首页> 外文会议>Earth Observing Systems XI >QuickBird relative radiometric performance and on-orbit long term trending
【24h】

QuickBird relative radiometric performance and on-orbit long term trending

机译:QuickBird相对辐射性能和在轨长期趋势

获取原文
获取原文并翻译 | 示例

摘要

One of the top radiometric priorities of the high-spatial resolution, commercial remote sensing industry is to achieve a superior level of image quality in all imagery products. Errors in detector gain and offset correction during product generation create noticeable image artifacts such as banding and streaking that degrade the overall image quality. Banding and streaking can be minimized by relative radiometric calibration, however, this calibration is only a temporary solution as the gain and offset of each detector will drift over time. The work presented here examines the relative radiometric performance of the QuickBird panchromatic and multispectral bands and tracks the performance from January 2005 until the present. During radiometric operations, uniform scenes of desert, ocean, forest, and snow areas are identified in the DigitalGlobe ImageLibrary. Products for these uniform scenes are generated and detector statistics are calculated for each scene. The QuickBird focal plane contains detectors that are masked so that no light reaches them. The statistics for the masked detectors are analyzed to study image-to-image variability and determine changes in detector offsets over time. Next, the detector averages for all active detectors are radiometrically corrected, and banding and streaking metrics are applied. Banding and streaking are trended to monitor changes over time. Quality metrics are also established based on the banding and streaking results to determine when the detector's gains and offsets have drifted sufficiently to require recalibration. Relative radiometric performance of the uniform scenes is compared with and without recalibration.
机译:商业遥感行业是高空间分辨率的最重要的辐射优先事项之一,是在所有图像产品中都实现卓越的图像质量。产品生成过程中检测器增益和偏移校正的错误会产生明显的图像伪影,例如条带和条纹,会降低整体图像质量。相对辐射定标可以最大程度地减少条带和条纹,但是,此定标只是暂时的解决方案,因为每个检测器的增益和偏移都会随时间漂移。本文介绍的工作检查了QuickBird全色和多光谱波段的相对辐射测量性能,并跟踪了从2005年1月至今的性能。在辐射测量操作期间,DigitalGlobe ImageLibrary中会识别出沙漠,海洋,森林和积雪区域的统一场景。生成这些统一场景的产品,并为每个场景计算检测器统计信息。 QuickBird焦平面包含被遮盖的探测器,因此没有光到达它们。分析掩盖检测器的统计数据,以研究图像间的差异,并确定检测器偏移量随时间的变化。接下来,对所有活动探测器的探测器平均值进行辐射校正,并应用条带和条纹度量。趋势是条带化和条纹化以监视随时间的变化。还基于条带和条痕结果建立质量度量,以确定何时检测器的增益和偏移已充分漂移以至于需要重新校准。比较有无重新校准情况下统一场景的相对辐射性能。

著录项

相似文献

  • 外文文献
  • 中文文献
  • 专利
获取原文

客服邮箱:kefu@zhangqiaokeyan.com

京公网安备:11010802029741号 ICP备案号:京ICP备15016152号-6 六维联合信息科技 (北京) 有限公司©版权所有
  • 客服微信

  • 服务号