I am creating a metric for measuring similarity of binary and multiclass images. This is useful for, eg, comparing classifications of image data in a way that considers both the structure of the image and the binary/multiclass nature of the data. To judge how well it's doing, I have to get some human feedback.
The survey is a Shiny app you can take.
The survey tool is based on a Shiny app from Econometrics by Simulation.