I'm on a project that processes hundreds of forms. The forms have consistent formats but are filled out by hand by different people. I need a way to quickly process all of this data into electronic forms.
OCR recognition for typed documents seems mature but for handwriting is very lacking.
For this thought, let's consider a form with several fields structured like this:
Field1: Value1
For example:
Name: John
where Name
is the field and John
is the value.
Considering that forms are structured and typed, OCR should be able to recognize/interpret the fields. However, the values of the fields are handwritten and OCR will perform very poorly for them.
So, is there a way where the fields would be recognized on the image, and then an image chunk of the value would be returned?