July 21, 2025

Automated Detection of Violent Events in Video Streams Using Augmented Deep Learning with Pre-Trained VGG16

View Full Paper

Ask AI

Bookmark

Authors

Discussion

Automated detection identifies violent events in video streams, implying improved public safety through advanced image processing techniques.

The model demonstrates 99.86% training accuracy and 98.36% validation accuracy in detecting violent events.
Key metrics such as precision, recall, and F1-score all reached 0.98, indicating strong detection performance.
Utilizing transfer learning and a pre-trained VGG16 model, the approach is fine-tuned on over 8,000 video frames.
This methodology introduces potential for real-time violence detection in surveillance systems, enhancing public safety.

Saha et al. (2025) studied this question.

View Full Paper

Ask AI

Bookmark

July 21, 2025

View Full Paper

Ask AI

Bookmark

Authors

Automated detection identifies violent events in video streams, implying improved public safety through advanced image processing techniques.

The model demonstrates 99.86% training accuracy and 98.36% validation accuracy in detecting violent events.
Key metrics such as precision, recall, and F1-score all reached 0.98, indicating strong detection performance.
Utilizing transfer learning and a pre-trained VGG16 model, the approach is fine-tuned on over 8,000 video frames.
This methodology introduces potential for real-time violence detection in surveillance systems, enhancing public safety.

Saha et al. (2025) studied this question.

View Full Paper

Ask AI

Bookmark