• ANEESHA FATHIMA Z
• AMRIN THABSUM S
• HARINI G G
• JOSPHIN MARY S
• Mrs.Barveen

The exponential rise of user-generated video content on social media has led to increased exposure to violent scenes,which can negatively impact viewers and public perception.Manual moderation is inefficient and inadequate at handling the growing volume of content. This paper presents a real-time violence detection framework powered by the YOLOv11 algorithm combined with Convolutional Neural Networks (CNNs), capable of identifying aggressive actions such as punches, kicks, and weapon usage by analyzing objects, motion patterns, and contextual cues. The system utilizes YOLOv11 for fast and accurate object detection and CNNs for enhanced classification of violent behavior.Evaluated on datasets from Roboflow and real-world scenarios,the model achieved high accuracy even under challenging conditions like low lighting and crowding. With integrated alert generation and scalability for smart surveillance systems, the proposed model shows promise for use in content moderation, law enforcement, and public safety applications.

Violence Detection, YOLOv11, Deep Learning, Real-Time Surveillance, Social Media, CNN, Object Detection, Content Moderation.

"DEEP LEARNING-BASED VIOLENCE DETECTION IN REAL TIME VIDEOS USING YOLO", International Journal of Emerging Technologies and Innovative Research (www.jetir.org), ISSN:2349-5162, Vol.12, Issue 5, page no.h65-h73, May-2025, Available :http://www.jetir.org/papers/JETIR2505809.pdf

"DEEP LEARNING-BASED VIOLENCE DETECTION IN REAL TIME VIDEOS USING YOLO", International Journal of Emerging Technologies and Innovative Research (www.jetir.org | UGC and issn Approved), ISSN:2349-5162, Vol.12, Issue 5, page no. pph65-h73, May-2025, Available at : http://www.jetir.org/papers/JETIR2505809.pdf