I want a lean, lightning-fast Python script that sits on Windows, continuously grabs a small, predefined rectangle of my BlueStacks window, performs real-time blob detection with OpenCV, and the instant a qualifying blob appears fires off a single, specific key press back to the emulator. The key points that matter to me are low latency capture, accurate blob filtering (size, colour or contour thresholding we can fine-tune together) and rock-solid keystroke delivery so the command never misses its timing window in-game. Please code the solution in plain Python with OpenCV as the imaging backbone; feel free to pull in lightweight helpers such as mss, pywin32, pynput, or pyautogui for screen capture and keyboard output as long as the overall loop stays responsive. I’d like the script packaged in a single .py file plus any supplementary modules you write yourself, and a quick read-me that covers setup, configurable parameters for the screen region and thresholds, and how to map the outgoing key. On my side I will test by pointing the script at BlueStacks, watching for the target visual cue, and checking that the key press lands every time with minimal delay. Attached are some screenshots for reference. Image 1 shows a typical Arena match. Image 2 shows the target color I want to track. The script should compare the area of the central player (me) with the detected enemy blobs as described in the briefing