How can font types be recognized to allow code to act on text of different font type in a scanned textual document

I am new to Matlab, feel free to abuse me. I have performed internet search and search in Mathworks resources on this subject and not found what I am looking for. I have found this resource which I will read later today, CHARACTER RECOGNITION, Handwritten character Recognition: Training a Simple NN for classification using MATLAB; Mentor: prof. Primož Potočnik, Student: Žiga Zadni

I will be scanning a bible that has various fonts used for text, section headings and footnotes on each page. I need to be able to differentiate these fonts based on size, font, superscript/subscript, etc. to allow cross referencing within the document and possibly cross referencing with a different bible/book. I hope to be able to use the resulting project to link various documents together for any subject.

I have the Text Analytics, Statistics and Machine Learning, Image Processing, and Deep Learning toolboxes (and several others).

Before I scan the document, I would like to know the best format for the scan to be able to perform the task described above. I would also appreciate guidance to a good reference for this type of analysis or personal recommendations from experienced Matlab users and Mathworks.

Thank you