This project utilizes the Python OCR package called Tesseract and other image cleaning packages to create an OCR tool.
Our target is to find out the total payment amount and other relevant fields of the receipt accurately, so that manunal effort can be reduced.
09/15 Uploaded intial code, will do more NLP analysis to resolve the issues can not be resolved by OCR and also plan to test more receipts of different formats
09/17 Used NLP to post-process and extract the total payment amount, will try to extract more fields as such address in the future.