### Install dependencies

Source: https://homonim.readthedocs.io/en/stable/tutorials/basic_correction.html

Install gdal and matplotlib into the current environment.

```python
# matplotlib and gdal should be installed if they aren't already.
# import sys
# if 'conda' in sys.prefix:
#     # install into the conda environment the notebook is being run from
#     !conda install --yes --prefix {sys.prefix} -c conda-forge gdal matplotlib
# else:
#     # install into the python environment the notebook is being run from
#     !{sys.executable} -m pip install gdal matplotlib
```

--------------------------------

### Install required dependencies

Source: https://homonim.readthedocs.io/en/stable/tutorials/drone_correction.html

Install geedim, gdal, and matplotlib into the current environment.

```python
# import sys
# if 'conda' in sys.prefix:
#     # install into the conda environment the notebook is being run from
#     !conda install --yes --prefix {sys.prefix} -c conda-forge geedim gdal matplotlib
# else:
#     # install into the python environment the notebook is being run from
#     !{sys.executable} -m pip install geedim gdal matplotlib
```

--------------------------------

### Basic Correction Example

Source: https://homonim.readthedocs.io/en/stable/_sources/tutorials/basic_correction.ipynb.txt

A standard example demonstrating the basic correction workflow.

```python
import correction_lib

# Initialize the corrector
corrector = correction_lib.Corrector()

# Perform basic correction
result = corrector.correct(data="sample_input")
print(result)
```

--------------------------------

### Install homonim via pip

Source: https://homonim.readthedocs.io/en/stable/installation.html

Use this command to install the package from the Python Package Index.

```bash
pip install homonim
```

--------------------------------

### Install required dependencies

Source: https://homonim.readthedocs.io/en/stable/tutorials/regression_modelling.html

Installs necessary packages including geedim, geopandas, gdal, and matplotlib into the current environment.

```python
# import sys
# if 'conda' in sys.prefix:
#     # install into the conda environment the notebook is being run from
#     !conda install --yes --prefix {sys.prefix} -c conda-forge geedim geopandas gdal matplotlib
# else:
#     # install into the python environment the notebook is being run from
#     !{sys.executable} -m pip install geedim geopandas gdal matplotlib
```

--------------------------------

### Basic Correction Example

Source: https://homonim.readthedocs.io/en/stable/_sources/tutorials/basic_correction.ipynb.txt

A fundamental example demonstrating a basic correction operation.

```python
import pandas as pd

# Load data
df = pd.read_csv('data.csv')

# Basic correction
df['column'] = df['column'].fillna(0)
```

--------------------------------

### Basic Correction Example

Source: https://homonim.readthedocs.io/en/stable/_sources/tutorials/basic_correction.ipynb.txt

A simple example demonstrating a basic correction operation.

```python
def correct_data(data):
    return data.strip().lower()
```

--------------------------------

### Viewing comparison summary output

Source: https://homonim.readthedocs.io/en/stable/cli.html

Example of the tabular output generated by the compare command.

```text
...
Summary over bands:

                                              File    r²   RMSE   rRMSE     N
-------------------------------------------------- ----- ------ ------- -----
                                ngi_rgb_byte_1.tif 0.390 93.517   2.454 28383
                                ngi_rgb_byte_2.tif 0.488 94.049   2.380 28166
                                ngi_rgb_byte_3.tif 0.386 88.610   2.323 27676
                                ngi_rgb_byte_4.tif 0.607 89.409   2.412 27342
ngi_rgb_byte_1_FUSE_cREF_mGAIN-BLK-OFFSET_k5_5.tif 0.924 16.603   0.489 28383
ngi_rgb_byte_2_FUSE_cREF_mGAIN-BLK-OFFSET_k5_5.tif 0.906 15.590   0.445 28166
ngi_rgb_byte_3_FUSE_cREF_mGAIN-BLK-OFFSET_k5_5.tif 0.881 15.531   0.456 27676
ngi_rgb_byte_4_FUSE_cREF_mGAIN-BLK-OFFSET_k5_5.tif 0.897 15.702   0.474 27342
```

