Berk Belhan

Computer Vision & Game Automation

GoalSense - Real-Time Goal Detection Overlay

A real-time computer vision system that detects goals in EA FC 26 by monitoring scoreboard changes on screen and instantly triggers custom audio reactions, without accessing game memory or processes.

Explore the Project

Problem

Football simulation games provide no simple way for players to create personalized goal celebrations or automate reactions to in-game events. Traditional approaches often rely on game memory access or third-party modifications, which can introduce compatibility issues and anti-cheat risks.

Solution

Developed GoalSense, a lightweight desktop application that uses screen capture and computer vision techniques to detect score changes directly from the game's user interface. When a goal is confirmed, the system instantly plays custom audio effects while maintaining full compatibility with anti-cheat systems by operating exclusively on captured screen pixels.

Architecture

Built with Python using a modular event-driven architecture. The application continuously captures a configurable scoreboard region using MSS, performs template-based digit recognition with OpenCV, and validates score transitions through a multi-frame confirmation pipeline. A state management layer handles debounce logic, cooldown protection, and goal attribution, while asynchronous audio playback ensures non-blocking real-time performance.

Technologies

PythonOpenCVComputer VisionMSSPygameTemplate MatchingImage ProcessingPyTestEvent-Driven Architecture

Demo Video

Demo Recording In Progress

A detailed walkthrough video will be available soon.