2026-10-10T04:38:48Z https://api.figshare.com/v2/oai

oai:figshare.com:article/33977301 2026-09-23T17:53:59Z category_734 category_931 category_106 portal_5 item_type_3 month_year_09_2026

<p>Comparison with different variants.</p> Manoj Kumar (205700) Mantosh Biswas (16030751) Anoop Kumar Patel (18865480) Abhay Kumar (697860) Kumar Abhishek (848214) Ahamed Shafeeq B. M. (25099473) Biological Sciences not elsewhere classified Information Systems not elsewhere classified Science Policy shows actual situations recommended performs better make anomaly identification finding unusual activity challenging monitoring situations bidirectional long short unexpected human actions time video monitoring describe temporal relationships current surveillance systems simulated video datasets model performs effective lstm &# 8217 keep people secure extracts spatial data convolutional neural network spatial data world surveillance video sequences video frames temporal representations human beings datasets frequently xlink "> useful way ucf50 dataset transfer learning term memory still restricted smart gadgets results suggest life settings gets 95 crime dataset computers everywhere art methods <div><p>With smart gadgets and computers everywhere, real-time video monitoring is now necessary to keep people secure and lower the chance of unexpected human actions. However, current surveillance systems have a lot of trouble effectively spotting unusual actions in real life. Most of the advanced anomaly detection algorithms are trained and tested on synthetic or simulated video datasets. These datasets frequently do not show how complicated and different real-life settings can be. Because of this, their usefulness and capacity to be used in real life are still restricted. We propose a transfer learning-based hybrid deep framework for finding unusual activity in real-world surveillance. We test the model using raw video streams from a subset of the UCF-Crime dataset that shows actual situations of people doing things. First, a pre-trained DenseNet-201, a convolutional neural network, extracts spatial data from video frames that have already been processed. A Bidirectional Long Short-Term Memory (Bi-LSTM) network is used to describe temporal relationships between video sequences in a useful way. The spatial data and the Bi-LSTM’s temporal representations are then integrated to make anomaly identification more precise and dependable. This integrated spatiotemporal architecture identifies abnormal actions of human beings in challenging monitoring situations. The model that was recommended performs better than the state of art methods. It gets 95.04% accuracy on the UCF50 dataset and 62.04% on the untrimmed UCF-Crime dataset. These results suggest that the model performs effective in real-life scenarios.</p></div> 2026-09-23T17:53:44Z Dataset Dataset 10.1371/journal.pone.0355690.t004 https://figshare.com/articles/dataset/_p_Comparison_with_different_variants_p_/33977301 CC BY 4.0