CNN-Based Crowd Counting Through IoT: Application For Saudi Public Places

Abstract

Crowd counting in specific places has recently been considered as a significant contribution in many applications in terms of security and economic values. Recently, the Kingdom of Saudi Arabia has considered new ways and methods to diversify sources of income, where many non-traditional establishments in several fields have been initiated and put in place. However, controlling the number of visitors and participants to events and exhibitions has always been a challenge, as it has always been considered as an important success factor to any event. Smart public places approach is one of the inevitable directions of development in Saudi Arabia, where security, comfort, and safety of crowds is to be controlled and managed using machine learning techniques, more specifically, IoT-based crowd counting techniques. Such a technology will not only help in resolving security and safety problems, but also will play a significant role in reducing waiting time for visitors, by giving indicators, projections and advices on crowded places. In this paper, a mobile-based model is proposed for counting people in high and low crowded public places in Saudi Arabia under various scene conditions with no prior knowledge. The proposed model is built based on pre-trained convolutional neural network (CNN) called VGG-16 with some modifications on the last layer of the CNN to increase the efficiency of the training model. In addition to the improvement of efficiency, the proposed method accepts images of arbitrary sizes/scales as inputs. The applicability of the proposed method has been evaluated by incorporating IoT architecture, where surveillance cameras to be connected to the Internet to capture live pictures of different public places. To achieve this goal, New and special Saudi people dataset as well as some other existing dataset, have been produced and used to train the network. The result shows a significant improvement to the efficiency of the DCNN over the existing counting networks.