The Image Processing Division of Bundito Heavy Industries reports success in analyzing the first batch of Propaganda tiles.

Terminal output attached as proof of life.

UPDATE (19:03)

I’ve crunched through all 19 (or is it 18.5?) volumes of the Propaganda tiles and stored all the color information in a SQLite database. I’ve got the three most-prominent colors in each image extracted and filed away. There are a total of 1,047 tiles, by the way.

The next trick will be matching that color data to a color chosen by a user. Since it’s unlikely that the user choice will hit an exact match, I need to find colors that are close to the user’s picked color. Finding colors that are close isn’t a straightforward endeavor. There’s going to be more math tomorrow. Lots more math.

In the mean time, I got feedback on some bug fixes I’d submitted for KDE. Henrik has more guidance and specific methods he’d like me to employ. More learning!


Leave a Reply

Your email address will not be published. Required fields are marked *