Cloud Detection and Haze Removal module


Quick linksDescription :: Parameters :: Parameter descriptions :: Details

Back to top

Description


The Cloud Detection and Haze Removal module generates cloud masks and removes haze from multiple images.
Back to top

Parameters


Name Caption
Input Orthos Input orthorectified scenes
Output Folder Output directory
Output File Type Output file type
Output File Options Output file options
Overwrite Results Overwrite existing results
Send Email Email notification settings
Source Background Type Source background type
Source Background Value Source background pixel value
Lower Cloud Reflectance Threshold Lower cloud reflectance threshold (haze band)
Upper Cloud Reflectance Threshold Upper cloud reflectance threshold (haze band)
Water Threshold Upper water reflectance thresholds (NIR, SWIR)
Cloud Dilation Cloud dilation factor
Calibration File Text file of calibration coefficients
Remove Small Clouds Whether to remove small clouds
Minimum Size Clouds smaller than the minimum size will be removed
Unit Unit for minimum size
Smooth Cloud Polygons Generate smooth cloud polygons
Remove Haze Remove haze from output image
Haze Coverage Percentage of pixels covered by haze
Back to top

Parameter descriptions

Input Orthos

Specifies the path of the folder containing orthorectified satellite scenes to be atmospherically corrected.

Input scenes must follow the standard PCI naming convention; that is, supported file names end with "ORTHO_MS.PIX".

This parameter is mandatory. See Details for more information.

Output Folder

The path of the output folder to which to write the coarse classification masks, and the haze-corrected (HZC) images.

This parameter is mandatory.

Output File Type

The format of the output file.

For more information on the supported file formats, see GDB-supported file formats.

Output File Options

The options to apply when creating the output file or files. The available options are specific to the file format; in each case, the default of no options is allowed.

For more information on the options available for the output file type you specify, see GDB-supported file formats.

Overwrite Results

Select this check box to overwrite the existing output files, if any exist. If this check box is left clear, and an output file exists in the relevant folder, the status of the job displays a message informing you of the existence and name of the output file. The message is also written to the event log of the job.

Send Email

If necessary, you can set up CATALYST Enterprise to send an email notification on job start and job completion.

With this check box selected, an email message is sent to each address specified in the Email Addresses box after the job starts and on completion.

You can specify one or more addresses, and each must be separated by a comma or a semi-colon. The email address of the user currently logged in displays by default.

Source Background Type

The method to use to determine which pixels in the source image to process as background (NoData) pixels. In general, if a pixel is considered NoData, the module processes it in a specific manner.

If the Any option or the All option is selected, a value must be specified for the Source Background Value parameter.

Available options are:

For specific examples, see the Source Background Value parameter description.

Source Background Value

The source background value or values when the Source Background Type parameter is set to:

The source background value is provided as either a single number (applied to all channels) or as a pixel "stack" (a comma-delimited list of values). If a pixel stack is provided, but the number of values does not equal the number of channels, the list is truncated or the last value is repeated as necessary. The background values provided is truncated to the range allowed by the source image data type.

The following examples apply to a 3-channel, 8-bit unsigned image:

Lower Cloud Reflectance Threshold

Optionally specifies, in percentage (1-100), the lower empirical threshold (bound) used to identify the minimum TOA (Top of Atmosphere) reflectance value in the blue or green haze band used to determine the seeds of clouds (the central part of the clouds).

Note: Lower thresholds tend to identify more pixels as cloud, but may also misclassify more urban features with unnaturally high reflectance as clouds.
If this parameter is not specified, the default Lower Cloud Reflectance Threshold value is 20 percent, except for the following sensors:

Upper Cloud Reflectance Threshold

Optionally specifies, in percentage (1-100), the minimum Top of the Atmosphere reflectance (in percentage) in the blue (or green) band, but it is used to capture the area of the cloud not captured by the lower threshold (the extension of the cloud). This value must be higher than that specified for the Lower Cloud Reflectance Threshold.

Note: Lower thresholds may help to cover a larger area of the cloud, but may also misclassify brighter features as cloud.
If this parameter is not specified, the default Upper Cloud Reflectance Threshold value is 26 percent, except for the following sensors:

Water Threshold

Specifies, in percentage, the upper reflectance threshold used to identify water pixels in the atmosphere.

