ASF_Tools.RGBDecomp.pyt.xml

<?xml version="1.0"?>
<metadata xml:lang="en"><Esri><CreaDate>20200918</CreaDate><CreaTime>13534800</CreaTime><ArcGISFormat>1.0</ArcGISFormat><SyncOnce>TRUE</SyncOnce><ModDate>20201002</ModDate><ModTime>15000700</ModTime><scaleRange><minScale>150000000</minScale><maxScale>5000</maxScale></scaleRange><ArcGISProfile>NAP</ArcGISProfile><DataProperties><itemProps><imsContentType export="False"/></itemProps></DataProperties></Esri><tool name="RGBDecomp" displayname="RGB Decomposition" toolboxalias="ASF Tools" xmlns=""><arcToolboxHelpPath>c:\program files\arcgis\pro\Resources\Help\gp</arcToolboxHelpPath><parameters><param name="indir" displayname="Input directory containing dual-pol RTC data" type="Required" direction="Input" datatype="Workspace" expression="indir"><dialogReference>&lt;DIV STYLE="text-align:Left;"&gt;&lt;DIV&gt;&lt;DIV&gt;&lt;P&gt;&lt;SPAN&gt;Navigate to the directory for an &lt;/SPAN&gt;&lt;A href="https://hyp3-docs.asf.alaska.edu/guides/rtc_product_guide/" target="_blank" STYLE="text-decoration:underline;"&gt;On Demand RTC product generated using HyP3&lt;/A&gt;&lt;SPAN&gt;, or other directory containing the co- and cross-pol datasets. You can also enter the path by typing or pasting it, or by dragging the folder from the Catalog window.&lt;/SPAN&gt;&lt;/P&gt;&lt;P&gt;&lt;SPAN&gt;The two input rasters (co-pol and dual-pol) must be in the same folder, with filenames ending in XX.tif where XX is the polarization (VV, VH, HV, HH). There can only be one co-pol and one cross-pol raster in the input directory. The datasets can be in power, amplitude, or dB scale.&lt;/SPAN&gt;&lt;/P&gt;&lt;P&gt;&lt;SPAN&gt;The base filename of the input RTC products must be the same as the directory name (i.e. directory_name_VV.tif), and must follow the pattern used by HyP3 to name On Demand RTC products.&lt;/SPAN&gt;&lt;/P&gt;&lt;P&gt;&lt;SPAN&gt;If the directory or RTC products have been renamed or generated by a different process, they will need to be renamed to match the HyP3 file naming system: S1X_XX_yyyymmddThhmmss_DPX_RTCXX_G_xsxxxx_XXXX, where the polarization (P in DPX) is populated with the primary polarization (DVX for VV/VH, DHX for HH/HV), and the scale (s in xsxxxx) is populated with p (power) a (amplitude) or d (decibel). You could use the example string with all the x values (or populate the rest of the string with values that help identify the product), as long as those two variables are populated and the number of characters does not change.&lt;/SPAN&gt;&lt;/P&gt;&lt;/DIV&gt;&lt;/DIV&gt;&lt;/DIV&gt;</dialogReference></param><param name="scale" displayname="Scale of input RTC (power, amplitude, or dB)" type="Required" direction="Input" datatype="String" expression="Power | Amplitude | dB"><dialogReference>&lt;DIV STYLE="text-align:Left;"&gt;&lt;DIV&gt;&lt;DIV&gt;&lt;P&gt;&lt;SPAN&gt;Select the scale of the input datasets from the dropdown menu (Power, Amplitude, or dB).&lt;/SPAN&gt;&lt;/P&gt;&lt;P&gt;&lt;SPAN&gt;If the input directory has been extracted from an &lt;/SPAN&gt;&lt;A href="https://hyp3-docs.asf.alaska.edu/guides/rtc_product_guide/" target="_blank" STYLE="text-decoration:underline;"&gt;On Demand RTC product generated using HyP3&lt;/A&gt;&lt;SPAN&gt;, this parameter will automatically populate based on the directory name.