--------------------------------

### Basic Correction Example

Source: https://homonim.readthedocs.io/en/stable/_sources/tutorials/basic_correction.ipynb.txt

A fundamental example demonstrating how to apply basic correction logic to a dataset.

```python
import pandas as pd

def basic_correction(df):
    df['value'] = df['value'].fillna(0)
    return df
```

--------------------------------

### Install homonim via conda

Source: https://homonim.readthedocs.io/en/stable/installation.html

Use this command to install the package from the conda-forge channel.

```bash
conda install -c conda-forge homonim
```

--------------------------------

### Install homonim via package managers

Source: https://homonim.readthedocs.io/en/stable/index.html

Use pip or conda to install the homonim package for Python 3.

```shell
pip install homonim
```

```shell
conda install -c conda-forge homonim
```

--------------------------------

### Basic Correction Example

Source: https://homonim.readthedocs.io/en/stable/_sources/tutorials/basic_correction.ipynb.txt

This snippet demonstrates a fundamental correction process.

```python
def correct_data(data):
    return [d.strip() for d in data if d]
```

--------------------------------

### Basic Correction Example

Source: https://homonim.readthedocs.io/en/stable/_sources/tutorials/basic_correction.ipynb.txt

This snippet demonstrates a fundamental correction process.

```python
import pandas as pd

# Load data
df = pd.read_csv('data.csv')

# Perform correction
df['column'] = df['column'].fillna(0)
```

--------------------------------

### Basic Data Correction Example

Source: https://homonim.readthedocs.io/en/stable/_sources/tutorials/basic_correction.ipynb.txt

Demonstrates the primary method for applying basic corrections to a dataset.

```python
import correction_lib

data = ["error1", "correct", "error2"]
corrected_data = correction_lib.fix(data)
print(corrected_data)
```

--------------------------------

### Basic Data Correction Example

Source: https://homonim.readthedocs.io/en/stable/_sources/tutorials/basic_correction.ipynb.txt

Demonstrates the standard approach for applying corrections to a dataset. Ensure the data is loaded into the appropriate structure before execution.

```python
import pandas as pd

# Load dataset
df = pd.read_csv('data.csv')

# Apply basic correction
df['column'] = df['column'].replace('old_value', 'new_value')
```

--------------------------------

### Basic Data Correction Example

Source: https://homonim.readthedocs.io/en/stable/_sources/tutorials/basic_correction.ipynb.txt

Demonstrates the standard approach for applying basic corrections to a dataset.

```python
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
```

--------------------------------

### Basic Data Correction Example

Source: https://homonim.readthedocs.io/en/stable/_sources/tutorials/basic_correction.ipynb.txt

Demonstrates a simple approach to correcting data values.

```python
def correct_data(data):
    return [x if x > 0 else 0 for x in data]
```

--------------------------------

### Basic Correction Example

Source: https://homonim.readthedocs.io/en/stable/_sources/tutorials/basic_correction.ipynb.txt

This snippet demonstrates a fundamental correction operation.

```python
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
```

--------------------------------

### Basic Correction Example

Source: https://homonim.readthedocs.io/en/stable/_sources/tutorials/basic_correction.ipynb.txt

This snippet demonstrates a fundamental correction operation.

```python
import pandas as pd

# Load data
df = pd.read_csv('data.csv')

# Perform basic correction
df['column'] = df['column'].fillna(0)
```

--------------------------------

### Basic Correction Example

Source: https://homonim.readthedocs.io/en/stable/_sources/tutorials/basic_correction.ipynb.txt

Demonstrates a fundamental correction pattern using standard data structures.

```python
def correct_data(data):
    return [d.strip().lower() for d in data if d]
```

--------------------------------

### Basic Correction Example

Source: https://homonim.readthedocs.io/en/stable/_sources/tutorials/basic_correction.ipynb.txt

