But why bother with image recognition? And pay a heavy latency penalty which makes it difficult to implement PID-style controls on player's aim? Its a game of milliseconds and opencv is comparably slow, as is the I/O involved.
Far easier to do a MiTM on the network traffic and recreate state in another process.
Still easier to clone memory pages used by the target app, or use hardware that allows dma.
Nobody really cares because most players are honest and play to have fun.