&lt;/SPAN&gt;&lt;/P&gt;&lt;P&gt;&lt;SPAN&gt;If the directory or RTC products have been renamed or generated by a different process, they will need to be renamed to match the HyP3 file naming system: S1X_XX_yyyymmddThhmmss_DPX_RTCXX_G_xsxxxx_XXXX, where the polarization (P in DPX) is populated with the primary polarization (DVX for VV/VH, DHX for HH/HV), and the scale (s in xsxxxx) is populated with p (power) a (amplitude) or d (decibel). You could use the example string with all the x values (or populate the rest of the string with values that help identify the product), as long as those two variables are populated and the number of characters does not change.&lt;/SPAN&gt;&lt;/P&gt;&lt;P&gt;&lt;SPAN STYLE="font-style:italic;"&gt;Note that if you change the input directory after this parameter is populated, it does not automatically update to match the new input directory information. Delete the outdated parameter and press tab or click away from the field to reset it to reflect the new input directory if necessary.&lt;/SPAN&gt;&lt;/P&gt;&lt;/DIV&gt;&lt;/DIV&gt;&lt;/DIV&gt;</dialogReference></param><param name="pol" displayname="Primary polarization (V or H)" type="Required" direction="Input" datatype="String" expression="V | H"><dialogReference>&lt;DIV STYLE="text-align:Left;"&gt;&lt;DIV&gt;&lt;DIV&gt;&lt;P&gt;&lt;SPAN&gt;Select the primary polarization from the dropdown menu (either V for VV/VH datasets or H for HH/HV datasets).&lt;/SPAN&gt;&lt;/P&gt;&lt;P&gt;&lt;SPAN&gt;If the input directory has been extracted from an &lt;/SPAN&gt;&lt;A href="https://hyp3-docs.asf.alaska.edu/guides/rtc_product_guide/" target="_blank" STYLE="text-decoration:underline;"&gt;On Demand RTC product generated using HyP3&lt;/A&gt;&lt;SPAN&gt;, this parameter will automatically populate based on the directory name.&lt;/SPAN&gt;&lt;/P&gt;&lt;P&gt;&lt;SPAN&gt;If the directory or RTC products have been renamed or generated by a different process, they will need to be renamed to match the HyP3 file naming system: S1X_XX_yyyymmddThhmmss_DPX_RTCXX_G_xsxxxx_XXXX, where the polarization (P in DPX) is populated with the primary polarization (DVX for VV/VH, DHX for HH/HV), and the scale (s in xsxxxx) is populated with p (power) a (amplitude) or d (decibel). You could use the example string with all the x values (or populate the rest of the string with values that help identify the product), as long as those two variables are populated and the number of characters does not change.&lt;/SPAN&gt;&lt;/P&gt;&lt;P&gt;&lt;SPAN STYLE="font-style:italic;"&gt;Note that if you change the input directory after this parameter is populated, it does not automatically update to match the new input directory information. Delete the outdated parameter and press tab or click away from the field to reset it to reflect the new input directory if necessary.&lt;/SPAN&gt;&lt;/P&gt;&lt;/DIV&gt;&lt;/DIV&gt;&lt;/DIV&gt;</dialogReference></param><param name="rb_thresh_db" displayname="Threshold cutoff value for red/blue in dB (default: -24.0 dB)" type="Required" direction="Input" datatype="Double" expression="rb_thresh_db"><dialogReference>&lt;DIV STYLE="text-align:Left;"&gt;&lt;DIV&gt;&lt;DIV&gt;&lt;P&gt;&lt;SPAN&gt;Set the red/blue threshold (dB).&lt;/SPAN&gt;&lt;/P&gt;&lt;P&gt;&lt;SPAN&gt;The default value is -24 dB, which works well for most Sentinel-1 IW datasets, but you can adjust the threshold if desired. Suitable values typically fall in the range of -19 to -25 dB.&lt;/SPAN&gt;&lt;/P&gt;&lt;/DIV&gt;&lt;/DIV&gt;&lt;/DIV&gt;</dialogReference></param><param name="outdir" displayname="Output directory for new RGB file" type="Required" direction="Input" datatype="Workspace" expression="outdir"><dialogReference>&lt;DIV STYLE="text-align:Left;"&gt;&lt;DIV&gt;&lt;DIV&gt;&lt;P&gt;&lt;SPAN&gt;Navigate to the destination directory where the output will be saved. &lt;/SPAN&gt;&lt;/P&gt;&lt;P&gt;&lt;SPAN&gt;By default this is set to be the same as the input directory. You can amend the path, or click the folder icon to navigate to a different output directory if desired.