William Huang Wang ¹ and Andrew Park ² , ¹ USA, ² California State Polytechnic University, USA

Notation forces players to pause and record every move, which disrupts focus and wastes valuable time during games. Existing solutions either digitize written score sheets or depend on expensive smart boards, both of which remain impractical for most players. This project introduces an affordable camera-based system that automatically records moves in real-time using OpenCV for image calibration, YOLOv8 for object detection, and Python for validation and PGN generation [6]. The program captures live video, detects and flattens the chessboard, identifies piece types and positions, and updates the game state instantly. During testing, experiments have evaluated speed, accuracy, and resilience under varied lighting and occlusions. Results confirmed performance in reliable move recognition even in challenging conditions [7]. By combining cost-efficiency, portability, and real-time precision, the system offers a practical and accessible way to modernize chess notation, allowing players to focus entirely on strategy and gameplay

Computer Vision, Chess Notation, Object Detection, Real-Time Systems