Deep learning and object detection
Βαθιά μάθηση και ανίχνευση αντικειμένου
View/
Keywords
Ανίχνευση αντικειμένου ; Τεχνητή νοημοσύνηAbstract
The rise of COVID-19 disease came up with new restrictions in everyday life, like quarantine, social distancing, wearing face masks and more. In the current diploma thesis object detection algorithms are applied on image and video data in order to detect if a person wears or not a face mask as a preventative measure against the spread of COVID-19 disease. More specifically, in this dissertation a comparative study was conducted between one-stage detectors YOLOv3, YOLOv4 and YOLOv5 on an RGB image dataset, which consists of 1694 images and 7791 labels. The performance of YOLOv5 model is the best one between the three models.
Postgraduate Studies Programme
Πληροφοριακά Συστήματα και ΥπηρεσίεςDepartment
Σχολή Τεχνολογιών Πληροφορικής και Επικοινωνιών. Τμήμα Ψηφιακών ΣυστημάτωνNumber of pages
40Language
EnglishCollections
Except where otherwise noted, this item's license is described as
Αναφορά Δημιουργού-Μη Εμπορική Χρήση-Όχι Παράγωγα Έργα 3.0 Ελλάδα
Αναφορά Δημιουργού-Μη Εμπορική Χρήση-Όχι Παράγωγα Έργα 3.0 Ελλάδα