&lt;/SPAN&gt;&lt;/P&gt;&lt;P&gt;&lt;SPAN STYLE="font-style:italic;"&gt;Note that if you change the input directory after the default directory populates, the output directory parameter does not automatically update. Delete the outdated directory and press tab or click away from the field to reset the output directory to reflect the new input directory.&lt;/SPAN&gt;&lt;/P&gt;&lt;/DIV&gt;&lt;/DIV&gt;&lt;/DIV&gt;</dialogReference></param><param name="outname" displayname="Filename for new RGB raster" type="Required" direction="Input" datatype="String" expression="outname"><dialogReference>&lt;DIV STYLE="text-align:Left;"&gt;&lt;DIV&gt;&lt;DIV&gt;&lt;P&gt;&lt;SPAN&gt;Enter the filename for the output raster, including the .tif extension.&lt;/SPAN&gt;&lt;/P&gt;&lt;P&gt;&lt;SPAN&gt;By default, this parameter is set to add an _RGB.tif tag to the input directory name. You can amend the name, or enter an entirely different filename if desired.&lt;/SPAN&gt;&lt;/P&gt;&lt;P&gt;&lt;SPAN STYLE="font-style:italic;"&gt;Note that if you change the input directory after this parameter is populated, it does not automatically update to match the new input directory information. Delete the outdated filename and press tab or click away from the field to reset the parameter to reflect the new input directory if necessary.&lt;/SPAN&gt;&lt;/P&gt;&lt;/DIV&gt;&lt;/DIV&gt;&lt;/DIV&gt;</dialogReference></param><param name="out_yn" displayname="Add output to map" type="Required" direction="Input" datatype="Boolean" expression="out_yn"><dialogReference>&lt;DIV STYLE="text-align:Left;"&gt;&lt;P&gt;&lt;SPAN&gt;Select this option to automatically add the output raster to the map. &lt;/SPAN&gt;&lt;/P&gt;&lt;P&gt;&lt;SPAN&gt;This option is selected by default. Remove the check if you do not want the output raster added to the map.&lt;/SPAN&gt;&lt;/P&gt;&lt;/DIV&gt;</dialogReference></param></parameters><summary>&lt;DIV STYLE="text-align:Left;"&gt;&lt;DIV&gt;&lt;DIV&gt;&lt;P&gt;&lt;SPAN&gt;This tool generates an RGB Decomposition raster from a dual-pol RTC dataset. It was designed for use with Sentinel-1 RTC products generated with HyP3, including the &lt;/SPAN&gt;&lt;A href="https://hyp3-docs.asf.alaska.edu/guides/rtc_product_guide/" target="_blank" STYLE="text-decoration:underline;"&gt;On Demand RTC products you can request in Vertex&lt;/A&gt;&lt;SPAN&gt;, but can be used with any dual-pol RTC dataset that is in power, amplitude, or dB scale. &lt;/SPAN&gt;&lt;/P&gt;&lt;P&gt;&lt;SPAN&gt;-----&lt;/SPAN&gt;&lt;/P&gt;&lt;P&gt;&lt;SPAN&gt;The parameters are populated with default values once you select the input directory containing the RTC datasets. You can amend any of the auto-populated outputs or set them to something completely different if desired. Note that if you change your input directory, the default values will not automatically change. You can reset the auto-populated outputs based on the new parameters by deleting the default string and pressing tab or clicking away from the field.&lt;/SPAN&gt;&lt;/P&gt;&lt;P&gt;&lt;SPAN&gt;-----&lt;/SPAN&gt;&lt;/P&gt;&lt;P&gt;&lt;SPAN&gt;For best results when viewing the output image, set the symbology to be an RGB Composite stretched to Minimum-Maximum with no gamma stretch applied. The default symbology is generally an RGB Composite stretched to a Percent Clip with a gamma stretch applied.