Algorithm for Tracking Multiple Moving Targets from a Video Feed Acquired by a Group of Drones
Adam B. Mazoya, Nunna Joseph Mlyahilu, Jacques B. Ngoua Ndong Avele, Mba Dit O. Medfranck · 2024
Object detection is a computer vision technique for detecting and locating instances of objects in images or videos. Computer vision enables computer and other interconnected systems to derive meaningful information from visual inputs. Besides, the main tasks of computer vision are image classification, object detection, semantic and instance segmentation. Moreover, special algorithms for object detection that have been developed are two-stage, one-stage object detection algorithms and transformer-based approaches.In this paper we implemented a one-stage object detection algorithm named YOLO-v7 framework on a video feed acquired using a swarm of drones. This algorithm has surpassed all known object detectors algorithm in both speed and accuracy in the range from 5FPS to 160FPS and the highest accuracy of 56.8% mean average precision among all known real time object detectors.