Recognition of complex human behaviors in pool environment using foreground silhouette

How Lung Eng, Kar Ann Toh, Wei Yun Yau, Tuan Kiang Chiew

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

2 Citations (Scopus)

Abstract

This paper presents a vision system which allows real-time recognition of temporal swimming activities and the detection of drowning incident. Operating with a set of techniques, the developed system focuses on two fundamental issues: i) way to analyze temporal behavior and ii) way to incorporate expert knowledge. To perform the recognition of different behaviors, data fusion and Hidden Markov Model (HMM) techniques are implemented. A polynomial classifier is introduced to deal with noisy foreground descriptors caused by poor resolution and sensory noise. It addresses the nonlinear interactions among different dimensions of foreground descriptors while preserving the linear estimation property. HMM is used to model the state transition process that yields a simple and efficient probabilistic inference engine. This work reports the results of extensive on-site experiments carried out. The results demonstrate reasonably good performance yielded, specifically, in terms of false alarm rates and detection of genuine water crises.

Original languageEnglish
Title of host publicationAdvances in Visual Computing - First International Symposium, ISVC 2005, Proceedings
Pages371-379
Number of pages9
DOIs
Publication statusPublished - 2005 Dec 1
EventFirst International Symposium on Advances in Visual Computing, ISVC 2005 - Lake Tahoe, NV, United States
Duration: 2005 Dec 52005 Dec 7

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume3804 LNCS
ISSN (Print)0302-9743
ISSN (Electronic)1611-3349

Other

OtherFirst International Symposium on Advances in Visual Computing, ISVC 2005
CountryUnited States
CityLake Tahoe, NV
Period05/12/505/12/7

Fingerprint

Silhouette
Human Behavior
Hidden Markov models
Markov Model
Descriptors
Probabilistic Inference
Linear Estimation
Inference engines
Inference Engine
False Alarm Rate
Nonlinear Interaction
Data Fusion
Data fusion
State Transition
Vision System
Classifiers
Classifier
Polynomials
Real-time
Water

All Science Journal Classification (ASJC) codes

  • Theoretical Computer Science
  • Computer Science(all)

Cite this

Eng, H. L., Toh, K. A., Yau, W. Y., & Chiew, T. K. (2005). Recognition of complex human behaviors in pool environment using foreground silhouette. In Advances in Visual Computing - First International Symposium, ISVC 2005, Proceedings (pp. 371-379). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 3804 LNCS). https://doi.org/10.1007/11595755_45
Eng, How Lung ; Toh, Kar Ann ; Yau, Wei Yun ; Chiew, Tuan Kiang. / Recognition of complex human behaviors in pool environment using foreground silhouette. Advances in Visual Computing - First International Symposium, ISVC 2005, Proceedings. 2005. pp. 371-379 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).
@inproceedings{650d63c62e014e6d95765b66c85bb77c,
title = "Recognition of complex human behaviors in pool environment using foreground silhouette",
abstract = "This paper presents a vision system which allows real-time recognition of temporal swimming activities and the detection of drowning incident. Operating with a set of techniques, the developed system focuses on two fundamental issues: i) way to analyze temporal behavior and ii) way to incorporate expert knowledge. To perform the recognition of different behaviors, data fusion and Hidden Markov Model (HMM) techniques are implemented. A polynomial classifier is introduced to deal with noisy foreground descriptors caused by poor resolution and sensory noise. It addresses the nonlinear interactions among different dimensions of foreground descriptors while preserving the linear estimation property. HMM is used to model the state transition process that yields a simple and efficient probabilistic inference engine. This work reports the results of extensive on-site experiments carried out. The results demonstrate reasonably good performance yielded, specifically, in terms of false alarm rates and detection of genuine water crises.",
author = "Eng, {How Lung} and Toh, {Kar Ann} and Yau, {Wei Yun} and Chiew, {Tuan Kiang}",
year = "2005",
month = "12",
day = "1",
doi = "10.1007/11595755_45",
language = "English",
isbn = "3540307508",
series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
pages = "371--379",
booktitle = "Advances in Visual Computing - First International Symposium, ISVC 2005, Proceedings",

}

Eng, HL, Toh, KA, Yau, WY & Chiew, TK 2005, Recognition of complex human behaviors in pool environment using foreground silhouette. in Advances in Visual Computing - First International Symposium, ISVC 2005, Proceedings. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 3804 LNCS, pp. 371-379, First International Symposium on Advances in Visual Computing, ISVC 2005, Lake Tahoe, NV, United States, 05/12/5. https://doi.org/10.1007/11595755_45

Recognition of complex human behaviors in pool environment using foreground silhouette. / Eng, How Lung; Toh, Kar Ann; Yau, Wei Yun; Chiew, Tuan Kiang.

Advances in Visual Computing - First International Symposium, ISVC 2005, Proceedings. 2005. p. 371-379 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 3804 LNCS).

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

TY - GEN

T1 - Recognition of complex human behaviors in pool environment using foreground silhouette

AU - Eng, How Lung

AU - Toh, Kar Ann

AU - Yau, Wei Yun

AU - Chiew, Tuan Kiang

PY - 2005/12/1

Y1 - 2005/12/1

N2 - This paper presents a vision system which allows real-time recognition of temporal swimming activities and the detection of drowning incident. Operating with a set of techniques, the developed system focuses on two fundamental issues: i) way to analyze temporal behavior and ii) way to incorporate expert knowledge. To perform the recognition of different behaviors, data fusion and Hidden Markov Model (HMM) techniques are implemented. A polynomial classifier is introduced to deal with noisy foreground descriptors caused by poor resolution and sensory noise. It addresses the nonlinear interactions among different dimensions of foreground descriptors while preserving the linear estimation property. HMM is used to model the state transition process that yields a simple and efficient probabilistic inference engine. This work reports the results of extensive on-site experiments carried out. The results demonstrate reasonably good performance yielded, specifically, in terms of false alarm rates and detection of genuine water crises.

AB - This paper presents a vision system which allows real-time recognition of temporal swimming activities and the detection of drowning incident. Operating with a set of techniques, the developed system focuses on two fundamental issues: i) way to analyze temporal behavior and ii) way to incorporate expert knowledge. To perform the recognition of different behaviors, data fusion and Hidden Markov Model (HMM) techniques are implemented. A polynomial classifier is introduced to deal with noisy foreground descriptors caused by poor resolution and sensory noise. It addresses the nonlinear interactions among different dimensions of foreground descriptors while preserving the linear estimation property. HMM is used to model the state transition process that yields a simple and efficient probabilistic inference engine. This work reports the results of extensive on-site experiments carried out. The results demonstrate reasonably good performance yielded, specifically, in terms of false alarm rates and detection of genuine water crises.

UR - http://www.scopus.com/inward/record.url?scp=33744804016&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=33744804016&partnerID=8YFLogxK

U2 - 10.1007/11595755_45

DO - 10.1007/11595755_45

M3 - Conference contribution

SN - 3540307508

SN - 9783540307501

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 371

EP - 379

BT - Advances in Visual Computing - First International Symposium, ISVC 2005, Proceedings

ER -

Eng HL, Toh KA, Yau WY, Chiew TK. Recognition of complex human behaviors in pool environment using foreground silhouette. In Advances in Visual Computing - First International Symposium, ISVC 2005, Proceedings. 2005. p. 371-379. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). https://doi.org/10.1007/11595755_45