&lt;/SPAN&gt;&lt;/P&gt;&lt;/DIV&gt;&lt;/DIV&gt;&lt;/DIV&gt;</summary><usage>&lt;DIV STYLE="text-align:Left;"&gt;&lt;DIV&gt;&lt;DIV&gt;&lt;P&gt;&lt;SPAN&gt;This tool generates an RGB Decomposition raster from a dual-pol RTC dataset. It was designed for use with Sentinel-1 RTC products generated with HyP3, including the &lt;/SPAN&gt;&lt;A href="https://hyp3-docs.asf.alaska.edu/guides/rtc_product_guide/" target="_blank" STYLE="text-decoration:underline;"&gt;On Demand RTC products you can request in Vertex&lt;/A&gt;&lt;SPAN&gt;, but can be used with any dual-pol dataset that is in power, amplitude, or dB scale. &lt;/SPAN&gt;&lt;/P&gt;&lt;P&gt;&lt;SPAN&gt;The two input rasters (co-pol and dual-pol) must be in the same folder, with filenames ending in XX.tif where XX is the polarization (VV, VH, HV, HH). There can only be one co-pol and one cross-pol raster in the input directory. &lt;/SPAN&gt;&lt;/P&gt;&lt;P&gt;&lt;SPAN&gt;The parameters will be set to default values once the input directory is selected. If the input directory has been extracted from an &lt;/SPAN&gt;&lt;A href="https://hyp3-docs.asf.alaska.edu/guides/rtc_product_guide/" target="_blank" STYLE="text-decoration:underline;"&gt;On Demand RTC product generated using HyP3&lt;/A&gt;&lt;SPAN&gt;, all of the parameters will be automatically populated based on the information contained in the directory name. If you are not using a product directory generated by HyP3, or if the directory name has been changed, you will need to rename the directory and RTC products to match the HyP3 naming pattern. &lt;/SPAN&gt;&lt;/P&gt;&lt;P&gt;&lt;SPAN&gt;The red/blue threshold can be adjusted if desired. The default is -24 dB, and the appropriate value range is generally -19 to -25 dB. &lt;/SPAN&gt;&lt;/P&gt;&lt;P&gt;&lt;SPAN&gt;The output directory defaults to the input directory, and the output name defaults to the name of the directory tagged with _RGB.tif. &lt;/SPAN&gt;&lt;/P&gt;&lt;P&gt;&lt;SPAN&gt;You can amend any of the auto-populated outputs or set them to something completely different if desired. Note that if you change your input directory, the default values will not automatically change. You can reset the auto-populated outputs based on the new parameters by deleting the default string and pressing tab or clicking away from the field. &lt;/SPAN&gt;&lt;/P&gt;&lt;P&gt;&lt;SPAN&gt;----- &lt;/SPAN&gt;&lt;/P&gt;&lt;P&gt;&lt;SPAN&gt;For best results when viewing the output image, set the symbology to be an RGB Composite stretched to Minimum-Maximum with no gamma stretch applied. The default symbology is generally an RGB Composite stretched to a Percent Clip with a gamma stretch applied.&lt;/SPAN&gt;&lt;/P&gt;&lt;P&gt;&lt;SPAN&gt;In products generated using this tool, vegetated areas will generally appear more green, water (or desert sand) blue, and other areas such as urban landscapes more red. Where there are stronger cross-pol returns, indicating more volume scatter, the image will be more dominated by the green band, while areas with more surface scatter will have a greater contribution from the blue and red bands. &lt;/SPAN&gt;&lt;/P&gt;&lt;P&gt;&lt;A href="https://github.com:443/ASFHyP3/hyp3-lib/blob/develop/docs/rgb_decomposition.md" target="_blank" STYLE="text-decoration:underline;"&gt;Additional information&lt;/A&gt;&lt;SPAN&gt; about interpreting RGB Decomposition images and the processing steps used to generate the color values is available.&lt;/SPAN&gt;&lt;/P&gt;&lt;/DIV&gt;&lt;/DIV&gt;&lt;/DIV&gt;</usage><scriptExamples><scriptExample><title>RGB Decomposition</title><code>class RGBDecomp(object):
    def __init__(self):