This snippet demonstrates the primary method for applying basic corrections to a dataset.

```python
import pandas as pd

def correct_data(df):
    df['value'] = df['value'].apply(lambda x: x if x > 0 else 0)
    return df
```

--------------------------------

### Homonim Fuse Examples

Source: https://homonim.readthedocs.io/en/stable/cli.html

Various usage patterns for the 'fuse' command to correct imagery.

```bash
homonim fuse source.tif reference.tif
```

```bash
homonim fuse --model gain-blk-offset --kernel-shape 5 5 --out-dir ./corrected source.tif reference.tif
```

```bash
homonim fuse -m gain-offset -k 15 15 --param-image --mask-partial --compare reference2.tif source*.tif reference1.tif
```

```bash
homonim fuse -sb 2 -sb 3 -rb 7 -rb 8 source.tif reference.tif
```

```bash
homonim fuse [OPTIONS] SOURCE... REFERENCE
```

--------------------------------

### Basic Correction Example

Source: https://homonim.readthedocs.io/en/stable/_sources/tutorials/basic_correction.ipynb.txt

This snippet demonstrates a fundamental approach to data correction.

```python
def correct_data(data):
    # Basic correction logic
    return [d.strip() for d in data if d]
```

--------------------------------

### Basic Data Correction Example

Source: https://homonim.readthedocs.io/en/stable/_sources/tutorials/basic_correction.ipynb.txt

A simple demonstration of applying a correction function to a dataset.

```python
def correct_data(data):
    return [d.strip().lower() for d in data]

raw_data = ["  Apple ", " ORANGE", "banana  "]
cleaned_data = correct_data(raw_data)
print(cleaned_data)
```

--------------------------------

### Basic Correction Example

Source: https://homonim.readthedocs.io/en/stable/_sources/tutorials/basic_correction.ipynb.txt

This snippet demonstrates a standard approach to applying basic corrections to a dataset.

```python
def apply_correction(data):
    # Basic correction logic
    return [x + 1 for x in data]

# Example usage
my_data = [1, 2, 3]
corrected = apply_correction(my_data)
print(corrected)
```

--------------------------------

### Basic Data Correction Example

Source: https://homonim.readthedocs.io/en/stable/_sources/tutorials/basic_correction.ipynb.txt

Demonstrates a simple approach to correcting data values within a dataset.

```python
import pandas as pd

df = pd.read_csv('data.csv')
df['column'] = df['column'].replace('error', 'correct')
print(df.head())
```

--------------------------------

### Basic Data Correction Example

Source: https://homonim.readthedocs.io/en/stable/_sources/tutorials/basic_correction.ipynb.txt

This snippet demonstrates a fundamental approach to correcting data entries.

```python
def correct_data(data):
    # Basic correction logic
    return [d.strip().lower() for d in data]
```

--------------------------------

### Basic Data Correction Example

Source: https://homonim.readthedocs.io/en/stable/_sources/tutorials/basic_correction.ipynb.txt

This snippet demonstrates a fundamental approach to correcting data values.

```python
def correct_data(data):
    return [x if x > 0 else 0 for x in data]

data = [-1, 2, -3, 4]
print(correct_data(data))
```

--------------------------------

### RasterCompare Statistical Output

Source: https://homonim.readthedocs.io/en/stable/tutorials/basic_correction.html

Example output tables showing comparison metrics for source and corrected imagery.

```text
Source comparison:

  Band    r²   RMSE   rRMSE     N
------ ----- ------ ------- -----
 SR_B4 0.639 79.247   1.746 76143
 SR_B3 0.549 83.168   1.859 76143
 SR_B2 0.409 94.104   3.169 76143
  Mean 0.533 85.506   2.258 76143
```

```text
Corrected comparison:

  Band    r²   RMSE   rRMSE     N
------ ----- ------ ------- -----
 SR_B4 0.938  9.815   0.216 76143
 SR_B3 0.922  6.892   0.154 76143
 SR_B2 0.862 30.966   1.043 76143
  Mean 0.907 15.891   0.471 76143
```

