Ahmed Alia

Ahmed Alia

I am a researcher and a Ph.D. student at Forschungszentrum Jülich in collaboration with Wuppertal University. My research concentrates on developing intelligent systems for crowds data analysis, with a focus on detecting crowded spots and pushing behavior in large-scale events. The developed intelligent systems are mainly based on:

1. Deep learning algorithms.
2. Internet of things.
3. Cloud technology

First System: On the exploitation of GPS-based data for real-time visualization of pedestrian dynamics in open environments

This work proposed an efficient and accurate system for real-time acquiring, processing, and visualizing pedestrians’ dynamic behavior. Our goal in this context is to produce GPS-based heat maps that assist event organizers and visitors in dynamically find crowded spots using their smartphone devices.

The architecture of the proposed system.

Published paper

GitHub repo

Second System: A Hybrid Deep Learning and Visualization Framework for Pushing Behavior Detection in Pedestrian Dynamics

This work developed a hybrid deep learning and visualization framework that aims to assist researchers in automatically identifying pushing behavior in videos.

An example of a video for a crowded event entrance. Circles represent the ground truth of pushing behavior, while the red rectangles denote predicted pushing patches.

The architecture of the proposed system.

Published paper

GitHub repo

Third System: A Cloud-based Deep Learning System for Improving Crowd Safety at Event Entrances

This work developed a cloud-based deep-learning system for early pushing detection automatically in the live video stream of entrances at the patch level.

Research Interests:

• Data analysis.
• Deep learning.
• Machine learning.
• Computer vision.
• Intelligent systems.
• Crowd behavior analysis.

Others

1. Google Scholar
2. ResearchGate