Trying to automate a system here, either through off-the-shelf or roll-your-own solutions: how to take a PDF form (e.g. membership, taxes, etc.) and detect fields? Preferably along withe the label(s), type, length, etc..
All the OCR I've seen is directed at reading text, but not other forms of content.
Note: I'm NOT trying to capture responses, but to capture blank fields on an empty form.