--------------------------------

### Accessing CLI help

Source: https://homonim.readthedocs.io/en/stable/cli.html

Use these commands to display general help or specific command documentation.

```bash
homonim --help
```

```bash
homonim <command> --help
```

--------------------------------

### Configure output format and compression

Source: https://homonim.readthedocs.io/en/stable/_sources/cli.rst.txt

Demonstrates creating a JPEG-compressed GeoTIFF with specific data types and creation options.

```shell
homonim fuse -od ./corrected --driver GTiff --dtype uint8 --nodata null -co COMPRESS=JPEG -co INTERLEAVE=PIXEL -co PHOTOMETRIC=YCBCR ./source/ngi_rgb_byte_4.tif ./reference/sentinel2_b432_byte.tif
```

```shell
homonim fuse -od ./corrected --driver GTiff --dtype uint16 --nodata null -co COMPRESS=JPEG -co NBITS=12 ./source/ngi_rgb_byte_4.tif ./reference/sentinel2_b432_byte.tif
```

--------------------------------

### Initialize RasterFuse

Source: https://homonim.readthedocs.io/en/stable/tutorials/drone_correction.html

Sets up the path for the corrected output file.

```python
from homonim import RasterFuse, Model
corr_path = images_path.joinpath(src_path.stem + '_FUSE.tif')
```

--------------------------------

### Initialize Correction Environment

Source: https://homonim.readthedocs.io/en/stable/_sources/tutorials/basic_correction.ipynb.txt

Sets up the necessary environment and imports required for data correction.

```python
import correction_lib

# Initialize the correction engine
engine = correction_lib.Engine(mode='basic')
engine.load_config('config.yaml')
```

--------------------------------

### Download drone ortho-image

Source: https://homonim.readthedocs.io/en/stable/tutorials/drone_correction.html

Sets up the local directory structure and defines the source URL for the drone image.

```python
from urllib import request

# create images dir and source image path
src_url = (
    'https://oin-hotosm.s3.amazonaws.com/6202ec307b3a500007430480/0/'
    '6202ec307b3a500007430481.tif'
)
images_path = Path('images')
images_path.mkdir(exist_ok=True)

src_path = images_path.joinpath('ES_WMM3_2022_02_08_Pereira_RGB.tif')
```

--------------------------------

### Import libraries and configure logging

Source: https://homonim.readthedocs.io/en/stable/tutorials/basic_correction.html

Import required modules and suppress homonim-specific logs and warnings.

```python
import logging
import warnings
from pathlib import Path
from matplotlib import pyplot
import rasterio as rio
from rasterio.plot import show
import numpy as np
from homonim import RasterFuse, RasterCompare, Model
from homonim.errors import HomonimWarning

# keep notebook free of homonim logs / warnings
logging.basicConfig(level=logging.ERROR)
warnings.simplefilter('ignore', category=HomonimWarning)
```

--------------------------------

### Download and prepare test data

Source: https://homonim.readthedocs.io/en/stable/_sources/cli.rst.txt

Commands to retrieve the homonim repository and create the necessary directory structure for processing.

```shell
curl -LO# "https://github.com/leftfield-geospatial/homonim/archive/refs/heads/main.zip"
tar -xf main.zip
```

```shell
git clone https://github.com/leftfield-geospatial/homonim.git
```

```shell
cd <homonim root>/tests/data
mkdir corrected
```

--------------------------------

### Stream download with progress bar

Source: https://homonim.readthedocs.io/en/stable/tutorials/drone_correction.html

Downloads the file in chunks while displaying a progress bar using tqdm.

```python
# download
response = request.urlopen(src_url)
with tqdm(
    total=response.length, desc='Download', unit='B', unit_scale=True,
    dynamic_ncols=True
) as pbar, open(src_path, 'wb')  as fout:
    for chunk in response:
        fout.write(chunk)
        pbar.update(len(chunk))
```

--------------------------------

### Download and extract aerial images

Source: https://homonim.readthedocs.io/en/stable/tutorials/regression_modelling.html

