Python3 application for image noise removal by aligning and averaging many images.
It uses OpenCV3 and NumPy.
OpenCV is used for image analysis (feature detection, matching, finding homography, transforming).
NumPy is used for image processing and averaging.
Images are processed with float32 accuracy and saved as 16bpc PNGs.
Samples included.
- place images in input directory
- optionally put 16bpp raster darkframe into input_darkframe dir
- adjust some params in script (at the moment this is the only way to customize processing)
- run application
- wait until it finishes or press Ctrl + C, it should exit gracefully and save output image
- check output directory
YouTube videos:
Divided view, upper-left - original, bottom-right - averaged, center boosted 8x. Images were taken with smartphone camera:
