article · Scientific Reports
Monitoring abnormal human behaviour such as assault and web-based violence is increasingly critical for preventing harm. While deep learning offers promise for processing large volumes of data, categorising aberrant activities remains a challenging task. A deep-learning framework has been developed that combines convolutional neural networks, bidirectional long short-term memory networks, and an attention mechanism to process the spatiotemporal features of raw video streams. This system analyses video input to identify and reliably categorise anomalous actions. Experimental evaluations against benchmark datasets demonstrate that the architecture achieves classification accuracies of 98.9 percent on UCF11, 96.04 percent on UCF50, and 61.04 percent on the subUCF crime dataset. These results show strong performance in standard activity detection alongside significant challenges when categorising complex crime scenarios.
Automatically identifying abnormal behaviour in video feeds can assist in mitigating physical violence and online harm such as hate crimes. By effectively capturing both spatial and temporal features in video data, automated systems can help detect harmful activities more reliably than previous methods, although complex crime detection still presents notable performance hurdles.
The framework could enable automated video surveillance and content monitoring tools to flag assault and violent incidents. Potential users include security operators and digital platforms seeking to identify harmful behaviour. The research represents applied, laboratory-tested technology validated on standard benchmark datasets, though the lower accuracy of 61.04 percent on crime data indicates that it remains at an early developmental stage before practical operational deployment.
AI-generated from the published abstract. Always read the original work before citing.
Abnormal human behavior must be monitored and controlled in today's technology-driven era, since it may cause damage to society in the form of assault or web-based violence, such as direct harm to a person or the propagation of hate crimes through the internet. Several authors have attempted to address this issue, but no one has yet come up with a solution that is both practical and workable. Recently, deep learning models have become popular as a means of handling massive amounts of data but their potential to categorize the aberrant human activity remains unexplored. Using a convolutional neural network (CNN), a bidirectional long short-term memory (Bi-LSTM), and an attention mechanism to pay attention to the unique spatiotemporal characteristics of raw video streams, a deep-learning approach has been implemented in the proposed framework to detect anomalous human activity. After analyzing the video, our suggested architecture can reliably assign an abnormal human behavior to its designated category. Analytic findings comparing the suggested architecture to state-of-the-art algorithms reveal an accuracy of 98.9%, 96.04%, and 61.04% using the UCF11, UCF50, and subUCF crime datasets, respectively.
This page summarises published work. The authoritative version sits with the publisher.
DOI: 10.1038/s41598-023-41231-0
Is something wrong with this record? Report it or request removal.
Discussion
Have you built on this work, tried to replicate it, or seen it applied in practice? Share what you know. Verified researchers and MARATTO™ domain experts can open a discussion, and any member can reply. Contributions are reviewed before they appear.
No discussion yet. Open the first thread.
New to MARATTO™? Create a free account.
