School Bullying Identification Based on LSSVM Algorithm
Keywords:
Intelligent Monitoring System, Crowd Density, Movement Intensity, Degree Of Movement Clutter, Optical Flow Algorithm, LSSVM.Abstract
This research is about developing a monitoring system that can supervise school bullying. Based on the analysis of characteristics of crowd behaviors, we extract 3 features of crowd behaviors to identify the normal incidents and school bullying, which are crowd density, movement intensity, and degree of movement clutter. The crowd density is estimated with the method-based pixel statistics. According to the optical flow algorithm, we can get the movement vectors of all people in each frame image. Then calculate the mean amplitude of all the vectors as the movement intensity. As for the degree of movement clutter, firstly get all the angles of movement vectors according to the vectors’ coordinates of x and y. Then calculate the standard deviation of all the angles as the degree of movement clutter. Finally, We apply the LSSVM algorithm to identify school bullying incidents. Randomly choose 30 groups of images from a video that contains normal incidents and school bullying incidents. Extract the 3 movement features and label all the images as 0 or 1, with 0 representing a normal incident and 1 representing a school bullying incident. Apply the training data to train the LSSVM model and then apply the 20 groups of testing data to test the prediction accuracy. The results show that the classification accuracy is only about 60%. Through the characteristics analysis of school bullying, we only take movement intensity and degree of movement clutter as input parameters to train the SVM model. The results show that the prediction accuracy can be improved to about 90%. So we can use the SVM model to identify school bullying incidents and take the crowd density as an early warning parameter to detect the school bullying ahead.
Downloads
References
Ma, Shuoping et al. ““School bullying”: How the middle school system affects school bullying.” Advances in Education, Humanities and Social Science Research (2023): n. pag.
Gaffney, Hannah et al. “Examining the Effectiveness of School-Bullying Intervention Programs Globally: a Meta-analysis.” International Journal of Bullying Prevention 1 (2019): 14-31.
Farrington, David P. and Maria M. Ttofi. “School-Based Programs to Reduce Bullying and Victimization.” (2012).
Barbero, José Antonio Jiménez et al. “Effectiveness of antibullying school programmes: A systematic review by evidence levels.” Children and Youth Services Review 34 (2012): 1646-1658.
iménez-Barbero, José Antonio et al. “Physical education and school bullying: a systematic review.” Physical Education and Sport Pedagogy 25 (2020): 100 - 79.
Miraclin, G.Litisha et al. “Population Density Estimation using Single Shot Detection Algorithm.” 2023 International Conference on Sustainable Computing and Smart Systems (ICSCSS) (2023): 695-699.
McIntyre, Anne F. et al. “Key population size estimation in Nigeria: applying novel Bayesian methods for analysis of three-source capture-recapture data (Preprint).” (2021).
Yang, Kai et al. “Population counting method based on semantic perceptual network.” 2020 39th Chinese Control Conference (CCC) (2020): 6475-6479.
Wang Erdan, Li Xiaohua, Shen Lansun. Based on multi-scale analysis and fractal crowd density estimation method [J]. Computer engineering and application, 2005, 041 (029) : 35-38, DOI: 10.3321 / j.i SSN: 1002-8331.2005.29.010.
XIE P C. Research and implementation of crowd density estimation in real-time surveillance in complex scenes [D]. University of electronic science and technology [2024-01-03]. DOI: CNKI: CDMD: 2.1011.193549.
Kishore, Polurie Venkata Vijay et al. “Crowd Density Analysis and tracking.” 2015 International Conference on Advances in Computing, Communications and Informatics (ICACCI) (2015): 1209-1213.
Chakole, Pallavi D et al. “Crowd behavior anomaly detection using correlation of optical flow magnitude.” Journal of Physics: Conference Series 2273 (2022): n. pag.
Sajid, Amna et al. “Crowd Behavior Categorization using Live Stream based on Motion Vector Estimation.”Computer Science (2018).
Nam, Yunyoung and Sangjin Hong. “Real-time abnormal situation detection based on particle advection in crowded scenes.” Journal of Real-Time Image Processing 10 (2015): 771-784.
Zhi, Huilai. “Anomaly Detection Based on Motion Direction.” Acta Automatica Sinica (2008): n. pag.
Grant, Jason M. “Analysis of Crowd Behavior Based on Optical Flow: Detection,Classification, and Clustering.” (2018).
Yogameena, B. et al. “Support Vector Machine-Based Human Behavior Classification in Crowd through Projection and Star Skeletonization.” Journal of Computer Science 6 (2010): 1008-1013.
Hu Zhengping et al. “Overview of Abnormal Target Detection and Localization in Video Surveillance Systems.” Journal of Yanshan University. 2019 (01)
Li Ming et al. “Laboratory Abnormal Behavior Detection Based on Video Image Processing.” Experimental Technology and Management. 2018 (11)
Zhang Dongxiao and Li Jiale. “Research on Video Intelligent Alarm System Based on Abnormal Behavior Detection Technology.” Computer Programming Skills and Maintenance. 2017 (14)
Chen Jia et al. “A moving object detection method based on the combination of frame difference method and fast image segmentation.” Modern Electronic Technology. 2016 (03)
Downloads
Published
How to Cite
Issue
Section
License
This work is licensed under a Creative Commons Attribution-ShareAlike 4.0 International License.
All papers should be submitted electronically. All submitted manuscripts must be original work that is not under submission at another journal or under consideration for publication in another form, such as a monograph or chapter of a book. Authors of submitted papers are obligated not to submit their paper for publication elsewhere until an editorial decision is rendered on their submission. Further, authors of accepted papers are prohibited from publishing the results in other publications that appear before the paper is published in the Journal unless they receive approval for doing so from the Editor-In-Chief.
IJISAE open access articles are licensed under a Creative Commons Attribution-ShareAlike 4.0 International License. This license lets the audience to give appropriate credit, provide a link to the license, and indicate if changes were made and if they remix, transform, or build upon the material, they must distribute contributions under the same license as the original.
