I have several images of graphs of one-variable functions. I'd like to get samples of such functions and thus generate a table.

Right now, my best bet is using GIMP: crop, show grid, resize (so the grid fits the sample size I want). But I imagine there's a more efficient way to do this.

(The images are from different sources and quite heterogeneous. Different resolution and quality. Some of the graphs are multiplot, some of them have grids, etc)

What software would accomplish the task?


You could do this with python and numpy/Scipy the steps would be something along the lines of:

  1. Load the image into an array
  2. Scan for distinct colours
  3. Prompt the user for which colours are of interest
  4. Prompt the user for scale factors & offsets
  5. Take the row/col of each entry that matches the colour(s) of interest, (be sure to allow some leeway for anti-aliasing, ect.), as the X, Y values and multiply/offset to get the actual values
  6. Output as a table

Not the answer you're looking for? Browse other questions tagged or ask your own question.