ArrNorm

ArrNorm is a command line program for apply the radiometric normalization to the target image based on reference image using the IR-MAD algorithm to locate invariant/variant pixels for a relative radiometric normalization.

The algorithm takes advantage of the linear and affine invariance of the Multivariate alteration detection (MAD)
transformation to perform a relative radiometric normalization of the images involved in the transformation, using the correlation of the iteratively reweighted MAD (IR-MAD) [1]

Stop condition is set by max iteration or with a minimum no-change probability threshold. With more iterations the algorithm try to find a better match to the reference image, decreasing the delta, the plugin select the best delta for the final result. However, after several iterations the changes in the delta are imperceptible.

[1] M. J. Canty (2014): Image Analysis, Classification and Change Detection in Remote Sensing, with Algorithms for ENVI/IDL and Python (Third Revised Edition), Taylor and Francis CRC Press.

Fig.1 - Example of a Landsat image normalization, using a multi-year average (reference) to normalize a scene. Those pixel changes are because, some remote sensing imagery pixel values are affected by different causes such as: sensor angle, sun position, clouds and seasons. Be sure that you are using the same ‘style’ for all layers in Qgis (use copy/paste style for that) to visualize and check the normalization visually.

Check the ArrNorm-Qgis-processing implementation of Arrnorm as a Qgis processing.

Installation

For example with Anaconda/Conda environment:

conda install -c conda-forge gdal numpy scipy matplotlib
pip install https://github.com/SMByC/ArrNorm/archive/master.zip

Parameters

-ref reference image: The reference image to normalize the target image.
-p number of threads: Number of threads to process several files at the same time (default: 1).
-i max iterations: Maximum number of iterations (default: 10).
-m create mask: Create a mask file with the valid data (default: False).

For other parameters check the help:

$ arrnorm -h

Examples:

$ arrnorm -ref reference.tif target.tif

$ arrnorm -i 15 -p 3 -ref reference.tif target01.tif target02.tif target03.tif

About us

ArrNorm was developing, designed and implemented by the Group of Forest and Carbon Monitoring System (SMByC), operated by the Institute of Hydrology, Meteorology and Environmental Studies (IDEAM) - Colombia.

Author and developer: Xavier C. Llano xavier.corredor.llano@gmail.com
Theoretical support, tester and product verification: SMByC-PDI group

License

ArrNorm is a free/libre software and is licensed under the GNU General Public License.