Atmospheric correction


Atmospheric correction is the process of adjusting images taken by satellite or airborne sensors to remove distortions caused by the atmosphere. These distortions—mainly due to the scattering and absorption of sunlight by particles and gases—can affect how accurately the sensor captures the true reflectance of the Earth's surface.
In remote sensing, atmospheric effects can significantly alter the spectral characteristics of the radiation detected by sensors. This occurs because light must pass through the atmosphere twice—first as sunlight traveling to the Earth's surface, and again as reflected light returning to the sensor—undergoing both absorption and scattering along the way. These distortions can affect the accuracy of surface reflectance measurements and are typically corrected through a range of physical and statistical methods.

Examples of atmospheric correction methods

SensorApproach
MSSband-to-band regression
MSSall-band spectral covariance
airborne MSSband-to-band regression
AVHRRiterative estimation
MSS, TMDOS with exponential scattering model
TMDOS with exponential scattering model, downwelling atmospheric radiance measurements
TMpixel-by-pixel tasseled cap haze parameter
AVHRRDOS, NDVI, AVHRR band 3
airborne TMS, Landsat TMground and airborne solar measurements, atmospheric modeling code
TMcomparison of ten DOS and atmospheric modeling code variations with field data
TMdark target, modeling code
TM atmospheric modeling code, region histogram matching
TMDOS with estimated atmospheric transmittance
TMdark target, atmospheric modeling code
TM, ETM+empirical line method, single target, ground measurements
TMwater reservoirs, comparison of 7 methods for 12 dates
AVHRR2-band PCT used to separate aerosol components