Human Activity Recognition based on Deep Learning
Pretrained VGG-16, LSTM (Long Short-Term Memory)
Real-time security surveillance is a critical aspect of public safety and crime prevention. In this research project, we propose a solution for detecting fights in videos using computer vision algorithms and machine learning models. Our approach uses a deep learning model that leverages a pre-trained VGG 16 model followed by a Long Short-Term Memory (LSTM) layer to analyze video frames and identify potential fights based on changes in motion and object behavior. To improve the accuracy of the detection system, we trained our model on a large dataset of videos, which we preprocessed by extracting frames and applying data augmentation techniques. Our experimental results demonstrate that our approach achieves high accuracy in detecting fights in videos, which can be useful for enhancing public safety and security.
"Human Activity Recognition based on Deep Learning", JETNR - JOURNAL OF EMERGING TRENDS AND NOVEL RESEARCH (www.JETNR.org), ISSN:2984-9276, Vol.4, Issue 4, page no.a359-a366, April-2026, Available :https://rjpn.org/JETNR/papers/JETNR2604045.pdf
Volume 4
Issue 4,
April-2026
Pages : a359-a366
Paper Reg. ID: JETNR_233605
Published Paper Id: JETNR2604045
Downloads: 00042
Research Area: Science and Technology
Country: Harda, Madhya Pradesh , India
ISSN: 2984-9276 | IMPACT FACTOR: 9.87 Calculated By Google Scholar | ESTD YEAR: 2023
An International Scholarly Open Access Journal, Peer-Reviewed, Refereed Journal Impact Factor 9.87 Calculate by Google Scholar and Semantic Scholar | AI-Powered Research Tool, Multidisciplinary, Monthly, Multilanguage Journal Indexing in All Major Database & Metadata, Citation Generator
Publisher: RJPN (IJPublication) Janvi Wave
