2   

  1. 1. Hebei Utilization and Planning Institute of Land Resources, Shijiazhuang 050051, China;
    2. Hebei Key Laboratory of Environmental Change and Ecological Construction, College of Resources and Environmental Science, Hebei Normal University, Shijiazhuang 050024, China;
    3. Department of Traffic Engineering, Hebei Vocational and Technical College of Communications, Shijiazhuang 050035, China;
    4. Hebei Institute of Geographical Sciences, Shijiazhuang 050011, China
  • Received:2011-11-05 Revised:2012-04-15 Online:2012-11-10 Published:2012-11-10

摘要: 为获得反演海岸带地区地表真实反射率的最佳大气校正方法,分别利用DOS、6S及FLAASH三种模型对SPOT卫星影像进行大气校正,对比分析校正前后影像的典型地物光谱曲线及NDVI指数的变化情况,并对结果进行精度验证。研究表明:三种大气校正模型均能有效去除大气对遥感影像的影响,获取较为准确的地表反射率信息;对比河流、水库、海水、植被、城镇、裸地等六种典型地物校正前后的光谱曲线,与DOS及6S模型相比,FLAASH模型校正结果中各地物反射率差别较大、特征明显,能较好的还原地表真实情况;NDVI指数检验显示,FLAASH模型对地物NDVI响应更强烈,在地物光谱响应特征分析等方面具有更强的优势,是三种模型中最适宜海岸带地区SPOT影像的大气反演方法。研究方法及结论可为同类研究提供参考与借鉴。

关键词: 海岸带, SPOT影像, DOS模型, 6S模型, FLAASH模型, 大气校正, 精度验证

Abstract: In order to obtain the best atmospheric correction about true surface reflectance in the coastal zones, three models, i.e, DOS, 6S and FLAASH, were used to correct the atmospheric condition of SPOT image.In comparison of the change conditions by different index factors, including spectrum curves of typical surface features and NDVI, the accuracy of results were verified.The results are shown as follows.Firstly, three atmospheric correction models could be used to obtain surface reflectance information correctly, and to remove the effect of the atmosphere on remote sensing images, which is caused by different atmospheric conditions.Secondly, the paper took spectral curves of six typical objects as an example to compare correction effects, such as river, reservoir, sea, vegetation, urban and bare soil.FLAASH model, compared with DOS and 6S model, showed greater differences and more obvious characteristics in the reflectivity of surface features.Except that, it was better in coinciding with spectrum curves of real typical surface features than the others.Thirdly, FLAASH model responded to NDVI index of surface features more strongly.Furthermore, it was the best one about atmospheric correction of SPOT satellite images in coastal zones.The study methods and results could provide the reference for similar studies.

Key words: coastal zone, SPOT image, DOS model, 6Smodel, FLAASH model, atmospheric correction, verification