An automated snick detection and classification scheme as a cricket decisionreview system

Abstract

Umpire decisions can greatly affect the outcome of a cricket game. When there is doubt about the umpire's call for a decision, a decision review system (DRS) may be brought into play by a batsman or bowler to validate the decision. Recently, the latest technologies, including Hotspot, Hawk-eye, and Snickometer, have been employed when there is doubt among the on-field umpire, batsman, or bowlers. This research is a step forward in gaging the true class of a snick generated from the contact of the cricket ball with either (i) the bat, (ii) gloves, (iii) pad, or (iv) a combination of bat and pad. Preprocessing included noise removal from the snick audios using the Audacity program.Machine learning-based classification is achieved by training neural networks with audio features of the snick. A support vector machine was employed to enhance the classification system. Twenty-one features comprising time and frequency domain characteristics were compiled. After one-way analysis of variance was employed and multicomparison analysis was performed, a set of seventeen features was selected and utilized to increase the accuracy of the proposed DRS.The system was trained on indigenous snick data gathered by ourselves and then tested for real cricket snick scenarios.The system achieves a classification rate of 98.3 percent for the self-collected data while presenting an accuracy of 85.7 percentage for the real cricket snick scenarios. The research also developed a dataset of snick audios comprising 132 signals for the four categories to aid researchers working in the same field.

Technological assistance in highly competitive sports for referee decision making: A systematic literature review.

“…(16,17,26) Likewise, we found that, in cricket, figure skating and wushu, similar problems are evident, such as having an obsolete and tedious process for detecting sporting repercussions in games of these sports, as well as for modifying the score of game markers and for detection of movements of players as referees for the issuance of a precise judgment, this being the origin of the reason for their studies, which seek to detect the movements of sports objects and referees for the detection of game moments such as automating processes to updating game scores. (6,8,10,11,12,18,21,22,23,25) Likewise, in other sports such as basketball, hockey, fencing, handball, race-walking, skateboarding, we also find problems with presence in more than one sporting discipline, which lie in having a traditional approach for detecting sports incidents. (7,9,11,13,14,19,20) All…”

“…Ravi et al (22) Khan et al (23) Kamble et al (24) Shahjalal et al (25) Reno et al (26) Figure 3. Number of scientific publications this, in order to achieve precise decision making and a fair game, through event detection by having automated approaches for the detection and monitoring of players, balls and referee movements.…”

“…Having proposals of various characteristics as well as various artificial intelligence approaches for its development, it is thus that we have the development of an algorithm focused on the detection and monitoring of the trajectory of the ball as the game progresses, based on Transfer learning using principles of Deep learning by Kamble et al (24) , the development of systems capable of detecting and classifying sports actions with respect to snick moments in cricket games as well as for the detection of referee movements, making use of Neural networks Inception V3(CNN), VGG19 networks, techniques related to an approach based on Deep learning and Machine Learning. (22,23) Likewise, there is a segmentation method for predicting and evaluating the quality of actions performed in a figure skating competition, using Deep Learning. (6) Thus, we can show that the most useful artificial intelligence approaches for the development of solutions focused on the detection and monitoring of players, sports objects, and the identification of referee movements make use of Deep Learning and Machine learning principles and techniques.…”

“…Abdullah et al (20) Skateboarding Kowsher et al (21) Cricket Ravi et al (22) Cricket Khan et al (23) Cricket Kamble et al (24) Soccer Shahjalal et al (25) Cricket Reno et al (26) Tennis…”

“…sports in different disciplines. Having analyzed the algorithms used in the research presented, it was found that there is a greater presence in the use of the support vector machine (SVM) algorithm, this is due to the fact that a large percentage of articles by Mili et al (8) , Iyer et al (18) , Kowsher et al (21) , Khan et al (23) and Shahjalal et al (25) aims to detect and/ or identify movements of players and referees, thus allowing the use of this algorithm, as it is focused on the detection of events that occur through the processing of images and videos.…”