PreCount: a predictive model for correcting real-time occupancy count data

Fisayo Caleb Sangogboye,Mikkel Baun Kjærgaard

doi:10.1186/s42162-018-0016-4

Fisayo Caleb Sangogboye, Mikkel Baun Kjærgaard

Open Access

https://doi.org/10.1186/s42162-018-0016-4

Copy DOI

Abstract

Sensing the number of people occupying a building in real-time facilitates a number of pervasive applications within the area of building energy optimization and adaptive control. To ascertain occupant counts, the adoption of camera-based sensors i.e. 3D stereo-vision and thermal cameras have grown significantly. However, camera-based sensors can only produce occupant counts with accumulating errors. Existing methods for correcting such errors can only correct erroneous count data at the end of the day and not in real-time. However, many applications depend on real-time corrected counts. In this paper, we present an algorithm named PreCount for accurately correcting raw counts in real-time. The core idea of PreCount is to learn error estimates from the past. We evaluated the accuracy of the PreCount algorithm using datasets from four buildings. Also, the Normalized Root Mean Squared Error was used to evaluate the performance of PreCount. Our evaluation results show that in real-time PreCount achieved a significantly lower Normalized Root Mean Squared Error compared to raw counts and other correction approach with a maximum error reduction of 68% when benchmarked with ground truth data. By presenting a more accurate algorithm for estimating occupant counts in real-time, we hope to enable buildings to better serve the actual number of people to improve both occupant comfort and energy efficiency.

Highlights

Estimating the number of people in commercial and public buildings with the aid of pervasive sensors is receiving increasing attention
We reviewed the accuracy of the methods for correcting occupant counts in the past and how PreCount leverages the accuracy of the probabilistic approach to accurately produce a real-time correction
The first two evaluation cases benchmark the overall performance of both correction methods with CCp dataset from the first three building and ground truth data from the last building respectively

Summary

Introduction

Estimating the number of people in commercial and public buildings with the aid of pervasive sensors is receiving increasing attention. This is because the permeation of pervasive computing have prospects in facilitating several building applications. Most commercial outlets are mandated by government laws to provide accurate estimates of occupant counts in real-time. Several approaches for simulating, monitoring and optimizing energy consumption in buildings including the model based approaches have become prominent and have received significant research attention (Arendt et al 2016). A similar research attention is currently given to the concept of demand response (DR) in commercial buildings for facilitating the control of energy demand (Kjærgaard et al 2016).

Methods

Results

Discussion

Conclusion