Abstract
Sports image classification using image processing and machine vision is a growing area of research that involves the use of algorithms and techniques to identify and analyze objects in sports images and videos. This technology has a wide range of applications, including detecting illegal plays, analyzing team performance, and creating highlight reels. Additionally, it can provide valuable visual feedback during training and competition. In this paper, we propose a novel deep learning and optimization hybrid framework for sports image classification. Specifically, we use a modified version of the Battle Royal optimization algorithm as a feature selector to reduce the dimensionality of the images and achieve higher accuracy with only the essential features. We evaluate the proposed framework using sports images and demonstrate that our WOA-based framework outperforms other methods in terms of both classification accuracy and dimensionality reduction. Our results highlight the effectiveness of the proposed approach and its potential to improve sports image classification.