Implementation of Convolutional Neural Networks (CNN) for Crowd Counting in Shopping Mall Environments

Prihandoko, P
Wulandari, Natasya
Eska, Juna

IJISTECH
Vol 8, No 4 (2024): The December edition

Accurate crowd counting is crucial in public spaces such as shopping malls to ensure safety and optimize resource management. This article explores the use of Convolutional Neural Networks (CNN), specifically a modified VGG16 architecture, for real-time crowd counting in shopping mall environments. Using a dataset collected from various crowd scenarios, the model was trained and tested using evaluation metrics such as Mean Absolute Error (MAE) and Root Mean Squared Error (RMSE). The results indicate that the proposed model is effective, achieving higher accuracy compared to traditional methods, thanks to advanced feature extraction techniques. This research offers a robust and scalable solution to enhance security and improve crowd management in commercial spaces.