Help in developing app

We are now in developing our thesis system, this is about using NLP to automate librarian task such as abstracting, cataloging, classification and indexing.

What i want in finished mobile app is first it will scan the book(not all pages, some important part), then it will extract the scanned (images) to text, after that it will comes now the NLP to do the task (the 4 task ex. cataloging) and output it.

i also forgot the needed databases.

Can someone help me or give me some idea how to develop this pleasee. super beginner here

Thank youuuuu!