This document describes a study that developed a deep learning model to detect weapons like pistols and knives in CCTV footage. The researchers collected datasets from various online sources and performed data augmentation to increase the size and variety of images. They used the YOLOv5 object detection model and trained it on the preprocessed datasets. The trained model was then evaluated on real CCTV footage to test its ability to accurately detect and classify weapons. The goal was to develop an automated system that can monitor CCTV footage and alert operators when dangerous weapons are detected, in order to enhance security.