Research on compliance of intelligent penalty system of tennis match based on multi-source heterogeneous data fusion

To address scattered data sources and inconsistent rule interpretation in tennis match officiating, this paper constructs an intelligent officiating system based on multi-source heterogeneous data fusion, with rule compliance as its core contribution rather than mere performance enhancement.The system integrates match video, on-court sensor data, and referee records to support -not replace -human officials by providing evidence-based decision recommendations.Four key metrics evaluate system performance: decision accuracy (agreement with official final calls), consistency rate (agreement with review panel judgements), misjudgement rate (proportion of incorrect in/out calls), and rule conflict rate (deviations from ITF rule interpretations).In sample verification from professional events, the complete fusion model achieved a decision accuracy above 0.96 in hard-court events, a consistency rate of approximately 0.93 with review referees, a misjudgement rate below 0.03, and a rule conflict rate under 0.02.Results demonstrate that prioritising rule-aligned decision-making through multi-source collaboration enhances both penalty stability and regulatory conformity, offering empirical support for compliant intelligent officiating in tennis competitions.