>  Docs Center  >  Using ENVI  >  Perform Atmospheric Correction


Perform Atmospheric Correction

Perform Atmospheric Correction

In Step 2 of the Target Detection Wizard, you can optionally perform atmospheric correction on the input image. Most images should be converted to reflectance through atmospheric correction before running target detection, especially if library spectra or other external spectra are used to define targets in the process. If external spectra are not used in the processing, you can use radiance or uncalibrated data as input.

If your image is already corrected, or if you choose to use ROIs as target spectra, select None / Already Corrected to skip this step.

  1. Select one of the following atmospheric correction methods:
    • QUick Atmospheric Correction (QUAC®): (Requires a licensed installation of the Atmospheric Correction Module.) Determines the atmospheric compensation parameters directly from the information contained within the scene using the observed pixel spectra. This method assumes that the average reflectance of a collection of diverse material spectra, such as the endmember spectra in a scene, is effectively scene-independent. This enables the retrieval of reasonably accurate reflectance spectra even with radiometrically uncalibrated data. Select the Sensor Type from the drop-down list. If the sensor type is not available, select Unknown. If the sensor type is in the data header file and supported in the drop-down list, ENVI selects it by default.
    • IAR Reflectance: Normalizes images to a scene average spectrum. This is particularly effective for reducing spectral data to relative reflectance when little is known about the scene. It works best for arid areas with no vegetation. If you selected a spatial subset in the input file, you can choose to compute the Average Area over the spatial subset or the entire image.
    • Log Residuals: Creates a pseudo reflectance image that is useful for analyzing mineral-related absorption features. The logarithmic residuals of a dataset are defined as the input spectrum divided by the spectral geometric mean, then divided by the spatial geometric mean.
    • Dark Subtraction: Subtracts the dark noise from the image. The digital number to subtract from each band can be either Band Minimums, or User Values for each band. If you select Band Minimums, you can choose a Dark Search Area over the spatial subset or, if you selected a spatial subset at file input, over the entire image. You can also specify a Data Ignore Value. If you select User Values, you can edit Current Subtraction Values for each band.
    • Flat Field: Similar to the IAR Reflectance method, Flat Field normalizes images to an area of known “flat” reflectance. This is particularly effective for reducing hyperspectral data to relative reflectance. You can create a Flat Field ROI, or select an existing Flat Field ROI. The average spectrum from the ROI is used as the reference spectrum, which is then divided into the spectrum at each pixel of the image.
    • Empirical Line: Forces the image spectra to match selected field reflectance spectra. A linear regression is used for each band to equate DN and reflectance. This method is capable of producing the most accurate results, but it requires ground truth information. You need to select at least one Spectra Pair to continue. See Empirical Line Calibration for details.
  2. Click Next in the Wizard. The Select Target Spectra panel appears.

© 2019 Harris Geospatial Solutions, Inc. |  Legal
My Account    |    Store    |    Contact Us