Automated Solutions for Crowd Size Estimation

Abstract

The crowd phenomenon frequently occurs in dense urban living environments. Crowd counting or estimation helps to develop management strategies such as designing safe public places and evacuation plan for emergencies. These strategies are different depending upon the type of event such as political and public demonstrations, sports, and religious events. However, estimating the number of people in crowds at closed or open environments is quite challenging because of the dynamics involved in the process. In addition, crowd estimation itself poses challenges due to randomness in crowd behavior, motion, and an area's geometric specifications. Crowd behavior as well as the area parameters is studied before suggesting any possible technological solution for managing a crowd. This article presents a theoretical understanding of the major crowd size estimation approaches that cannot be achieved through the study of existing survey papers in this area, because the existing survey papers focus on particular technologies/specific areas with no or brief description of the involved steps. Besides, this article also highlights the strength and weakness of crowd size estimation solutions and their possible applications. It is, therefore, believed that the provided information would assist in developing an intelligent system for crowd management.

Formal Specification Approach for Smart Space Development

“…With the rapidly increasing growth of the number of pilgrims coming to the Holy Mosque, there is always a mounting need of proposing and implementing new solutions for crowd control. Although the conventional methods for controlling the crowd such as human interventions can be improved by using the fast-growing Information and Communication Technologies, they are often unsuccessful and uncomfortable for the pilgrims [9]. Controlling and managing the ever-increasing occurrences of crowd poses several challenges to the authorities.…”

Automated Visual Content Analysis (AVCA) in Communication Research: A Protocol for Large Scale Image Classification with Pre-Trained Computer Vision Models

“…Art historians have used computer vision to identify analogies in large samples of art-related images (Rosado, 2017) or to retrieve artworks from an existing database of paintings (Lang & Ommer, 2018). In the social sciences more broadly, for example, a study has employed computer vision to estimate the size and dynamics of human crowds (Aziz et al, 2018). In psychology, Reece and Danforth (2017) applied computer vision APIs to analyze over 43,000 Instagram images for features that may predict depression in users.…”

Sensing Thousands of Visitors Using Radio Frequency

“…Estimating the number of people present in a particular environment is critical to ensure crowd safety [1]. Many tragic cases exist of fatalities caused by mass panic breaking out in large crowds [2].…”