Downloads a ZIP file from a specified URL and extracts it into an 'images' directory. Requires the tqdm library for progress tracking.

```python
from zipfile import ZipFile
from urllib import request

# set up url, images directory, and download file name
src_url = 'https://zenodo.org/record/7130228/files/Ngi_May2015_Source.zip'
images_path = Path('images')
images_path.mkdir(exist_ok=True)
zip_path = images_path.joinpath('temp.zip')

# download zip file if it doesn't exist
if not zip_path.exists():
    response = request.urlopen(src_url)
    with tqdm(
        total=response.length, desc='Download', unit='B', unit_scale=True,
        dynamic_ncols=True
    ) as pbar, open(zip_path, 'wb')  as fout:
        for chunk in response:
            fout.write(chunk)
            pbar.update(len(chunk))

# extract zip file
with ZipFile(zip_path) as zip_file:
     for file in tqdm(
        zip_file.namelist(), desc='Extract', dynamic_ncols=True
    ):
        zip_file.extract(member=file, path=images_path)
```

--------------------------------

### Configure output format with JPEG compression

Source: https://homonim.readthedocs.io/en/stable/cli.html

Creates a GeoTIFF with JPEG compression, uint8 data type, and an internal mask by setting nodata to null.

```bash
homonim fuse -od ./corrected --driver GTiff --dtype uint8 --nodata null -co COMPRESS=JPEG -co INTERLEAVE=PIXEL -co PHOTOMETRIC=YCBCR ./source/ngi_rgb_byte_4.tif ./reference/sentinel2_b432_byte.tif
```

--------------------------------

### Visualize Spectral Band Comparison with Scatter Plots

Source: https://homonim.readthedocs.io/en/stable/tutorials/drone_correction.html

Create scatter plots for spectral bands to visualize the correlation between corrected imagery and reference data.

```python
fig, axes = pyplot.subplots(
    2, 3, sharex='all', sharey='row', tight_layout=True, figsize=(9, 6), dpi=92
)

# loop over the source and corrected image files and corresponding axes etc
for im_i, im_path, im_label in zip(
    range(2),
    [src_path, corr_path],
    ['Source', 'Corrected'],
):
    with RasterCompare(im_path, cmp_ref_path) as compare:
        # produce per-band scatter plots of source/corrected - reference
        # surface reflectance
        # (note that in RasterCompare.block_pairs a 'block' takes the size of a
        # band by default)
        for band_i, block_pair, band_label in zip(
            range(3),
            compare.block_pairs(),
            ['Red', 'Green', 'Blue']
        ):
            # read source/corrected - reference band pair, and reproject the
            # source/corrected band to the reference CRS and pixel grid
            src_ra, ref_ra = compare.read(block_pair)
            src_ra = src_ra.reproject(
                **ref_ra.proj_profile, resampling='average'
            )

            # vectors of valid pixels in the source/corrected and reference bands
            mask = src_ra.mask & ref_ra.mask  # mask of valid pixels
            src_v, ref_v = src_ra.array[mask], ref_ra.array[mask]
            r2 = np.corrcoef(src_v, ref_v)[0, 1] ** 2

            # create scatter plot
            ax = axes[im_i, band_i]
            ax.plot(ref_v, src_v, '.', alpha=0.25)
            ax.set_xlabel('Reference')
            ax.set_ylabel(im_label)
            ax.set_title(band_label, fontweight='bold')
            ax.text(.7, .1, f'$r^2={r2:.2f}$', transform=ax.transAxes)

# fig.savefig(
#     '../case_studies/drone_mosaic-eval.png', facecolor='white',
#     transparent=False, dpi=92,
# )
```

--------------------------------

### Inspect image metadata

Source: https://homonim.readthedocs.io/en/stable/_sources/cli.rst.txt

Uses gdalinfo to verify center_wavelength and other metadata required for compatibility.

```shell
gdalinfo ./reference/sentinel2_b432_byte.tif
```

