Affiliation:
1. Department of Mathematics, Computer Science and Physics, University of Udine, 33100 Udine, Italy
Abstract
This paper presents a novel low-cost integrated system prototype, called School Violence Detection system (SVD), based on a 2D Convolutional Neural Network (CNN). It is used for classifying and identifying automatically violent actions in educational environments based on shallow cost hardware. Moreover, the paper fills the gap of real datasets in educational environments by proposing a new one, called Daily School Break dataset (DSB), containing original videos recorded in an Italian high school yard. The proposed CNN has been pre-trained with an ImageNet model and a transfer learning approach. To extend its capabilities, the DSB was enriched with online images representing students in school environments. Experimental results analyze the classification performances of the SVD and investigate how it performs through the proposed DSB dataset. The SVD, which achieves a recognition accuracy of 95%, is considered computably efficient and low-cost. It could be adapted to other scenarios such as school arenas, gyms, playgrounds, etc.
Reference38 articles.
1. Sudhakaran, S., and Lanz, O. (September, January 29). Learning to detect violent videos using convolutional long short-term memory. Proceedings of the 14th IEEE International Conference on Advance Video and Signal Based Suirveillance (AVSS), Lecce, Italy.
2. Violence Detection in Videos by Combining 3D Convolutional Neural Networks and Support Vector Machines;Accattoli;Appl. Artif. Intell.,2020
3. Cheng, M., Cai, K., and Li, M. (2021, January 10–15). RWT-2000: An open large scale video database for violence detection. Proceedings of the 2020 25th International Conference on Pattern Recognition (ICPR), Milano, Italy.
4. Nievas, E.B., Suarez, O.D., Garcia, G.B., and Sukthankar, R. (2011). Computer Analysis of Images and Patterns, Springer.
5. A dataset for automatic violence detection in videos;Bianculli;Data Brief,2020