Digital Repository

"Eagle-Eye” Legality Analysis of Bowling Action in Cricket Using Deep Learning

Show simple item record

dc.contributor.author Wijayabahu, I.T
dc.date.accessioned 2022-03-14T08:48:31Z
dc.date.available 2022-03-14T08:48:31Z
dc.date.issued 2021
dc.identifier.citation "Wijayabahu, I.T (2021) “Eagle-Eye” Legality Analysis of Bowling Action in Cricket Using Deep Learning . BSc. Dissertation Informatics Institute of Technology" en_US
dc.identifier.issn 2017296
dc.identifier.uri http://dlib.iit.ac.lk/xmlui/handle/123456789/958
dc.description.abstract " Cricket in present day has a major concern of identifying the legality of a bowler’s bowling action. Due to the complexity of the bowling arm's biomechanical movement, the umpire on-field is unable to determine the legality of the bowling action. The state of-the-art method for determining the legality of their action is using inertial sensors. The legality of the bowling action is a crucial aspect of a cricket player's career. To do a clearance test with existing sensor-based systems is costly and time-consuming due to resource scarcity. Further to thorough analysis and research, this study introduces the Eagle-Eye system for analyzing bowling action in cricket. The Eagle-Eye system calculates the bowler's elbow extension in order to determine whether the action is legal or illegal. 3D human pose estimation, Detectron2 object detection, and a mathematical approach have been used to measure the bowler's elbow extension in order to determine if the action is legal or not. Both the mathematical approach and the system for calculating angles had been tested and evaluated. " en_US
dc.language.iso en en_US
dc.subject Computer Vision en_US
dc.subject Classification en_US
dc.subject Inertial sensors en_US
dc.subject 3D Analysis en_US
dc.subject Human Pose Estimation en_US
dc.subject Deep Learning en_US
dc.subject Bowler en_US
dc.subject Cricket en_US
dc.subject Bowling Action Analysis en_US
dc.title "Eagle-Eye” Legality Analysis of Bowling Action in Cricket Using Deep Learning en_US
dc.type Thesis en_US


Files in this item

This item appears in the following Collection(s)

Show simple item record

Search


Advanced Search

Browse

My Account