        """Generates an RGB image from co- and cross-pol RTC data"""
        self.label = "RGB Decomposition"
        self.description = "This tool generates an RGB image using co- and cross-pol RTC data."
        self.canRunInBackground = True

    def getParameterInfo(self):
        """Define parameter definitions"""

        # First parameter: input directory
        indir = arcpy.Parameter(
            name="indir",
            displayName="Input directory containing dual-pol RTC data",
            datatype="DEWorkspace",
            parameterType="Required",
            direction="Input")

        # Second parameter: scale of input dataset
        scale = arcpy.Parameter(
            name="scale",
            displayName="Scale of input RTC (Power, Amplitude, or dB)",
            datatype="GPString",
            parameterType="Required",
            direction="Input")

        scale.filter.type = "ValueList"
        scale.filter.list = ["Power", "Amplitude", "dB"]

        # Third parameter: Primary polarization
        pol = arcpy.Parameter(
            name="pol",
            displayName="Primary polarization (V or H)",
            datatype="GPString",
            parameterType="Required",
            direction="Input")

        pol.filter.type = "ValueList"
        pol.filter.list = ["V", "H"]

        # Fourth parameter: R/B threshold in dB
        rb_thresh_db = arcpy.Parameter(
            name="rb_thresh_db",
            displayName="Threshold cutoff value for red/blue in dB (default: -24.0 dB)",
            datatype="GPDouble",
            parameterType="Required",
            direction="Input")

        # Fifth parameter: output directory for RGB file
        outdir = arcpy.Parameter(
            name="outdir",
            displayName="Output directory for new RGB file",
            datatype="DEWorkspace",
            parameterType="Required",
            direction="Input")

        # Sixth parameter: output name for RGB file
        outname = arcpy.Parameter(
            name="outname",
            displayName="Filename for new RGB raster",
            datatype="GPString",
            parameterType="Required",
            direction="Input")

        # Seventh parameter: select if output is added to the map
        out_yn = arcpy.Parameter(
            name="out_yn",
            displayName="Add output to map",
            datatype="GPBoolean",
            parameterType="Required",
            direction="Input")

        out_yn.value = "true"

        # Eighth parameter: output layer to add to project
        outlayer = arcpy.Parameter(
            name="outlayer",
            displayName="Derived output for final product raster",
            datatype="GPRasterLayer",
            parameterType="Derived",
            direction="Output")

        params = [indir, scale, pol, rb_thresh_db, outdir, outname, out_yn, outlayer]
        return params

    def isLicensed(self):
        """This tool requires the Spatial Analyst Extension"""
        arcpy.AddMessage("Checking Spatial Analyst Extension status...")
        try:
            if arcpy.CheckExtension("Spatial") != "Available":
                raise Exception
            else:
                arcpy.AddMessage("Spatial Analyst Extension is available.")
                if arcpy.CheckOutExtension("Spatial") == "CheckedOut":
                    arcpy.AddMessage("Spatial Analyst Extension is checked out and ready for use.")
                elif arcpy.CheckOutExtension("Spatial") == "NotInitialized":
                    arcpy.CheckOutExtension("Spatial")
                    arcpy.AddMessage("Spatial Analyst Extension has been checked out.")
                else:
                    arcpy.AddMessage("Spatial Analyst Extension is not available for use.")
        except Exception:  # noqa: B902
            arcpy.AddMessage(
                "Spatial Analyst extension is not available for use. Check your licensing to make sure you have "
                "access to this extension.")
            return False

        return True

    def updateParameters(self, parameters):
        """Modify the values and properties of parameters before internal
        validation is performed.  This method is called whenever a parameter
        has been changed."""

        # Set the default scale for the input file to be selected based on the indir name
        if parameters[0].value:
            indirbase = os.path.basename(parameters[0].value.value)
            inscale = indirbase[36]
            if inscale == 'a':
                insc = 'Amplitude'
            elif inscale == 'p':
                insc = 'Power'
            elif inscale == 'd':
                insc = 'dB'
            else:
                insc = ''
            if not parameters[1].altered:
                parameters[1].value = insc

        # Set the default primary polarization
        if parameters[0].value:
            pol = indirbase[24]
            if pol in ("V", "H"):
                if not parameters[2].altered:
                    parameters[2].value = pol
            else:
                if not parameters[2].altered:
                    parameters[2].value = ''

        # Set the default R/B threshold value
        if not parameters[3].altered:
            parameters[3].value = -24

        # Set the default output directory to be the same as the input directory
        if parameters[0].value:
            if not parameters[4].altered:
                parameters[4].value = parameters[0].value.value

        # Set the default output filename to be the basename_RGB.tif
        if parameters[0].value:
            if not parameters[5].altered:
                parameters[5].value = ("%s_RGB.tif" % indirbase)

        return

    def updateMessages(self, parameters):
        """Modify the messages created by internal validation for each tool
        parameter.  This method is called after internal validation."""
        return

    def execute(self, parameters, messages):
        """The source code of the tool."""

