Continuously scanning and recognising text with Unity3D on Android

I have inherited a Unity3D project where the deployed application (on an Android device) allows the user to take a picture (UI button) of some serial number and then the app recognizes the serial number string.

Apparently, as there was no freely-available Unity3D plug-in for accurate OCR (Optical character recognition) they built an Android plug-in based on tess-two (a fork of Tesseract Tools for Android which is a public set of Android APIs and build files for the Tesseract OCR and Leptonica image processing libraries).

Tesseract is free and extensible, and also open source and available under Apache 2.0 License and backed by Google, so I can ask this question on a public forum, though of course I cannot share any code, even though it is simple code that achieves everything I describe here...

Now, I have to automate the OCR operation; meaning the user would like the OCR operation to be done automatically and continuously, and not only by pressing a button to take a picture, saving the screenshot on a 2D texture, sending it off to receive the recognized text, etc...

My first thought was to still take a picture every second, while scanning the serial number, and assuming that performance will not be affected much, running the OCR operation on those taken images per second, until the serial number string is recognised.

But I wondered if someone might have done something similar (scanning an area, looking for text, recognizing it while scanning, etc...) where thay can share with me any thoughts or suggestions or even code snippets, specially any Unity-specific classes that might be helpful to me.