added initial version of Example 8 (pansharpening)

EX08_pansharpenMeasurement.py

@@ -0,0 +1,135 @@
+import os
+import platform
+from pathlib import Path
+import time
+
+import cuvis
+import tifffile
+
+def run_example_pansharpening(userSettingsDir,
+                              measurementLoc,
+                              panToCube,
+                              panScale,
+                              exportDir):
+
+    print("loading user settings...")
+    cuvis.init(userSettingsDir)
+    cuvis.set_log_level("info")
+
+    print("loading session file...")
+    session = cuvis.SessionFile(measurementLoc)
+    mesu = session[0]
+    assert mesu._handle
+
+    if 'pan' not in mesu.data.keys():
+        raise ValueError('The measurement does not contain a Pan Image.')
+
+    # settings pansharpening
+    '''
+    cuvis.PansharpeningAlgorithm:
+    Defines, which algorithm is used to calculate the pansharpened image. Available options are:
+    - Noop:
+        "dummy" algorithm, only applies scaling.
+    - CubertMacroPixel:
+        Generates intensity scaling values for the spectral data from the panimage. Can operate without
+        reference measurements.
+    - CubertPanRatio:
+        Preferred algorithm for reflectance measurements. The measurement's and white calibration's panimages
+        are used to generate the panratio (or „panchromatic reflectance“ image). The panratio is then applied
+        to the hyperspectral cube to scale the intensity of the spectra without changing their shape. 
+    - AlphaBlendOverlay:
+        Technically not a pan-sharpening algorithm. Quick workaround to display correlation results on top of
+        the pan image using an opacity value selected by the user.
+    '''
+    panAlgo = cuvis.PanSharpeningAlgorithm.CubertPanRatio # algorithm for pansharpening
+
+    '''
+    cuvis.PanSharpeningInterpolationType:
+    This setting controls which conventional algorithm is utilized for upscaling the hyperspectral data cube.
+    Available options are: Nearest Neighbor, Linear, Cubic and Lanczos
+    '''
+    panInterp = cuvis.PanSharpeningInterpolationType.Cubic # interpolation method
+
+    multi_tiff_settings = cuvis.TiffExportSettings(
+        export_dir = str(exportDir),
+        format = cuvis.TiffFormat.MultiChannel,
+        pan_sharpening_algorithm = panAlgo,
+        pan_sharpening_interpolation_type = panInterp, 
+        pan_scale = panScale,
+        add_pan = panToCube 
+        )
+
+    print("size before pansharpening: ", mesu.cube.width, "x", mesu.cube.height, "x", mesu.cube.channels)
+    print("Pansharpening and tiff export in progress...")
+
+    multiTiffExporter = cuvis.TiffExporter(multi_tiff_settings)
+    multiTiffExporter.apply(mesu) 
+
+    time.sleep(5)
+
+    exportPath= Path(exportDir) / "x20p_flight_data_0000_raw.tiff"
+    pansharpened_mesu = tifffile.imread(str(exportPath))
+
+    print("size after pansharpening: ", pansharpened_mesu.shape[0], "x", pansharpened_mesu.shape[1], "x", pansharpened_mesu.shape[2] )
+    print("algorithm used:", str(panAlgo).split('.')[-1])
+    print("Interpolation method used:", str(panInterp).split('.')[-1])
+    print("Pansharpening amount:", panScale)
+    print("Panimage saved as 0th channel:", panToCube)
+
+    cuvis.shutdown()
+    print("finished.")
+
+
+if __name__ == "__main__":
+
+    if platform.system() == "Windows":
+        data_dir = Path(os.getenv("CUVIS")).parent / "sdk" / \
+            "sample_data" / "set_examples"
+
+    elif platform.system() == "Linux":
+        data_dir = Path(os.getenv("CUVIS_DATA")) / \
+            "sample_data" / "set_examples"
+
+    loc_file = data_dir / "set2_x20p_flight_data" / "x20p_flight_data.cu3s"
+    loc_settings = data_dir / "settings"
+
+    pan_to_cube = True
+    pan_size = 1
+
+    loc_output = Path(os.getcwd()) / "EX08_pansharpened_measurement"
+
+    print("Example 08: Pansharpen X20 Plus measurement. Please provide:")
+
+    userSettingsDir = input(
+        "User settings directory (default: {}): ".format(loc_settings))
+    if userSettingsDir.strip().lower() in ["", "default"]:
+        userSettingsDir = loc_settings
+
+    measurementLoc = input(
+        "Measurement file (.cu3s) (default: {}): ".format(loc_file))
+    if measurementLoc.strip().lower() in ["", "default"]:
+        measurementLoc = loc_file
+
+    panToCube = input(
+        "Save panimage as 0th channel - True/False? (default: {}): ".format(pan_to_cube))
+    if panToCube.lower() in ["", "default"]:
+        panToCube = pan_to_cube  # use the default boolean value
+    elif panToCube.lower() == "true":
+        panToCube = True
+    elif panToCube.lower() == "false":
+        panToCube = False
+    else:
+        raise ValueError("Invalid input. Please enter True or False.")
+
+    panScale = input(
+        "Define the amount of pansharpening between 0 and 1 in respect to the size of the panimage (default: {}): ".format(pan_size))
+    if panScale.strip().lower() in ["", "default"]:
+        panScale = pan_size
+
+    exportDir = input(
+        "Name of export directory (default: {}): ".format(loc_output))
+    if exportDir.strip().lower() in ["", "default"]:
+        exportDir = loc_output
+
+
+    run_example_pansharpening(str(userSettingsDir), str(measurementLoc), panToCube, panScale, exportDir)