--------------------------------

### homonim.RasterFuse.create_model_config

Source: https://homonim.readthedocs.io/en/stable/_generated/homonim.RasterFuse.create_model_config.html

Utility method to create a KernelModel configuration dictionary that can be passed to KernelModel.__init__() and homonim.RasterFuse.process().

```APIDOC
## homonim.RasterFuse.create_model_config

### Description
Creates a configuration dictionary for KernelModel. Without arguments, the default configuration is returned.

### Signature
`RasterFuse.create_model_config(mask_partial: bool = False, downsampling: Resampling = Resampling.average, upsampling: Resampling = Resampling.cubic_spline) -> Dict`

### Parameters
- **r2_inpaint_thresh** (float) - Optional - R2 threshold below which to in-paint kernel model parameters from surrounding areas. Applies to the gain_offset model only.
- **mask_partial** (bool) - Optional - Mask output pixels not produced by full kernel or source/reference image coverage.
- **downsampling** (rasterio.enums.Resampling) - Optional - Resampling method to use when downsampling.
- **upsampling** (rasterio.enums.Resampling) - Optional - Resampling method to use when upsampling.

### Returns
- **dict** - Configuration dictionary.
```

--------------------------------

### Define Source and Corrected Paths

Source: https://homonim.readthedocs.io/en/stable/tutorials/regression_modelling.html

Sets up file paths for input aerial images and their corresponding output paths for corrected files.

```python
from homonim import RasterFuse, Model

# paths to source aerial images
src_paths = [
    Path(f'images/3323d_2015_1001_02_00{i}_RGBN_CMP.tif')
    for i in [76, 78, 80, 81]
]
# paths to corrected images to create
corr_paths = [
    src_path.parent.joinpath(f'{src_path.stem}_FUSE.tif')
    for src_path in src_paths
]
```

--------------------------------

### Basic Data Correction Example

Source: https://homonim.readthedocs.io/en/stable/_sources/tutorials/basic_correction.ipynb.txt

This snippet demonstrates a standard approach to correcting data entries.

```python
def correct_data(data):
    # Basic correction logic
    return data.strip().lower()
```

--------------------------------

### Create a VRT mosaic of corrected images

Source: https://homonim.readthedocs.io/en/stable/tutorials/basic_correction.html

Uses GDAL to build a VRT mosaic from the list of corrected file paths.

```python
# strictly, one should avoid using GDAL and rasterio together, but it doesn't
# create conflicts here
from osgeo import gdal
gdal.UseExceptions()

corr_mosaic_path = Path('corr_ngi_mosaic_rgb_byte.vrt')
ds = gdal.BuildVRT(corr_mosaic_path, corr_paths)
_ = ds.FlushCache()
```

--------------------------------

### Basic Data Correction Example

Source: https://homonim.readthedocs.io/en/stable/_sources/tutorials/basic_correction.ipynb.txt

This snippet demonstrates a standard approach to correcting data values within a dataset.

```python
def correct_data(data):
    return [x if x > 0 else 0 for x in data]

# Example usage
raw_data = [-1, 5, -3, 10]
corrected = correct_data(raw_data)
print(corrected)
```

--------------------------------

### Homonim CLI Usage

Source: https://homonim.readthedocs.io/en/stable/cli.html

General syntax for invoking the Homonim command-line interface.

```bash
homonim [OPTIONS] COMMAND [ARGS]...
```

--------------------------------

### Visualize Source, Reference, and Corrected Images

Source: https://homonim.readthedocs.io/en/stable/_downloads/44f4de01c410955a735860c0ce80ac6b/basic_correction.ipynb

