Detecting Foreground Disambiguation of Depth Images using Fuzzy Logic

Tanvi Banerjee; James M. Keller; Marjorie Skubic

doi:10.1109/FUZZ-IEEE.2013.6622364

Detecting Foreground Disambiguation of Depth Images using Fuzzy Logic

Tanvi Banerjee, James M. Keller, Marjorie Skubic

Computer Science and Engineering

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution

Abstract

We present a unique occlusion and foreground overlap detection technique from depth sensor data using a fuzzy rule-based system. Features such as bounding box parameters and skeletonization were extracted from the foreground images and then input to the Fuzzy Inference System. Overlap and occlusion confidence measures were taken for each frame in the image sequence and compared against the extracted ground truth. This technique can help filter out occluded regions in the image sequence which, in an Eldercare environment, can then be used to compute accurate estimates of fall risk parameters such as stride time, stride length, and walking speed on a daily basis in in order to monitor the well-being of older adults in an ambient assisted living facility.

Original language	American English
Title of host publication	FUZZ-IEEE 2013 - 2013 IEEE International Conference on Fuzzy Systems
Publisher	IEEE
ISBN (Electronic)	978-1-4799-0022-0, 978-1-4799-0020-6
DOIs	https://doi.org/10.1109/FUZZ-IEEE.2013.6622364
State	Published - Oct 7 2013
Event	2013 IEEE International Conference on Fuzzy Systems, FUZZ-IEEE 2013 - Hyderabad, India Duration: Jul 7 2013 → Jul 10 2013

Conference

Conference	2013 IEEE International Conference on Fuzzy Systems, FUZZ-IEEE 2013
Country/Territory	India
City	Hyderabad
Period	7/7/13 → 7/10/13

ASJC Scopus Subject Areas

Software
Theoretical Computer Science
Artificial Intelligence
Applied Mathematics

Keywords

Activity analysis
Depth image
Fuzzy rules
Machine learning
Occlusion

Disciplines

Bioinformatics
Communication Technology and New Media
Databases and Information Systems
OS and Networks
Science and Technology Studies

Access to Document

10.1109/FUZZ-IEEE.2013.6622364License: Unspecified

Cite this

@inproceedings{cb6e5712b4bd41808ebcca47174e54a0,

title = "Detecting Foreground Disambiguation of Depth Images using Fuzzy Logic",

abstract = " We present a unique occlusion and foreground overlap detection technique from depth sensor data using a fuzzy rule-based system. Features such as bounding box parameters and skeletonization were extracted from the foreground images and then input to the Fuzzy Inference System. Overlap and occlusion confidence measures were taken for each frame in the image sequence and compared against the extracted ground truth. This technique can help filter out occluded regions in the image sequence which, in an Eldercare environment, can then be used to compute accurate estimates of fall risk parameters such as stride time, stride length, and walking speed on a daily basis in in order to monitor the well-being of older adults in an ambient assisted living facility.",

keywords = "Activity analysis, Depth image, Fuzzy rules, Machine learning, Occlusion",

author = "Tanvi Banerjee and Keller, {James M.} and Marjorie Skubic",

year = "2013",

month = oct,

day = "7",

doi = "10.1109/FUZZ-IEEE.2013.6622364",

language = "American English",

booktitle = "FUZZ-IEEE 2013 - 2013 IEEE International Conference on Fuzzy Systems",

publisher = "IEEE",

note = "2013 IEEE International Conference on Fuzzy Systems, FUZZ-IEEE 2013 ; Conference date: 07-07-2013 Through 10-07-2013",

}

TY - GEN

T1 - Detecting Foreground Disambiguation of Depth Images using Fuzzy Logic

AU - Banerjee, Tanvi

AU - Keller, James M.

AU - Skubic, Marjorie

PY - 2013/10/7

Y1 - 2013/10/7

N2 - We present a unique occlusion and foreground overlap detection technique from depth sensor data using a fuzzy rule-based system. Features such as bounding box parameters and skeletonization were extracted from the foreground images and then input to the Fuzzy Inference System. Overlap and occlusion confidence measures were taken for each frame in the image sequence and compared against the extracted ground truth. This technique can help filter out occluded regions in the image sequence which, in an Eldercare environment, can then be used to compute accurate estimates of fall risk parameters such as stride time, stride length, and walking speed on a daily basis in in order to monitor the well-being of older adults in an ambient assisted living facility.

AB - We present a unique occlusion and foreground overlap detection technique from depth sensor data using a fuzzy rule-based system. Features such as bounding box parameters and skeletonization were extracted from the foreground images and then input to the Fuzzy Inference System. Overlap and occlusion confidence measures were taken for each frame in the image sequence and compared against the extracted ground truth. This technique can help filter out occluded regions in the image sequence which, in an Eldercare environment, can then be used to compute accurate estimates of fall risk parameters such as stride time, stride length, and walking speed on a daily basis in in order to monitor the well-being of older adults in an ambient assisted living facility.

KW - Activity analysis

KW - Depth image

KW - Fuzzy rules

KW - Machine learning

KW - Occlusion

UR - http://www.scopus.com/inward/record.url?scp=84887840690&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84887840690&partnerID=8YFLogxK

UR - https://corescholar.libraries.wright.edu/knoesis/1109

U2 - 10.1109/FUZZ-IEEE.2013.6622364

DO - 10.1109/FUZZ-IEEE.2013.6622364

M3 - Conference contribution

BT - FUZZ-IEEE 2013 - 2013 IEEE International Conference on Fuzzy Systems

PB - IEEE

T2 - 2013 IEEE International Conference on Fuzzy Systems, FUZZ-IEEE 2013

Y2 - 7 July 2013 through 10 July 2013

ER -

Detecting Foreground Disambiguation of Depth Images using Fuzzy Logic

Abstract

Conference

ASJC Scopus Subject Areas

Keywords

Disciplines

Access to Document

Other files and links

Cite this