        # Check licensing
        self.isLicensed()

        # Define parameters
        indir = parameters[0].valueAsText
        scale = parameters[1].valueAsText
        pol = parameters[2].valueAsText
        rb_thresh_db = parameters[3].value
        outdir = parameters[4].valueAsText
        outname = parameters[5].valueAsText
        out_yn = parameters[6].valueAsText

        arcpy.AddMessage("Input parameters have been defined. Preparing workspace...")

        # Set the working directory containing the RTC images
        arcpy.env.workspace = indir
        indirbase = os.path.basename(indir)
        # Create a scratch directory for intermediate files
        arcpy.CreateFolder_management(indir, "temp")
        scratchpath = os.path.join(indir, "temp")

        arcpy.AddMessage("Workspace has been prepared. Defining input rasters...")

        # Confirm that the input dataset is dual-pol
        dpol = indirbase[23]
        if dpol == 'D':
            arcpy.AddMessage("Based on the input directory, this is a dual-pol dataset.")
        elif dpol == 'S':
            arcpy.AddError("This is a single-pol dataset. A dual-pol dataset is required for RGB Decomposition.")
            sys.exit(0)
        else:
            arcpy.AddMessage("Dual-polarization cannot be determined from the input directory.")

        # Set variables for co-pol and cross-pol GeoTIFFs
        if pol == 'V':
            vv_tif = arcpy.ListRasters('*VV.tif')[0]
            cp = os.path.join(indir, vv_tif)
            arcpy.AddMessage("Co-pol dataset: %s" % vv_tif)
            vh_tif = arcpy.ListRasters('*VH.tif')[0]
            xp = os.path.join(indir, vh_tif)
            arcpy.AddMessage("Cross-pol dataset: %s" % vh_tif)

        elif pol == 'H':
            hh_tif = arcpy.ListRasters('*HH.tif')[0]
            cp = os.path.join(indir, hh_tif)
            arcpy.AddMessage("Co-pol dataset: %s" % hh_tif)
            hv_tif = arcpy.ListRasters('*HV.tif')[0]
            xp = os.path.join(indir, hv_tif)
            arcpy.AddMessage("Cross-pol dataset: %s" % hv_tif)

        # Convert the scale to power if necessary
        if scale == 'Amplitude':
            cps = arcpy.sa.Square(cp)
            xps = arcpy.sa.Square(xp)
            arcpy.AddMessage("Input scale is amplitude. Converted to power scale.")

        elif scale == 'dB':
            cps = arcpy.sa.Power(10, arcpy.sa.Divide(cp, 10))
            xps = arcpy.sa.Power(10, arcpy.sa.Divide(xp, 10))
            arcpy.AddMessage("Input scale is dB. Converted to power scale.")

        elif scale == 'Power':
            cps = cp
            xps = xp
            arcpy.AddMessage("Input scale is power. No conversion necessary.")

        else:
            arcpy.AddError("Input scale must be set to Power Amplitude, or dB.")
            sys.exit(0)

        arcpy.AddMessage("Input rasters have been defined. Running pixel cleanup routine...")

        # Perform pixel cleanup on VV and VH RTC images, using -48 dB as cutoff for valid pixels
        pc_thresh = math.pow(10, -4.8)
        wc_pc = "VALUE &lt; %s" % (pc_thresh)
        cp0 = arcpy.sa.Con(cps, 0, cps, wc_pc)
        xp0 = arcpy.sa.Con(xps, 0, xps, wc_pc)

        arcpy.AddMessage("Pixel cleanup complete. Generating spatial masks...")

