This project aims to get rid of the vertical lines that occur on metOp ASCAT images.
For this script to work, you need Python 3 installed and added to PATH.
After you installed Python on your machine, you need to install Image and matplotlib by running
pip install Image and pip install matplotlib.
Simply put an ASCAT-image named input.png into the same directory as the script.
Then you just have to run the script. That's it. All results will be saved in the same directory.
The script first calculates the average pixel-brightness of each column by adding all values up and dividing by the height.
After we did this, we can see the average brightness of each column in the noisepattern.png file created by the script.
We then substract this noisepattern from every row, normalize the result and get ASCAT_denoised.png.
As an optional 4th step, the script uses matplotlib.pyplot to colorize the image.
Currently, each channel has to be treated seperately. The script is still WIP and unable to do batch-proccessing.
The algorithm can theoretically handle the ASCAT-ALL.png image created by satDump,
but the result for single-channel proccessing is better. Since this uses the average over each row, higher images should give a better result.