Cricket Bowling Technique Analysis Using Pose Estimation and Deep Neural Networks

Abstract

Now, human umpires watch the delivery and render an arbitrary ruling based on their interpretation of the rules to determine if a bowler's action was legal. This may result in errors and inconsistencies that affect the game's outcome. Incorrect bowling techniques can also result in injuries and have a bad effect on players' health. To develop a method for evaluating cricket bowlers' actions that employs machine learning to provide an accurate and trustworthy approach to judge a bowler's legality. This will advance the fields of computer vision and machine learning, as well as the consistency and fairness of cricket matches. It will also benefit players' health and wellbeing. Convolutional Neural Networks (CNNs), a type of deep learning algorithm, are employed in the system to extract information that can be used to categorize the action as either legal or illegal from images and videos of the bowler as they deliver the ball. The system's accuracy and dependability were examined, and the findings showed that it was highly accurate and reliable at spotting erroneous bowling motions. The system's great accuracy and dependability in identifying erroneous bowling actions during testing suggests that it can offer a trustworthy and objective means to judge whether a bowler's action is legal. The machine learning-based cricket bowling action inspection system may greatly enhance the uniformity and fairness of cricket matches as well as the players' health and wellbeing. The technology also offers a practical application for computer vision and machine learning, which has the potential to advance these fields. Future studies may examine the technology's additional advancements and uses outside of sports analytics.