Track papers across labs, topics, and trends.
Search papers, labs, and topics across Lattice.
This study introduces a real-time object detection framework for surveillance systems aimed at enhancing public safety in urban environments by identifying weapons and blunt objects. Utilizing a custom dataset of 336 images of blunt objects alongside a publicly available dataset of 7,623 images of guns and knives, the authors trained a YOLOv8 model to achieve high recall and precision. The results demonstrate that extended training improves detection performance without overfitting, making the system viable for real-world applications in public safety monitoring.
Real-time detection of weapons in urban surveillance could drastically reduce response times and improve public safety outcomes.
Ensuring public safety in densely populated urban environments remains a critical challenge, necessitating the deployment of intelligent and automated video surveillance systems. Traditional surveillance approaches rely heavily on manual monitoring, which is inefficient and susceptible to human fatigue, delayed response, and observational errors. To overcome these limitations, this work presents a real-time object detection-based surveillance framework. The proposed system focuses on detecting guns, knives, and region-specific blunt objects commonly involved in violent activities in Indian surveillance scenarios. A key contribution of this work is the use of a custom-created dataset collected using a mobile camera, consisting of 336 labeled images of blunt objects such as iron rods, wooden sticks, and plastic rods. This dataset is combined with a publicly available dataset of 7,623 images of guns and knives, forming a consolidated dataset of 7,959 images across three classes: gun, knife, and blunt object. The combined dataset is used to train a YOLOv8-based object detection model for real-time performance. Experimental evaluation shows that increasing the training duration significantly improves recall and average precision for the blunt object class without signs of overfitting. Overall, the proposed framework achieves an effective balance between accuracy and efficiency, making it suitable for deployment in real-world surveillance environments such as campuses, public spaces, and transportation areas.