Need to extract OCR data for specific use cases (different types of documents e.g. bank cheques, purchase receipts, business cards, etc. ) with high accuracy (99%+) , automated and dynamic ability to identify key areas of the document and extract the specific pieces of information and identifying what is each piece of information.
For example if it is a receipt, then total billed amount, date of transaction and name of vendor company and order ID should be extracted and identified.
Platforms where it should be able to run: iOS, Android, Web (PHP)
Languages to be OCR'd: Structured Arabic, Turkish, English.
Actual scope will be discussed and defined during the interview.