Bringing Sergei Prokudin-Gorskii's historical photographs to life through digital alignment
This project implements Sergei Mikhailovich Prokudin-Gorskii's technique for creating color photographs by digitally aligning and combining three separate glass plate negatives taken through red, green, and blue filters.
The goal is to automatically align the RGB channels to produce a single color image by computing optimal displacement vectors that minimize alignment error using Normalized Cross-Correlation (NCC) as the primary metric (with L2 distance used specifically for the emir image for better results).
For images under 1000px width
For high-resolution images
< 1 second per image
±15 pixels displacement
Normalized Cross-Correlation
< 4 seconds per image
±15 pixels (coarsest level)
±2 pixels at finer levelsNormalized Cross-Correlation
Emir uses L2 distance
I initially implemented an edge detection enhancement using Sobel filters from scikit-image to improve alignment accuracy by focusing on edge features rather than raw pixel intensities.
However, upon discovering that external image processing libraries were not permitted for the core implementation, I removed this code from the notebook file which I submitted. Luckily, I didn't delete the pictures processed with this method so I can still showcase the results here.
To see other pictures processed with this method, feel free to refer here.