Identifying Text on image (ebook reader testing)

Hi All,
Currently i am automation an ebook reader, where the book pages are displayed as images, if the user do long press on text highlights and adding user notes will appear. These and will be synced in other devices.

Problem i am facing is, UIautomator is not identifying the text, its showing entire page properties only, can someone please help me on this.

How we can verify in multiple devices ?(Like chatting program) ?

A lot thanks in advance.