Abstract: In cricket, Leg Before Wicket (LBW) decisions have long been a subject of controversy and debate, often relying on the subjective judgment of on-field umpires. With the advent of advanced technologies, Artificial Intelligence (AI) and Machine Learning (ML) have emerged as powerful tools to enhance decision-making accuracy in sports. The proposed system presents AI-based LBW detection systems, exploring the integration of computer vision, ball tracking procedures, pose estimation, and predictive models to improve real-time predictions and reduce human. The combination of Histogram of Oriented Gradients (HOG) and Support Vector Machine (SVM) classifiers can achieve approximately 80–90% accuracy in classification tasks such as ball detection and LBW decision-making using side-on video footage.

Keywords: Leg Before Wicket (LBW), Artificial Intelligence (AI), Machine Learning (ML), Umpire, HOG.

| DOI: 10.17148/IARJSET.2025.125255