This parameter is specified as follows:
Water Threshold = NIR_THRESH, SWIR_THRESH

If this parameter is not specified, a default value of 5 is used for the NIR band and a default value of 3 is used for the SWIR band (when applicable).

Cloud Dilation

Optionally specifies the amount, in pixels, by which to dilate, or expand, the cloud so that the 'halo' effect in the cloud is eliminated (that is, covered by the cloud mask).

If this parameter set to 0, no dilation occurs. If this parameter is not specified, the default cloud dilation factor is 2 pixels. The maximum dilation factor is 200 pixels.

Calibration File

The path and file name of the text file that contains, for each band, the calibration coefficients used to transform the values from the image to absolute radiance values. Typically, this information is provided with the data set as an offset and a gain for each channel. The specified text file should contain the band number, offset and gain, and the units for each. The following shows an example of a calibration file for SPOT Landsat-7 images:

7          c0       c1         [mW/cm2 sr micron]
1   	-0.6200   0.0776000
2   	-0.6400   0.0796000  
3   	-0.5000   0.0619000
4   	-0.5100   0.0637000
5   	-0.1002   0.0126000 
6   	 0.0000   0.00670866
7   	-0.0350   0.00437000				
				
The folder atcor\cal in your installation folder contains sample calibration files for each supported sensor. In a calibration file, each row lists the calibration coefficients for each sensor band; the bands are listed in increasing order. You must ensure that the offset (c0) and gain (c1) values correspond to the scene. These values must be provided in the expected units for the specific sensor (shown at the top right of the calibration file). If no value is specified for this parameter, the following metadata tags for each channel in the input file are used:

Also, if no value is specified for Calibration File, or if the coefficients cannot be derived from the metadata, an error message will be displayed.

Important: Do not modify the calibration files, because they indicate the nominal bands for each sensor.

Remove Small Clouds

Selected by default, this check box controls whether to remove small clouds from the output imagery.

You must enter a value for Minimum Size and select a value for Unit.

Minimum Size

The size of clouds smaller than the minimum to remove based on the value selected for Unit.

When you specify a value, you must select a value for Unit.

Unit

The unit of the value specified for Minimum Size, the available options are:

Smooth Cloud Polygons

Select this check box to smooth the cloud polygons to provide continuous change in curvature.

When selected, the smoothing algorithm cuts through some pixel edges rather than following the actual edge of each pixel.

Remove Haze

Select this check box to remove haze and, therefore, create haze-corrected (HZC) image files.

Haze Coverage

The approximate percentage of coverage of hazy pixels in the image.

This value you specify determines the size of the haze mask to generate to identify haze pixels. You can base the percetage you specify on the total number of pixels in the image, excluding saturated pixels. By default, haze coverage is approximately 50 percent.

This parameter is optional.

Back to top

Details

General job details

Preprocessing requirements

Before running this module, the following requirements must be met to ensure the job processes successfully and produces accurate results:

Module details

The input for the Cloud Detection and Haze Removal module is a folder of multispectral, orthorectified image scenes that have already been processed using the Orthorectification module. As such, the input files contain the “_ORTHO_MS” tag in the file name. For example:

<SCENE_ID>_ORTHO_MS.pix
      

The Cloud Detection and Haze Removal module reads the folder of input orthos for multispectral, orthorectified image scenes. A child job is created for each scene. The module generates coarse classification masks (HOTMASK file), which are used to perform the haze-removal operation for any of the supported multispectral sensors.

When the classification masks are created successfully, the haze-removal process is run on the multispectral image to create a haze-corrected (HZC) image.

For each HZC image created, the module creates the following masks to use during mosaic preparation: The masks are then exported to the final atmospherically corrected image.

Job results

On successful completion, the Cloud Detection and Haze Removal module produces the following output for each valid input scene:
  • A hazefree folder, which includes an atmospherically corrected image and polygon-cloud mask for each valid input scene.
  • A cloud folder, which includes each input scene containing its associated polygon-cloud mask. The folder can contain an additional masks folder, which contains only the polygon-cloud masks generated.

© PCI Geomatics Enterprises, Inc.®, 2026. All rights reserved.