Abstract
Crowd counting is of significant importance for numerous applications, e.g., urban security, intelligent surveillance and crowd management. Existing crowd counting methods typically require specialized hardware deployment and strict operating conditions, thereby hindering their widespread application. To acquire a more effective crowd counting approach, a device-free counting method based on Channel Status Information (CSI) is proposed. The wavelet domain denoising is introduced to mitigate environment noise. Furthermore, the amplitude or phase covariance matrix is extracted as the eigenmatrix. Moreover, both the spatial diversity and frequency diversity are leveraged to improve detection robustness. At the same experimental environment, the accuracy of the proposed CSI-based method is compared with a renowned crowd counting one, i.e., Electronic Frog Eye: Counting Crowd Using WiFi (FCC). The experimental results reveal an accuracy improvement of 30% over FCC.
Highlights
In some overpopulated countries, the contradiction between the limited indoor space and large population is becoming increasingly prominent
The coefficient of phase variation is the ratio of the standard deviation to the mean of the Channel Status Information (CSI) phase, in which human movement is detected when the averaged ratio is set within a predefined confidence interval
This paper only showed the CSI is related to the crowd counting in indoor environment, without illustrating phase is more sensitive and explaining between the relationship CSI and the number of crowd
Summary
The contradiction between the limited indoor space and large population is becoming increasingly prominent. It is of vital significance to implement crowd counting in public places, e.g., libraries, museums, shopping malls and college classrooms, which are with limited resources and strong mobility. In the meantime, it is a crucial and challenging task to acquire the human traffic or accurately calculate the population in some particular circumstances. Numerous crowd counting approaches have emerged over the past decades, e.g., video-based recognition, infrared-based induction, and non-image-based localization. These methods all require specialized hardware deployment and strict operating conditions which hinder their wide deployment. It is impractical and expensive to distribute equipments to each individual in a public place, and is not feasible under an emergent event [4,5]
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
