University of Twente Student Theses

Login

Crowd analysis on edge devices : a comparative study of neural networks on blurred images

Coste, S.F. (2023) Crowd analysis on edge devices : a comparative study of neural networks on blurred images.

[img] PDF
1MB
Abstract:Crowd counting is an important task in various applications such as public safety, traffic management, and surveillance. In recent years, edge devices have become increasingly popular for crowd counting due to their low cost and high efficiency. However, traditional image processing techniques for crowd counting often struggle to handle images taken in challenging environments, such as low light or crowded scenes. In this paper, we propose a new approach for crowd counting on edge devices using blurred images. The use of blurred images allows for crowd counting in challenging environments while also preserving privacy. This paper will compare the performance of different neural network architectures on a data set of blurred images of crowded scenes and evaluate their accuracy, robustness, and efficiency. We also examine the impact of different factors on the performance of our approach, such as image resolution, blur type and blur level. Our results show that crowd counting on edge devices using blurred images is a promising approach with potential applications in various fields such as crowd management in public events, retail, transportation, surveillance and security, public health, and emergency and disaster management.
Item Type:Essay (Bachelor)
Faculty:EEMCS: Electrical Engineering, Mathematics and Computer Science
Subject:54 computer science
Programme:Computer Science BSc (56964)
Link to this item:https://purl.utwente.nl/essays/94360
Export this item as:BibTeX
EndNote
HTML Citation
Reference Manager

 

Repository Staff Only: item control page