Displays a side-by-side comparison of the source mosaic, reference image, and the corrected output using matplotlib and rasterio.

```python
# VRT mosaic of the source files
src_mosaic_path = (
    'https://raw.githubusercontent.com/leftfield-geospatial/homonim/main/'
    'tests/data/source/ngi_mosaic_rgb_byte.vrt'
)

fig, axes = pyplot.subplots(
    1, 3, sharex=True, sharey=True, tight_layout=True, figsize=(9, 4), dpi=92
)

# loop over the source, reference and corrected images etc
for im_file, scale, ax, label in zip(
    [src_mosaic_path, ref_path, corr_mosaic_path],
    [255, 150, 150],
    axes,
    ['Source', 'Reference', 'Corrected'], 
):   
    # read, scale and display the image
    with rio.open(im_file, 'r') as ds:
        array = ds.read(out_dtype='float32') / scale
        
        show(array, transform=ds.transform, ax=ax, interpolation='nearest')

    ax.set_title(label)
    ax.set_xlim(-5.75e4, -5.50e4)  # zoom in
    ax.set_ylim(-3.733e6, -3.730e6) 
    ax.axis('off')

# fig.savefig('../case_studies/basic_correction-src_ref_corr.jpg', dpi=92)
```

--------------------------------

### homonim.RasterCompare.create_config

Source: https://homonim.readthedocs.io/en/stable/_generated/homonim.RasterCompare.create_config.html

Utility method to create a RasterCompare configuration dictionary whose items can be passed as keyword arguments to RasterCompare.process().

```APIDOC
## homonim.RasterCompare.create_config(threads=0, max_block_mem=512, downsampling=Resampling.average, upsampling=Resampling.cubic_spline)

### Description
Utility method to create a RasterCompare configuration dictionary whose items can be passed as keyword arguments to `RasterCompare.process()`. Without arguments, the default configuration is returned.

### Parameters
- **threads** (int, optional) - Number of image blocks to process concurrently. 0 = use all processors.
- **max_block_mem** (float, optional) - Maximum size of an image block in megabytes.
- **downsampling** (rasterio.enums.Resampling, optional) - Resampling method to use when downsampling.
- **upsampling** (rasterio.enums.Resampling, optional) - Resampling method to use when upsampling.

### Returns
- **dict** - Configuration dictionary.
```

--------------------------------

### homonim.RasterFuse.create_out_profile

Source: https://homonim.readthedocs.io/en/stable/_generated/homonim.RasterFuse.create_out_profile.html

Creates a profile dictionary for output images, defining format, data type, nodata values, and driver-specific creation options.

```APIDOC
## homonim.RasterFuse.create_out_profile

### Description
Utility method to create a profile for the output image(s) that can be passed to `RasterFuse.process()`. Without arguments, the default profile is returned.

### Parameters
- **driver** (str) - Optional - Output format driver. See the GDAL docs for available options. Default: 'GTiff'.
- **dtype** (str) - Optional - Output image data type. One of: uint8|uint16|int16|uint32|int32|float32|float64. Default: 'float32'.
- **nodata** (float) - Optional - Output image nodata value. If set to None, an internal mask is written.
- **creation_options** (Dict) - Optional - Driver specific creation options e.g. dict(compress='deflate').

### Returns
- **dict** - Output image profile.
```

--------------------------------

### RasterFuse.__init__

Source: https://homonim.readthedocs.io/en/stable/api.html

Initializes the RasterFuse class for correcting an image to surface reflectance by fusion with a reference.

```APIDOC
## RasterFuse.__init__(src_filename, ref_filename, proc_crs=None, src_bands=None, ref_bands=None, force=False)

### Description
Initializes the RasterFuse object. For best results, reference and source images should be concurrent, co-located, and spectrally similar.

### Parameters
- **src_filename** (str/Path) - Required - Path to a source image file.
- **ref_filename** (str/Path) - Required - Path to a reference image file.
- **proc_crs** (homonim.enums.ProcCrs) - Optional - Specifies which of the source/reference image CRS and pixel grid to use.
- **src_bands** (list of int) - Optional - Indexes of source spectral bands to be processed (1-based).
- **ref_bands** (list of int) - Optional - Indexes of reference spectral bands to match and fuse with source bands (1-based).
- **force** (bool) - Optional - Bypass auto wavelength matching and band-matching errors.
```