Pedestrian counting with grid-based binary sensors based on Monte Carlo method

Shuto Fujii, Yoshiaki Taniguchi, Go Hasegawa, Morito Matsuoka

Research output: Contribution to journalArticlepeer-review

10 Citations (Scopus)

Abstract

In this paper, we propose a method for estimating the number of pedestrians walking in opposite directions, as in cases of a shopping street or a sidewalk in a downtown area. The proposed method utilizes a compound-eye sensor that is constructed by placing two binary sensors for the pedestrians' movement direction and multiple binary sensors for the vertical direction of the pedestrians' movement direction. A number of Monte Carlo simulations about the movement of pedestrians are conducted, and the output history of the compound-eye sensor is obtained in each simulation. The simulation scenario with a small difference of the output history of the compound-eye sensor is selected to estimate the number of pedestrians. Evaluation results show that in the field whose width is 8 [m] the relative error in the proposed method is the smallest by using 2×8 binary sensors.

Original languageEnglish
Article number299
Pages (from-to)1-10
Number of pages10
JournalSpringerPlus
Volume3
Issue number1
DOIs
Publication statusPublished - 2014
Externally publishedYes

Keywords

  • Binary sensor
  • Monte Carlo method
  • Pedestrian counting
  • Sensor network

ASJC Scopus subject areas

  • General

Fingerprint Dive into the research topics of 'Pedestrian counting with grid-based binary sensors based on Monte Carlo method'. Together they form a unique fingerprint.

Cite this