Foreground segmentation-based human detection with shadow removal

Fadhlan Hafiz, A. A. Shafie, Othman Khalifa, M. H. Ali

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

9 Citations (Scopus)

Abstract

Recent research in video surveillance system has shown an increasing focus on creating reliable systems utilizing non-computationally expensive technique for observing humans' appearance, movements and activities, thus providing analytical information for advanced human behaviour analysis and realistic human modelling. In order for the system to function, it requires robust method for detecting human form from a given input of video streams. In this paper, we present a human detection technique suitable for video surveillance. The technique we propose includes background subtraction, foreground segmentation, and shadow removal. The proposed detection technique will first try to extract all foreground objects from the background and then moving shadows will be eliminated by a shadow detection algorithm. Finally, we perform a morphological reconstruction algorithm to recover the distorted foreground objects after shadow removal process. We define certain features that describe human and match them with the final objects obtained from earlier processing. The experimental result proves its validity and accuracy in various fixed outdoor and indoor video scenes.

Original languageEnglish
Title of host publicationInternational Conference on Computer and Communication Engineering, ICCCE'10
DOIs
Publication statusPublished - 2010
Externally publishedYes
EventInternational Conference on Computer and Communication Engineering, ICCCE'10 - Kuala Lumpur, Malaysia
Duration: May 11 2010May 12 2010

Other

OtherInternational Conference on Computer and Communication Engineering, ICCCE'10
CountryMalaysia
CityKuala Lumpur
Period5/11/105/12/10

Fingerprint

Processing

Keywords

  • Background subtraction
  • Foreground segmentation
  • Gaussian mixture model
  • Human detection
  • Shadow removal
  • Video surveillance

ASJC Scopus subject areas

  • Computer Networks and Communications

Cite this

Hafiz, F., Shafie, A. A., Khalifa, O., & Ali, M. H. (2010). Foreground segmentation-based human detection with shadow removal. In International Conference on Computer and Communication Engineering, ICCCE'10 [5556763] https://doi.org/10.1109/ICCCE.2010.5556763

Foreground segmentation-based human detection with shadow removal. / Hafiz, Fadhlan; Shafie, A. A.; Khalifa, Othman; Ali, M. H.

International Conference on Computer and Communication Engineering, ICCCE'10. 2010. 5556763.

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

Hafiz, F, Shafie, AA, Khalifa, O & Ali, MH 2010, Foreground segmentation-based human detection with shadow removal. in International Conference on Computer and Communication Engineering, ICCCE'10., 5556763, International Conference on Computer and Communication Engineering, ICCCE'10, Kuala Lumpur, Malaysia, 5/11/10. https://doi.org/10.1109/ICCCE.2010.5556763
Hafiz F, Shafie AA, Khalifa O, Ali MH. Foreground segmentation-based human detection with shadow removal. In International Conference on Computer and Communication Engineering, ICCCE'10. 2010. 5556763 https://doi.org/10.1109/ICCCE.2010.5556763
Hafiz, Fadhlan ; Shafie, A. A. ; Khalifa, Othman ; Ali, M. H. / Foreground segmentation-based human detection with shadow removal. International Conference on Computer and Communication Engineering, ICCCE'10. 2010.
@inproceedings{fd0fb666a93840e981d7c477e9147446,
title = "Foreground segmentation-based human detection with shadow removal",
abstract = "Recent research in video surveillance system has shown an increasing focus on creating reliable systems utilizing non-computationally expensive technique for observing humans' appearance, movements and activities, thus providing analytical information for advanced human behaviour analysis and realistic human modelling. In order for the system to function, it requires robust method for detecting human form from a given input of video streams. In this paper, we present a human detection technique suitable for video surveillance. The technique we propose includes background subtraction, foreground segmentation, and shadow removal. The proposed detection technique will first try to extract all foreground objects from the background and then moving shadows will be eliminated by a shadow detection algorithm. Finally, we perform a morphological reconstruction algorithm to recover the distorted foreground objects after shadow removal process. We define certain features that describe human and match them with the final objects obtained from earlier processing. The experimental result proves its validity and accuracy in various fixed outdoor and indoor video scenes.",
keywords = "Background subtraction, Foreground segmentation, Gaussian mixture model, Human detection, Shadow removal, Video surveillance",
author = "Fadhlan Hafiz and Shafie, {A. A.} and Othman Khalifa and Ali, {M. H.}",
year = "2010",
doi = "10.1109/ICCCE.2010.5556763",
language = "English",
isbn = "9781424462346",
booktitle = "International Conference on Computer and Communication Engineering, ICCCE'10",

}

TY - GEN

T1 - Foreground segmentation-based human detection with shadow removal

AU - Hafiz, Fadhlan

AU - Shafie, A. A.

AU - Khalifa, Othman

AU - Ali, M. H.

PY - 2010

Y1 - 2010

N2 - Recent research in video surveillance system has shown an increasing focus on creating reliable systems utilizing non-computationally expensive technique for observing humans' appearance, movements and activities, thus providing analytical information for advanced human behaviour analysis and realistic human modelling. In order for the system to function, it requires robust method for detecting human form from a given input of video streams. In this paper, we present a human detection technique suitable for video surveillance. The technique we propose includes background subtraction, foreground segmentation, and shadow removal. The proposed detection technique will first try to extract all foreground objects from the background and then moving shadows will be eliminated by a shadow detection algorithm. Finally, we perform a morphological reconstruction algorithm to recover the distorted foreground objects after shadow removal process. We define certain features that describe human and match them with the final objects obtained from earlier processing. The experimental result proves its validity and accuracy in various fixed outdoor and indoor video scenes.

AB - Recent research in video surveillance system has shown an increasing focus on creating reliable systems utilizing non-computationally expensive technique for observing humans' appearance, movements and activities, thus providing analytical information for advanced human behaviour analysis and realistic human modelling. In order for the system to function, it requires robust method for detecting human form from a given input of video streams. In this paper, we present a human detection technique suitable for video surveillance. The technique we propose includes background subtraction, foreground segmentation, and shadow removal. The proposed detection technique will first try to extract all foreground objects from the background and then moving shadows will be eliminated by a shadow detection algorithm. Finally, we perform a morphological reconstruction algorithm to recover the distorted foreground objects after shadow removal process. We define certain features that describe human and match them with the final objects obtained from earlier processing. The experimental result proves its validity and accuracy in various fixed outdoor and indoor video scenes.

KW - Background subtraction

KW - Foreground segmentation

KW - Gaussian mixture model

KW - Human detection

KW - Shadow removal

KW - Video surveillance

UR - http://www.scopus.com/inward/record.url?scp=77957772038&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=77957772038&partnerID=8YFLogxK

U2 - 10.1109/ICCCE.2010.5556763

DO - 10.1109/ICCCE.2010.5556763

M3 - Conference contribution

SN - 9781424462346

BT - International Conference on Computer and Communication Engineering, ICCCE'10

ER -