A Fast Hybrid Deep Neural Network Model for pushing behavior detection in human crowds

被引：2

作者：

Alia, Ahmed ^{[1
,2
,3
]}

Maree, Mohammed ^{[4
]}

Chraibi, Mohcine ^{[1
]}

机构：

[1] Forschungszentrum Julich, Inst Adv Simulat, D-52425 Julich, Germany

[2] Univ Wuppertal, Dept Comp Simulat Fire Protect & Pedestrian Traff, D-42285 Wuppertal, Germany

[3] An Najah Natl Univ, Dept Management Informat Syst, Nablus, Palestine

[4] Arab Amer Univ, Dept Informat Technol, Lenin, Palestine

来源：

2022 IEEE/ACS 19TH INTERNATIONAL CONFERENCE ON COMPUTER SYSTEMS AND APPLICATIONS (AICCSA) | 2022年

关键词：

Artificial Intelligence; Deep Neural Network; EfficientNetB1; Convolutional Neural Network; Crowd Behavior Analysis; Pushing Forward Motion Detection;

D O I：

10.1109/AICCSA56895.2022.10017883

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

Deep learning technology is regarded as one of the latest advances in data science and analytics due to its learning abilities from the data [1]. As a result, deep learning is widely applied in the human crowd analysis domain [2]. Although it has achieved remarkable success in this area, a fast and robust model for pushing behavior detection in the human crowd is unavailable. This paper proposes a model that allows crowd-monitoring systems to detect pushing behavior early, helping organizers make timely decisions before dangerous situations appear. This particularly becomes more challenging when applied to real-time video streams of crowded events, which the proposed model accomplishes with reasonable time latency. To achieve this, the model employs a hybrid deep neural network.

引用

页数：2

共 11 条

[1] Crowds in front of bottlenecks at entrances from the perspective of physics and social psychology [J].