How to classify,extract and segregate data from a bulk of document?

I can make the question more clear. I came across PiXL CLASSIFIER AND EXTRACTOR that can classify extract and segregate the data required from a bulk of documents. Let us consider a financial sector where each and every customer will have their own id cards copies it can be an aadhaar card,pan card or passports. But they are all in image formats. There will be thousands of accounts in a bank where is is very difficult to classify them as aadhaar card passport or else extract a particular data from this bulk of documents. PiXL CLASSIFIER AND EXTRACTOR do this with better accuracy and quality even if the images are having low quality. They can even deliver data with 98% accuracy. It was a new information for me. I would like to share with you folks.