PCI Geomatica‘s new Atmospheric Correction wizard provides users with the easiest and fastest way to perform a variety of atmospheric corrections. The wizard automatically populates most of the required parameters using the image metadata and guides the user through each major step.
Initial Project Setup
- Open Geomatica 2013’s Focus application
a. To do this through the Geomatica toolbar, click on the Focus button
2. Turn Auto-Re-enhance on
a. In the menu bar, click on the Tools dropdown menu and select Options…
b. In the Options panel, click on the Layers branch in the left pane
c. Click the checkbox next to Auto re-enhance grayscale and RGB layers
3. Navigate to the location of the data
4. Drag and drop the .pix file into Focus
Note that you must import most types of data into a PIX format before running ATCOR as much of the metadata is now read directly from the file. You can use one of CD* import program to import the data (for example, CDQB or CDIKONOS). You can find more information in the General Help in Focus. In Focus simply click on Help General. In the Contents tab navigate to Focus Atmospheric CorrectionPreparing data for atmospheric correction. If you are using Pleiades imagery, use the FIMPORT algorithm found in the Algorithm Librarian.
5. Correct the band-channel combination so that you render the true-color composite
a. Click on Focus’ Maps tab
b. Expand the image branch so that the three color guns are displayed
c. Right click on the Red color gun and select Channel 3
d. Right click on the Blue color gun and select Channel 1
- In the menu bar, select Analysis --> Atmospheric Correction --> ATCOR – Ground
- The Atmospheric Correction panel opens and contains 4 standalone workflows.
ATCOR Wizard General Design
The ATCOR wizard includes 4 completely standalone workflows that can be run completely independently of one another. The four (4) workflows are:
• TOA (Top of Atmosphere) Reflectance
Calculates top of atmosphere reflectance for any supported sensor
• Haze Removal
Performs Cloud masking and haze removal and outputs the result in DN values (Scaled at-sensor radiance)
• ATCOR – Ground Reflectance
Calculates ground reflectance values for supported optical imagery and optionally performs haze removal and cloud masking. The output is in reflectance values (0-100%)
• ATCOR – Surface Temperature
Calculates surface temperature for TIR bands from supported satellite sensors.
The ATCOR wizard is equipped with indicator symbols that indicate to a user that a given major step in the workflow is correctly setup and the user can proceed to the next step.
8. Select the Sensor and image settings step in the ATCOR – Ground Reflectance workflow
9. In the Input Image Files section, click on the Multispectral dropdown list and select the input .pix file from the list or Browse to the file location.
Automatic Parameter Setup:
The ATCOR wizard will automatically read the metadata that is ingested when the image is imported into a PIX file. Where necessary, these values will be converted so that they can be used inside ATCOR. The following is a list of Sensor parameters that are automatically set:
- Sensor Type
- Solar Zenith & Azimuth
- Acquisition Date
- Calibration Coefficients
- Band-Channel Combination
Note: This is only a partial listing of the parameters that are automatically set using the image metadata.
Channel-Band setup & Calculation of Calibration Coefficients
- In the Radiometric Information section, choose the calibration coefficients source: Channel Metadata (used in this example), Import from Text File or Manual Entry. Click on Band Setup… button
- The Band Setup panel opens with the following table
- In the Output Files section, click on the Browse button and select your output file directory
- Click OK
- Click Next to proceed to the Haze and cloud masking step
Haze & Cloud Masking:
PCI Geomatics has developed enhanced haze removal and cloud masking algorithms that can significantly improve results of the haze removal and cloud masking operations.
The haze removal is more stable and settings have been defined that are sensor specific to help ensure higher quality results on a sensor by sensor basis.
The haze removal algorithm can be operated as standalone, which will generate haze free images in the DN domain. However, in this workflow (ATCOR – Ground Reflectance) the haze removed results will be converted to reflectance values.
The cloud masking algorithm now uses a new seeding & filtering concept that provides the following benefits:
- Significantly less false positives in urban areas
- Cloud mask covers identified clouds right to the edges
- Dilation parameter allows users to have the mask extend beyond the cloud edge to prevent cloud rings in mosaics
- In the Water Masking section, click on the Omit radio button
- Use the default values for the remaining settings in this step
- Click Next to proceed to the Illumination conditions step
About Height Estimation for Flat Scenes
The constant height is automatically calculated as the average elevation of the scene bounds, as determined from the DEM distributed with Geomatica 2013.
- In the Elevation Information Setup section, click on the DEM file radio button
- Click Browse… and navigate to the DEM you wish to use
ATCOR’s New DEM Handling:
Geomatica 2013’s new Atmospheric Correction Wizard makes using DEMs, for the purpose of 3D atmospheric correction, easier than ever before.
Before now, users were required to perform time consuming preparation of DEMs before being able to use them in ATCOR. Users were required to ensure that the DEMs had identical extents, resolution and projection as the input image. This additional setup requirement was often very time consuming and complicated, because the optimal resampling technique depends on the DEMs resolution, relative to the input image.
Geomatica’s 2013 Atmospheric Correction Wizard now automatically performs all of the necessary DEM preparations. The wizard is will compare the resolution of the DEM to the input image and automatically determine the optimal resampling technique.
18. Click Next to proceed to the Visibility and ground reflectance step.
19. In the Visibility Map section, change the constant visibility value to 23km
20. Use the default settings for the remaining step.
Visibility and ground reflectance step:
The atmospheric Information parameters are used to define the atmospheric model that is used to convert the at-sensor radiance values to surface reflectance values. Three pieces of information are required:
- Aerosol type (molecular and particulate makeup of atmosphere)
- a. This parameter must be defined by the user
- Condition (Water vapor content of atmosphere
- a. This parameter is automatically estimated based on the scene’s latitude and acquisition date/time
- Visibility (aerosol optical depth (AOD)
- a. User can manually adjust this value until an optimum is found or select Calculate to auto-estimate the AOD from dark vegetation
The satellite viewing geometry is automatically calculated from sensor metadata for those sensors with tilt capabilities.
The default output is the reflectance value in floating point precision from 0-100%. The user can optionally scale the values to fit into 8 or 16 bit channels.
- Click Run
- After the Atmospheric correction processing completes, the outputs will be automatically displayed on the screen. You will have 1 RGB image layer and 2 bitmap (mask) layers and 2 greyscale images. They will appear in the Maps tab in the following order:
- RGB – Atmospherically corrected image
- Bitmap – Haze Mask
- Bitmap – Water Mask
- Bitmap – Cloud Mask
- Resampled DEM
2. Correct the band-channel combination for the atmospherically corrected image so that you render the true-color composite.
a. Red = Channel 3
b. Green = Channel 2
c. Blue = Channel 1
3. Apply an Enhancement
4. Zoom to overview of the image
5. Show the illumination map
About Illumination Map
This map helps to normalize affects caused by solar illumination. The map helps the algorithm determine how illuminated a pixel is regardless of the surface feature type. This is based off of the solar angle, relative to the slope and aspect angle of the pixel. For example, pixels oriented towards the sun are brighter, whereas pixels oriented away from the sun will be darker in the map6.
6. Uncheck the illumination map to show the haze mask. The haze mask covers pixels that the algorithm has identified as haze and can be corrected.
7. Uncheck the cloud mask to show the cloud underneath
8. Toggle the raw (input) image on and off to see the results of the atmospheric correction