Indian Journal of Science and Technology
DOI: 10.17485/IJST/v16i34.1777
Year: 2023, Volume: 16, Issue: 34, Pages: 2709-2718
Original Article
Snehil G Jaiswal1*, Sharad W Mohod2, Dinesh Sharma3
1Research Scholar, Sant Gadge Baba Amravati University Amravati and Registrar, G H Raisoni University, Amravati, Maharashtra, India
2Professor and Head of Department, Department of Electronics and Telecommunication Engineering, Prof. Ram Meghe Institute of Technology & Research, Amravati, Maharashtra, India
3Assistant Professor, Department of Electronics and Telecommunication Engineering, Chandigarh College of Engineering and Technology, Chandigarh, India
*Corresponding Author
Email: [email protected]
Received Date:16 July 2023, Accepted Date:03 August 2023, Published Date:12 September 2023
Background: With the proliferation of machine learning and its applications in a variety of spheres that are important to humans in their day-to-day lives, there is a pressing need for automatic detection models that can identify abnormal behaviors or acts of violence. Methods: This study examines a machine learning model that uses ensemble boosting and histograms of oriented gradients (HOG) to detect violent content from a feature vector with a single parameter. Findings: The tests performed on two benchmark datasets, such as the Hockey Dataset and the Peliculas dataset, reveal a high level of performance accuracy for the classification of violent videos. The experiment findings show that the suggested violence detection model performs well in terms of average metrics, with accuracy, precision, and recall being 90.50%, 91.80%, and 89.70%, respectively. Novelty and applications: The proposed method is capable of striking a balance between high performance and a limited number of parameters, and as a result, it can be implemented with a minimal investment of computational resources.
Keywords: Violence detection; Computer Vision; Action Recognition; Machine Learning; Histogram of Oriented Gradients (HOG); Ensemble Boosting
© 2023 Jaiswal et al. This is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. Published By Indian Society for Education and Environment (iSee)
Subscribe now for latest articles and news.