Hi there, my name's Matt. I think I could be a good fit for your project, computer vision is my area of interest and I did my masters at Cambridge in it. I also worked at a satellite image analysis company in san francisco. I'm coming in with a relatively low bid because I am new to this platform but it's not indicative of my experience level.
Looking at the image you've provided I think the easiest way to break up the image into boxes is to use a Hough Transform, which opencv has a method for. How many scans are you looking at extracting? Do they all have the same table layout?
Let me know if you want to have a chat about it.
Best,
Matt