AI systems bring new accuracy to basketball foul detection and analysis

Professional basketball is faster, more physical and more refined. Even the smallest fouls can tilt the balance of the game. Currently, research published in the International Journal of Computational Systems Engineering shows how artificial intelligence, particularly machine vision, can provide unprecedented accuracy in the detection of fouls, one of the most contested areas of sports.

Machine vision is a form of AI that allows computers to interpret and analyze visual data. In this study, researchers apply it to high-level basketball footage and focus on identifying and classifying fouls in real time. This technology separates subtle movements, points of contact, and spatial dynamics that often escape the naked eye, with even experienced judges from each video frame.

This study focused on analyzing games involving China’s national basketball team and its international opponents. This study found significant differences in two-sided foul patterns. For example, Chinese players committed more fouls during their opponent’s shooting attempts, known as shooting-related defensive fouls. In contrast, their opponents were more likely to foul during dribbling plays.

The AI ​​system also tracked subcategories of fouls, including illegal use of hand and player conflicts, revealing behavioral trends that far exceed standard box score statistics.

This new level of granularity that can be achieved with machine vision, if adopted as a standard, will have great significance in the future of the game. Traditionally, discovering a foul has rested on human observation, referees, coaches and analysts. Each of these characters is subject to the limits of perception, fatigue, and, inevitably, bias.

In contrast, machine vision provides a consistent, objective view. You can see the context as well as the contact information. Where did it occur on the court, at what angles and how both players were moving before the foul.

For coaching, this type of data can be used to diagnose weaknesses and imbalances in defensive or aggressive behaviors in a team. Specific foul patterns, if discovered early, could point to problems with footwork, positioning, or reaction time. All of these areas may be addressed through tailored training. Especially for players navigating the thin line of razors between offensive and reckless plays, this technology provides clear feedback that a simple post-match video review cannot provide.

Perhaps most importantly, research reconstructs the way judges train. Accurate, data-driven feedback on decision-making in live situations can improve both consistency and equity. These are two of the most sustainable challenges that typically host high-speed sports.

Machine vision is unlikely to be completely replaced by human judgement, but it can quickly become an essential tool for development, helping to reduce errors and improve the flow of professional gaming.