        # Generate spatial masks based on red/blue threshold
        rb_thresh = math.pow(10, rb_thresh_db / 10)

        # mb = xp0 &lt; k
        remap_mb = "0 %s 1;%s 100000 0" % (rb_thresh, rb_thresh)
        mb = arcpy.sa.Reclassify(xp0, "VALUE", remap_mb, "DATA")

        # mr = xp0 > k
        remap_mr = "0 %s 0;%s 100000 1" % (rb_thresh, rb_thresh)
        mr = arcpy.sa.Reclassify(xp0, "VALUE", remap_mr, "DATA")

        # mx = SXP > 0
        mx = arcpy.sa.Con(xp0, "1", "0", "VALUE > 0")

        arcpy.AddMessage("Spatial masks generated. Deriving red and blue components of surface scatter...")

        # The surface scattering component is divided into red and blue sections
        # Negative values are set to zero
        pr = arcpy.sa.Con((cp0 - (3 * xp0)), "0", (cp0 - (3 * xp0)), "VALUE &lt; 0")
        pb = arcpy.sa.Con(((3 * xp0) - cp0), "0", ((3 * xp0) - cp0), "VALUE &lt; 0")

        # Calculate the difference between the co- and cross-pol values
        # Negative values are set to zero
        sd = arcpy.sa.Con((cp0 - xp0), "0", (cp0 - xp0), "VALUE &lt; 0")

        arcpy.AddMessage(
            "Red and blue components have been derived. Applying spatial masks and scalars for each band...")

        # Apply spatial masks and specific scalars to stretch the values for each band from 1 to 255
        z = 2 / math.pi * mb * arcpy.sa.ATan(arcpy.sa.SquareRoot(sd))
        ir = 254 * mx * (2 * mr * arcpy.sa.SquareRoot(pr) + z) + 1
        ig = 254 * mx * (3 * mr * arcpy.sa.SquareRoot(xp0) + (2 * z)) + 1
        ib = 254 * mx * (2 * arcpy.sa.SquareRoot(pb) + (5 * z)) + 1

        arcpy.AddMessage(
            "Spatial masks and scalars have been applied. Converting bands to 8-bit unsigned integer GeoTIFFs...")

        # Create empty list for RGB bands
        band_list = []

        # Remove negative values and convert each band to an integer raster
        ar = arcpy.sa.Int(arcpy.sa.Con(ir, "255", ir, "VALUE > 255"))
        ag = arcpy.sa.Int(arcpy.sa.Con(ig, "255", ig, "VALUE > 255"))
        ab = arcpy.sa.Int(arcpy.sa.Con(ib, "255", ib, "VALUE > 255"))

        # Save bands as GeoTIFF rasters in the scratch folder
        arpath = os.path.join(scratchpath, "ar.tif")
        ar.save(arpath)
        band_list.append(arpath)

        agpath = os.path.join(scratchpath, "ag.tif")
        ag.save(agpath)
        band_list.append(agpath)

        abpath = os.path.join(scratchpath, "ab.tif")
        ab.save(abpath)
        band_list.append(abpath)

        arcpy.AddMessage(
            "GeoTIFF files for each band have been saved. Combining single-band rasters to generate RGB image...")

        # Combine the aRGB bands into a composite raster
        outpath = os.path.join(outdir, outname)
        arcpy.CompositeBands_management(band_list, outpath)
        arcpy.AddMessage("RGB Decomposition product has been generated: %s." % outpath)

        # Indicate process is complete
        arcpy.AddMessage("RGB Decomposition process is complete.")

        # Add the output product to the map
        if out_yn == "true":
            dispname = os.path.splitext(outname)[0]
            arcpy.MakeRasterLayer_management(outpath, dispname)
            arcpy.SetParameterAsText(7, dispname)
            arcpy.AddMessage("Added RGB raster layer to map display.")
        else:
            arcpy.AddMessage(
                "Option to add output layer to map was not selected. "
                "Output can be added manually if desired: %s" % outpath)

        """
        # Delete temporary files
        arcpy.AddMessage("Cleaning up...")

        # The deletion of the temp folder could be set as an option, if there's any chance that users might
        # want to have access to the individual color bands. I don't think that's likely, though.

