Drone Detection using YOLOV4 on Images and Videos

The paper covers the capstone project that I have been working on for last 5 months (June 2021 - October 2021).The Topic "Drone Detection using YOLOV4 on images and videos" speaks about training model using YOLOV4 algorithm resulting into detecting the drones in given images and videos for testing. It also speaks about how CCTV cameras could be operated to detect the drones and alert the concerned people. The idea behind this project to make CCTVs capable of detecting such kinds of drones. We could make use of OpenCV and develop a model to make it work accordingly. At this stage, we make use of yolov4 to generate a robust model to detect drones.The model could be made more robustly by adding up images and even by classifying different kinds of drones. It has got future scope because it could be even applied on video clips. Yolov4 offers the potential to produce accurate results if an optimised amount of data is provided to the model. Hence, this is just a start as it can be made more reliable and powerful by updating the model. The future would be more secure and safe by implementing such projects on a real-time basis with the help of such algorithms and technologies making it easier for people who keep gazing for hours at a computer screen to provide best safety.