OCR and Regular Expressions
Most of OCR data is not useful and might slow down the Full-Text search. Consider adding options for extracting specific values from OCR data using regular expressions. These values can be used to index the document, and the rest of OCR data can be ignored which will minimize the size of Dokmee database.
For example, in my case I know that a customer account number can only be 9 digits and always starts with "10", so when scanning a subscription form, the account number can be identified easily with regular expression.