        # There is not currently a mechanism for getting rid of the temp folder.
        # The arcpy.DeleteManagement approach doesn't work from within a tool, so that's not an option.
        # There seems to be a residual hold on at least one of the saved input bands for the RGB Composite,
        # so using shutil doesn't work (it won't delete a directory if a file contained in the directory
        # still has a file lock applied). The lock is probably ArcGIS-driven.
        # Not sure how best to proceed to get rid of the temp folder once the process is complete.

        arcpy.AddMessage("Temporary files have been deleted.")
        """

        # Check In Spatial Analyst Extension
        status = arcpy.CheckInExtension("Spatial")
        arcpy.AddMessage("The Spatial Analyst Extension is in %s status." % status)

        return</code><para>&lt;DIV STYLE="text-align:Left;"&gt;&lt;DIV&gt;&lt;DIV&gt;&lt;P&gt;&lt;SPAN&gt;This is the code for the RGBDecomp tool class within the ASF_Tools.pyt script&lt;/SPAN&gt;&lt;/P&gt;&lt;/DIV&gt;&lt;/DIV&gt;&lt;/DIV&gt;</para></scriptExample></scriptExamples></tool><dataIdInfo><idCitation><resTitle>RGB Decomposition</resTitle></idCitation><idAbs>&lt;DIV STYLE="text-align:Left;"&gt;&lt;DIV&gt;&lt;DIV&gt;&lt;P&gt;&lt;SPAN&gt;This tool generates an RGB Decomposition raster from a dual-pol RTC dataset. It was designed for use with RTC products generated with HyP3, including the &lt;/SPAN&gt;&lt;A href="https://hyp3-docs.asf.alaska.edu/guides/rtc_product_guide/" target="_blank" STYLE="text-decoration:underline;"&gt;On Demand RTC products you can request in Vertex&lt;/A&gt;&lt;SPAN&gt;, but can be used with any dual-pol RTC dataset that is in power, amplitude, or dB scale. &lt;/SPAN&gt;&lt;/P&gt;&lt;P&gt;&lt;SPAN&gt;-----&lt;/SPAN&gt;&lt;/P&gt;&lt;P&gt;&lt;SPAN&gt;The parameters are populated with default values once you select the input directory containing the RTC datasets. You can amend any of the auto-populated outputs or set them to something completely different if desired. Note that if you change your input directory, the default values will not automatically change. You can reset the auto-populated outputs based on the new parameters by deleting the default string and pressing tab or clicking away from the field.&lt;/SPAN&gt;&lt;/P&gt;&lt;P&gt;&lt;SPAN&gt;-----&lt;/SPAN&gt;&lt;/P&gt;&lt;P&gt;&lt;SPAN&gt;For best results when viewing the output image, set the symbology to be an RGB Composite stretched to Minimum-Maximum with no gamma stretch applied. The default symbology is generally an RGB Composite stretched to a Percent Clip with a gamma stretch applied.&lt;/SPAN&gt;&lt;/P&gt;&lt;/DIV&gt;&lt;/DIV&gt;&lt;/DIV&gt;</idAbs><searchKeys><keyword>ASF</keyword><keyword>Alaska Satellite Facility</keyword><keyword>SAR</keyword><keyword>Synthetic Aperture Radar</keyword><keyword>RGB Decomposition</keyword></searchKeys><idCredit>Alaska Satellite Facility</idCredit><dataLang><languageCode value="eng"/></dataLang><dataChar><CharSetCd value="004"/></dataChar><idPoC xmlns=""><rpIndName>Heidi Kristenson</rpIndName><rpOrgName>Alaska Satellite Facility</rpOrgName><rpPosName>GIS Specialist</rpPosName><role><RoleCd value="011"/></role></idPoC></dataIdInfo><distInfo><distributor><distorFormat><formatName>ArcToolbox Tool</formatName></distorFormat></distributor></distInfo><mdHrLv><ScopeCd value="005"/></mdHrLv><mdContact xmlns=""><rpIndName>Heidi Kristenson</rpIndName><rpOrgName>Alaska Satellite Facility</rpOrgName><rpPosName>GIS Specialist</rpPosName><role><RoleCd value="011"/></role><editorSave>False</editorSave><displayName>Heidi Kristenson</displayName></mdContact><Binary><Thumbnail><Data EsriPropertyType="PictureX">/9j/4AAQSkZJRgABAQEAHgAeAAD/2wBDAAMCAgMCAgMDAwMEAwMEBQgFBQQEBQoHBwYIDAoMDAsK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</Data></Thumbnail></Binary><mdDateSt Sync="TRUE">20201002</mdDateSt></metadata>