Project: Pedestrian Activity Measurement in the Transport Environment (PERMEATE)

Reference: STP 14/6/8

Last update: 27/07/2004 10:30:51

Objectives

The goal is to produce a pedestrian activity monitoring system that can be applied, over the course of a day, to count individual people at various datum points, determine the number of people alighting from and embarking upon bus journeys, count people entering and leaving each shop, and attempt to follow people as they enter and leave a shop to determine the number and type of intermediate goals in a journey.

Description

The project will investigate passive infra-red sensor array technology for monitoring pedestrian activity.

Contractor(s)

InfraRed Integrated Systems Ltd
Towcaster Mill, Towcaster, Morthants, NN12 6AD
01327 357824

City of Edinburgh Council (CEC)
Department of City Development, 1 Cockburn Street, Edinburgh, EH1 12W
0131 529 3749

Contract details

Cost to the Department: £203,893.00

Actual start date: 09 April 2001

Actual completion date: 25 February 2004

Publication(s)

Pedestrian Activity Measurement in the Transport Environment (PERMEATE)
Author: Napier University and IRISYS
Publication date: 01/01/2004
Source: IRISYS, Towcester, UK.

Summary of results

  1. The PERMEATE project investigated the use of low cost infrared detectors in tracking the movement of people in a space that is covered by more than one such detector. We have extended the operational performance of a simple one-dimensional infrared people counting technology toward an advanced two-dimensional people tracking system. The great advantage of the infrared based tracking technology is that it is able to generate the trajectories in real time using only an ordinary PC or embedded processor. Video-based tracking techniques, by contrast, require considerably more analysis time and resource and can be subject to inaccuracies due to light fluctuation and variation.

    The goal of the project was to produce a system that achieved the following key objectives over an extended time period:

    1. Count individual people at various datum points within the two-dimensional space.
    2. Determine the number of people alighting from and embarking on a bus.
    3. Count the number of people entering and leaving each shop.
    4. Make a reasonable attempt to follow a person as they leave a shop and enter another one to determine the number and type of intermediate goals in a journey.

    In pursuit of the key objectives, the following research results were obtained:

    1. We can track individuals from the field of view of one detector to the adjacent detectors with >80% accuracy
    2. We can dynamically create datum points over which we can collect various statistics.
    3. We can operate the system over a continuous period of 24 hours.
    4. We have written data analysis software for pedestrian flow analysis and interpretation.


    The impact of this system on the monitoring and measurement of pedestrian activity promises to be significant. The information obtained can be used to derive accurate estimates of walking speeds, deviation distances around obstacles, and interpersonal distances. Along with transport analysis, the applications for this technology lie in areas such as: retail space instrumentation, analysing the shopping habits of consumers; traffic flow design, analysing the use of space and designing better layout of public places; and security and surveillance, alerting guards to movement exceptions.

Departmental Assessment Status: Assessed by FIT Programme Advisory Group 15 July 2004. Given score 8/10 - Very good project, achieved targets and objectives, good quality research work, outputs with commercial potential.