Time Handling for Real-time Progressive Activity Recognition

Research output: Chapter in Book/Report/Conference proceedingConference contribution

2 Citations (Scopus)

Abstract

In a dense sensor-based smart home (SH), a significant challenge is to segment the sensor data stream in real-time to continuously support progressive activity recognition (AR). In this paper, we evaluate an approach that supports the segmentation of the sensor data stream used for knowledge-driven AR. The approach is based on the notion of dynamically varied sliding time windows, where data segments are formally modeled as time windows and ontological reasoning used to infer the ongoing activities of daily living (ADLs). We then present an algorithm that supports perpetual, real-time activity recognition and provide an implementation of both the proposed approach and the ADL ontology it uses. For evaluation, we developed a synthetic data generator and generated a set of synthetic ADLs. In addition, we implemented a real-time activity recognition system and a simple simulator that plays back a synthetic ADL as if the sensors are activated in real-time. We evaluated the real-time activity recognition algorithm, and obtained 99.2% average recognition accuracy on the synthetic ADLs tested. The ability of the algorithm to discriminate sensors that are activated in error was evaluated for selected ADLs and impressive results obtained.
LanguageEnglish
Title of host publicationUnknown Host Publication
Pages37-44
Number of pages8
DOIs
Publication statusPublished - Sep 2011
EventThe 2011 international workshop on Situation activity & goal awareness, in conjunction with Ubiquitous Computing 2011 -
Duration: 1 Sep 2011 → …

Workshop

WorkshopThe 2011 international workshop on Situation activity & goal awareness, in conjunction with Ubiquitous Computing 2011
Period1/09/11 → …

Fingerprint

Sensors
Ontology
Simulators

Cite this

@inproceedings{4b1708be3dac4ab5a980487b3dfc2856,
title = "Time Handling for Real-time Progressive Activity Recognition",
abstract = "In a dense sensor-based smart home (SH), a significant challenge is to segment the sensor data stream in real-time to continuously support progressive activity recognition (AR). In this paper, we evaluate an approach that supports the segmentation of the sensor data stream used for knowledge-driven AR. The approach is based on the notion of dynamically varied sliding time windows, where data segments are formally modeled as time windows and ontological reasoning used to infer the ongoing activities of daily living (ADLs). We then present an algorithm that supports perpetual, real-time activity recognition and provide an implementation of both the proposed approach and the ADL ontology it uses. For evaluation, we developed a synthetic data generator and generated a set of synthetic ADLs. In addition, we implemented a real-time activity recognition system and a simple simulator that plays back a synthetic ADL as if the sensors are activated in real-time. We evaluated the real-time activity recognition algorithm, and obtained 99.2{\%} average recognition accuracy on the synthetic ADLs tested. The ability of the algorithm to discriminate sensors that are activated in error was evaluated for selected ADLs and impressive results obtained.",
author = "George Okeyo and Liming Chen and Hui Wang and Roy Sterritt",
year = "2011",
month = "9",
doi = "10.1145/2030045.2030056",
language = "English",
isbn = "978-1-4503-0926-4",
pages = "37--44",
booktitle = "Unknown Host Publication",

}

Okeyo, G, Chen, L, Wang, H & Sterritt, R 2011, Time Handling for Real-time Progressive Activity Recognition. in Unknown Host Publication. pp. 37-44, The 2011 international workshop on Situation activity & goal awareness, in conjunction with Ubiquitous Computing 2011, 1/09/11. https://doi.org/10.1145/2030045.2030056

Time Handling for Real-time Progressive Activity Recognition. / Okeyo, George; Chen, Liming; Wang, Hui; Sterritt, Roy.

Unknown Host Publication. 2011. p. 37-44.

Research output: Chapter in Book/Report/Conference proceedingConference contribution

TY - GEN

T1 - Time Handling for Real-time Progressive Activity Recognition

AU - Okeyo, George

AU - Chen, Liming

AU - Wang, Hui

AU - Sterritt, Roy

PY - 2011/9

Y1 - 2011/9

N2 - In a dense sensor-based smart home (SH), a significant challenge is to segment the sensor data stream in real-time to continuously support progressive activity recognition (AR). In this paper, we evaluate an approach that supports the segmentation of the sensor data stream used for knowledge-driven AR. The approach is based on the notion of dynamically varied sliding time windows, where data segments are formally modeled as time windows and ontological reasoning used to infer the ongoing activities of daily living (ADLs). We then present an algorithm that supports perpetual, real-time activity recognition and provide an implementation of both the proposed approach and the ADL ontology it uses. For evaluation, we developed a synthetic data generator and generated a set of synthetic ADLs. In addition, we implemented a real-time activity recognition system and a simple simulator that plays back a synthetic ADL as if the sensors are activated in real-time. We evaluated the real-time activity recognition algorithm, and obtained 99.2% average recognition accuracy on the synthetic ADLs tested. The ability of the algorithm to discriminate sensors that are activated in error was evaluated for selected ADLs and impressive results obtained.

AB - In a dense sensor-based smart home (SH), a significant challenge is to segment the sensor data stream in real-time to continuously support progressive activity recognition (AR). In this paper, we evaluate an approach that supports the segmentation of the sensor data stream used for knowledge-driven AR. The approach is based on the notion of dynamically varied sliding time windows, where data segments are formally modeled as time windows and ontological reasoning used to infer the ongoing activities of daily living (ADLs). We then present an algorithm that supports perpetual, real-time activity recognition and provide an implementation of both the proposed approach and the ADL ontology it uses. For evaluation, we developed a synthetic data generator and generated a set of synthetic ADLs. In addition, we implemented a real-time activity recognition system and a simple simulator that plays back a synthetic ADL as if the sensors are activated in real-time. We evaluated the real-time activity recognition algorithm, and obtained 99.2% average recognition accuracy on the synthetic ADLs tested. The ability of the algorithm to discriminate sensors that are activated in error was evaluated for selected ADLs and impressive results obtained.

U2 - 10.1145/2030045.2030056

DO - 10.1145/2030045.2030056

M3 - Conference contribution

SN - 978-1-4503-0926-4

SP - 37

EP - 44

BT - Unknown Host Publication

ER -