This paper presents a weapon detection and email alert system that utilizes the YOLOv5 deep learning architecture and a custom dataset of pistol images. The system is designed to detect the presence of pistols in real-time video streams and send an email alert to the administrator in the event of a positive detection. The custom pistol dataset was created to train the YOLOv5 model and improve its ability to accurately detect pistols in a variety of environments and conditions. The system was tested on various videos and was found to achieve high accuracy in detecting pistols with low false positive rates. The email alert feature ensures that the administrator is immediately notified in case of weapon detection, allowing for quick and effective response. This system has the potential to be integrated into various settings such as schools, public spaces, and security systems to enhance security and prevent weapons-related incidents.