Master’s Thesis at the University of Basrah Explores the Development of Pedestrian Tracking Using Fuzzy Logic
A Master’s thesis at the College of Computer Science and Information Technology, University of Basrah, entitled Improving Multi-Object Tracking Using Artificial Intelligence Algorithms, investigated the development of multi-pedestrian tracking systems using fuzzy logic.
The thesis, presented by student Batool Baji Abboud, aimed to develop multi-object pedestrian tracking systems for crowded scenes and address detection-to-track association errors, particularly when pedestrians overlap, their paths intersect, or they are partially occluded, resulting in identity loss. The study employed a Mamdani-type fuzzy inference system to handle ambiguous association cases within the ByteTrack algorithm.
The results obtained using the MOT17 dataset demonstrated improvements in HOTA, AssA, IDF1, IDP, and AssPr metrics, along with reductions in identity switches and false positives.
Department of Media and Government Communication
