OCR - Detect Checkboxes in PDF Forms
I am developing cloud-base software for processing legal documents. My developers have already implemented the User Portal and UI of a working platform that utilizes API with 3rd party OCR products to convert the legal PDF document to text. I am able to successfully convert all file format PDF files (both containing True Type font as well as PDFs in FLAT file image format) into TEXT .
PROBLEM Remaining:
Some documents include several questions. Each question has an item #. Also each question has a Checkbox (slightly different size boxed) to its left side. My current OCR solution cannot 100% reliably detect which questions have a checkbox to tehir left side and also which checkboxes are checked and which are not!
My developer tried using OPenAI API to convert document and create list of all checkboxes, however some of the pages apparently do not get processed by OpenAI because of legal and privacy contents inside them!
Please see attached Screenshot for a snapshot of a document
I simply need a QUICK solution that my develop access via API that:
- They will pass a PDF file to our code (both true type as well as Flat image PDF)
- YOU would need to process PDF file, and create a list of all question numbers AND the status of each checkbox to pass back to my developer via API.
I have a very limited budget and even less time to complete this task. Looking for some one who has done this before and can provide a quick solution to my developer please.
Thanks
David
dnosrati@aol.com