Home > Research > Publications & Outputs > Algorithms for the detection of chewing behavio...
View graph of relations

Algorithms for the detection of chewing behavior in dietary monitoring applications

Research output: Contribution to Journal/MagazineJournal articlepeer-review

Published

Standard

Algorithms for the detection of chewing behavior in dietary monitoring applications. / Schmalz, M.S.; Helal, Sumi; Mendez-Vasquez, A.

In: Proceedings of SPIE, Vol. 7444, 2009.

Research output: Contribution to Journal/MagazineJournal articlepeer-review

Harvard

APA

Vancouver

Schmalz MS, Helal S, Mendez-Vasquez A. Algorithms for the detection of chewing behavior in dietary monitoring applications. Proceedings of SPIE. 2009;7444. doi: 10.1117/12.829205

Author

Schmalz, M.S. ; Helal, Sumi ; Mendez-Vasquez, A. / Algorithms for the detection of chewing behavior in dietary monitoring applications. In: Proceedings of SPIE. 2009 ; Vol. 7444.

Bibtex

@article{292622c5e0b8412193aff1af7b246575,
title = "Algorithms for the detection of chewing behavior in dietary monitoring applications",
abstract = "The detection of food consumption is key to the implementation of successful behavior modification in support of dietary monitoring and therapy, for example, during the course of controlling obesity, diabetes, or cardiovascular disease. Since the vast majority of humans consume food via mastication (chewing), we have designed an algorithm that automatically detects chewing behaviors in surveillance video of a person eating. Our algorithm first detects the mouth region, then computes the spatiotemporal frequency spectrum of a small perioral region (including the mouth). Spectral data are analyzed to determine the presence of periodic motion that characterizes chewing. A classifier is then applied to discriminate different types of chewing behaviors. Our algorithm was tested on seven volunteers, whose behaviors included chewing with mouth open, chewing with mouth closed, talking, static face presentation (control case), and moving face presentation. Early test results show that the chewing behaviors induce a temporal frequency peak at 0.5Hz to 2.5Hz, which is readily detected using a distance-based classifier. Computational cost is analyzed for implementation on embedded processing nodes, for example, in a healthcare sensor network. Complexity analysis emphasizes the relationship between the work and space estimates of the algorithm, and its estimated error. It is shown that chewing detection is possible within a computationally efficient, accurate, and subject-independent framework. {\textcopyright} 2009 SPIE.",
keywords = "Behavior detection, Computer vision, Dietary monitoring, Pattern recognition, Cardio-vascular disease, Complexity analysis, Computational costs, Computationally efficient, Distance-based, Embedded processing, Estimated error, Food consumption, Monitoring applications, Moving faces, Periodic motion, Spatiotemporal frequency, Spectral data, Surveillance video, Temporal frequency, Test results, Algorithms, Classifiers, Data processing, Food supply, Learning systems, Security systems, Sensor networks, Signal detection, Spectroscopy, Computational efficiency",
author = "M.S. Schmalz and Sumi Helal and A. Mendez-Vasquez",
year = "2009",
doi = "10.1117/12.829205",
language = "English",
volume = "7444",
journal = "Proceedings of SPIE",
issn = "0277-786X",
publisher = "SPIE",

}

RIS

TY - JOUR

T1 - Algorithms for the detection of chewing behavior in dietary monitoring applications

AU - Schmalz, M.S.

AU - Helal, Sumi

AU - Mendez-Vasquez, A.

PY - 2009

Y1 - 2009

N2 - The detection of food consumption is key to the implementation of successful behavior modification in support of dietary monitoring and therapy, for example, during the course of controlling obesity, diabetes, or cardiovascular disease. Since the vast majority of humans consume food via mastication (chewing), we have designed an algorithm that automatically detects chewing behaviors in surveillance video of a person eating. Our algorithm first detects the mouth region, then computes the spatiotemporal frequency spectrum of a small perioral region (including the mouth). Spectral data are analyzed to determine the presence of periodic motion that characterizes chewing. A classifier is then applied to discriminate different types of chewing behaviors. Our algorithm was tested on seven volunteers, whose behaviors included chewing with mouth open, chewing with mouth closed, talking, static face presentation (control case), and moving face presentation. Early test results show that the chewing behaviors induce a temporal frequency peak at 0.5Hz to 2.5Hz, which is readily detected using a distance-based classifier. Computational cost is analyzed for implementation on embedded processing nodes, for example, in a healthcare sensor network. Complexity analysis emphasizes the relationship between the work and space estimates of the algorithm, and its estimated error. It is shown that chewing detection is possible within a computationally efficient, accurate, and subject-independent framework. © 2009 SPIE.

AB - The detection of food consumption is key to the implementation of successful behavior modification in support of dietary monitoring and therapy, for example, during the course of controlling obesity, diabetes, or cardiovascular disease. Since the vast majority of humans consume food via mastication (chewing), we have designed an algorithm that automatically detects chewing behaviors in surveillance video of a person eating. Our algorithm first detects the mouth region, then computes the spatiotemporal frequency spectrum of a small perioral region (including the mouth). Spectral data are analyzed to determine the presence of periodic motion that characterizes chewing. A classifier is then applied to discriminate different types of chewing behaviors. Our algorithm was tested on seven volunteers, whose behaviors included chewing with mouth open, chewing with mouth closed, talking, static face presentation (control case), and moving face presentation. Early test results show that the chewing behaviors induce a temporal frequency peak at 0.5Hz to 2.5Hz, which is readily detected using a distance-based classifier. Computational cost is analyzed for implementation on embedded processing nodes, for example, in a healthcare sensor network. Complexity analysis emphasizes the relationship between the work and space estimates of the algorithm, and its estimated error. It is shown that chewing detection is possible within a computationally efficient, accurate, and subject-independent framework. © 2009 SPIE.

KW - Behavior detection

KW - Computer vision

KW - Dietary monitoring

KW - Pattern recognition

KW - Cardio-vascular disease

KW - Complexity analysis

KW - Computational costs

KW - Computationally efficient

KW - Distance-based

KW - Embedded processing

KW - Estimated error

KW - Food consumption

KW - Monitoring applications

KW - Moving faces

KW - Periodic motion

KW - Spatiotemporal frequency

KW - Spectral data

KW - Surveillance video

KW - Temporal frequency

KW - Test results

KW - Algorithms

KW - Classifiers

KW - Data processing

KW - Food supply

KW - Learning systems

KW - Security systems

KW - Sensor networks

KW - Signal detection

KW - Spectroscopy

KW - Computational efficiency

U2 - 10.1117/12.829205

DO - 10.1117/12.829205

M3 - Journal article

VL - 7444

JO - Proceedings of SPIE

JF - Proceedings of SPIE

SN - 0277-786X

ER -