Abstract:
One of the hot topics in the modern era of cricket is to decide whether
the bowling action of a bowler is legal or not. Because of the complex biomechanical
movement of the bowling arm, it is not possible for the umpire to
declare a bowling action either legal or illegal with the help of naked eye.
Inertial sensors are currently being used for activity recognition in cricket
for the coaching of bowlers and detecting the legality of their moves, since a
well trained and legal bowling action is highly signi cant for the career of a
cricket player. After extensive analysis and research, we present a system to
detect the legality of the bowling action based on real time multidimensional
physiological data obtained from inertial sensors mounted onto the bowlers
arm. We propose a method to examine the movement of the bowling arm
in the correct rotation order with a precise angle. The system evaluates the
bowling action using action pro les. The action pro les are used to simplify
the complex biomechanical movement of the bowling arm and is also used
to minimize the size of data provided to the classi er. The events of interest
are identi ed and tagged. Algorithms such as Support Vector Machines
(SVM), k-Nearest Neighbor (k-NN), Na ve Bayes, and Arti cial Neural Network
(ANN) are trained over statistical features extracted from the tagged
data. The proposed method achieves very high accuracies in the correct
classi cation of bowling action.
