Open brynpatel opened 3 years ago
I half started on another "image exploration" notebook (i.e. examining the object images not the object metadata), I wonder if your plan (which sounds great) would be better as a new notebook there, so there are data explorations and image explorations.
The down side of that is I then have to finish the data exploration one myself =)
I couldn't decide if every code sample should be self contained (hence importing panda each time) or not. I wasn't very consistent though so a tidy up of that makes sense.
On the Louvre image licencing - I think from reading this in the Louvre Terms of Use - "The downloading and re-use of medium-format photographs published on the collections website representing works that are not protected by copyright (hereinafter called the “Photographs”) are permitted, free of charge, for any non-collective use within a strictly private context " is a bit unclear as to what private context means. As both of us are not lawyers probably better for the moment to just link to the object on their site.
My plan is to compare the colour pallets of the spring pictures in the Louvre and the spring images in the V&A. I started off by scraping the HTML from a Louvre website, using the ARK code of an item and beautiful soup. I then used Open CV to do an image comparison of the two images, by creating a colour palette based on the main pixel colours used in an image. The next thing I would do is find some more interesting images to compare, and draw a conclusion over which image contains the nicest variety of colours. I also cleaned up some of the code, e.g. removing the lines of code with reimport panda multiple times.
Is this a direction that you think would work well for this page, or would you rather I focus on the metadata of the images and how that compares? Another question I have is how the Louvre's copyright policy relates to the displaying of images on this sight, and whether I should try to avoid displaying the Louvre's images?