Ανάπτυξη συστήματος εντοπισμού μικρών αντικειμένων με τεχνικές υπερ-ανάλυσης
Tiny object detection using super resolution techniques
View/
Keywords
Object detection ; Small object detection ; Super-resolution ; Convolutional Neural Networks (CNNs) ; Generative Adversarial Networks (GANs) ; ESRGAN (Enhanced Super-Resolution GAN)Abstract
This thesis focuses on the development of a small object detection system using super-resolution techniques to enhance the quality and resolution of images. The aim is to improve the accuracy of detecting small objects, which are often challenging to identify due to low resolution and limited features. By applying advanced super-resolution algorithms, such as ESRGAN (Enhanced Super-Resolution GAN), this research seeks to improve detail discrimination in images, achieving higher performance on object detection datasets, especially for small objects. The experimental results highlight the significance of super-resolution techniques in computer vision and open new avenues for applications in fields such as surveillance and autonomous driving
Postgraduate Studies Programme
Προηγμένα Συστήματα ΠληροφορικήςDepartment
Σχολή Τεχνολογιών Πληροφορικής και Επικοινωνιών. Τμήμα Ψηφιακών ΣυστημάτωνNumber of pages
81Language
GreekCollections
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