I'm wanting to develop a program that can interpret the visual things on screen and potentially output something--maybe audio--to represent it.
I know it would be extremely difficult to try and get everything that can be seen easily interpreted. For now, I'm just wanting to focus on something that can tell about where sprites are, maybe how big they are, things like that.
Problem is, I don't really know where to start. I'd greatly appreciate it if someone could point me in the right direction. (I can program in java... and know full well that this is useless for what I want to do.
