A robust multimedia surveillance system for people counting

Zeyad Q. H. Al-Zaydi, David Ndzi, Munirah L. Kamarudin, Ammar Zakaria, Ali Y. M. Shakaff

Research output: Contribution to journalArticle

Abstract

Closed circuit television cameras (CCTV) are widely used in monitoring. Most rely on human operators and controllers for detection and information gathering, e.g. the whether there is a person in the monitored area and how many people may be present. This paper presents an adaptive crowd counting system based on two algorithms that estimate the density of each pixel in each frame and use it as a basis for counting people. One algorithm uses scale-invariant feature transform (SIFT) features and clustering to represent pixels of frames (SIFT algorithm) and the other uses features from accelerated segment test (FAST) corner points with SIFT features (SIFT-FAST algorithm). Each algorithm is designed using a novel combination of pixel-wise, motion-region, grid map, background segmentation using Gaussian mixture model (GMM) and edge detection based on Canny algorithm. The Mall and University of California (UCSD) datasets have been used to evaluate the proposed algorithms. Results show that their average accuracies and processing times for each dataset are similar but their accuracies at frame level are different. A fusion technique is proposed and used to increase accuracy by combining the result of the algorithms at frame level. The mean deviation error and the mean absolute error for the two proposed algorithms are less than 0.1 and 3.1, respectively, for the Mall dataset and less than 0.07 and 1.9, respectively, for UCSD dataset.
Original languageEnglish
Pages (from-to)1
Number of pages28
JournalMultimedia Tools and Applications
Volume76
Issue number22
Early online date23 Nov 2016
DOIs
Publication statusPublished - 1 Nov 2017
Externally publishedYes

Fingerprint

Shopping centers
Pixels
Edge detection
Video cameras
Fusion reactions
Controllers
Networks (circuits)
Monitoring
Processing

Keywords

  • Crowd counting systems
  • Monitoring
  • CCTV cameras
  • Background segmentation

Cite this

Al-Zaydi, Zeyad Q. H. ; Ndzi, David ; Kamarudin, Munirah L. ; Zakaria, Ammar ; Shakaff, Ali Y. M. / A robust multimedia surveillance system for people counting. In: Multimedia Tools and Applications. 2017 ; Vol. 76, No. 22. pp. 1.
@article{dd0f498aaecf4711af4007039c9893ad,
title = "A robust multimedia surveillance system for people counting",
abstract = "Closed circuit television cameras (CCTV) are widely used in monitoring. Most rely on human operators and controllers for detection and information gathering, e.g. the whether there is a person in the monitored area and how many people may be present. This paper presents an adaptive crowd counting system based on two algorithms that estimate the density of each pixel in each frame and use it as a basis for counting people. One algorithm uses scale-invariant feature transform (SIFT) features and clustering to represent pixels of frames (SIFT algorithm) and the other uses features from accelerated segment test (FAST) corner points with SIFT features (SIFT-FAST algorithm). Each algorithm is designed using a novel combination of pixel-wise, motion-region, grid map, background segmentation using Gaussian mixture model (GMM) and edge detection based on Canny algorithm. The Mall and University of California (UCSD) datasets have been used to evaluate the proposed algorithms. Results show that their average accuracies and processing times for each dataset are similar but their accuracies at frame level are different. A fusion technique is proposed and used to increase accuracy by combining the result of the algorithms at frame level. The mean deviation error and the mean absolute error for the two proposed algorithms are less than 0.1 and 3.1, respectively, for the Mall dataset and less than 0.07 and 1.9, respectively, for UCSD dataset.",
keywords = "Crowd counting systems, Monitoring , CCTV cameras, Background segmentation",
author = "Al-Zaydi, {Zeyad Q. H.} and David Ndzi and Kamarudin, {Munirah L.} and Ammar Zakaria and Shakaff, {Ali Y. M.}",
year = "2017",
month = "11",
day = "1",
doi = "10.1007/s11042-016-4156-x",
language = "English",
volume = "76",
pages = "1",
journal = "Multimedia Tools and Applications",
issn = "1380-7501",
publisher = "Springer",
number = "22",

}

A robust multimedia surveillance system for people counting. / Al-Zaydi, Zeyad Q. H. ; Ndzi, David; Kamarudin, Munirah L.; Zakaria, Ammar ; Shakaff, Ali Y. M.

In: Multimedia Tools and Applications, Vol. 76, No. 22, 01.11.2017, p. 1.

Research output: Contribution to journalArticle

TY - JOUR

T1 - A robust multimedia surveillance system for people counting

AU - Al-Zaydi, Zeyad Q. H.

AU - Ndzi, David

AU - Kamarudin, Munirah L.

AU - Zakaria, Ammar

AU - Shakaff, Ali Y. M.

PY - 2017/11/1

Y1 - 2017/11/1

N2 - Closed circuit television cameras (CCTV) are widely used in monitoring. Most rely on human operators and controllers for detection and information gathering, e.g. the whether there is a person in the monitored area and how many people may be present. This paper presents an adaptive crowd counting system based on two algorithms that estimate the density of each pixel in each frame and use it as a basis for counting people. One algorithm uses scale-invariant feature transform (SIFT) features and clustering to represent pixels of frames (SIFT algorithm) and the other uses features from accelerated segment test (FAST) corner points with SIFT features (SIFT-FAST algorithm). Each algorithm is designed using a novel combination of pixel-wise, motion-region, grid map, background segmentation using Gaussian mixture model (GMM) and edge detection based on Canny algorithm. The Mall and University of California (UCSD) datasets have been used to evaluate the proposed algorithms. Results show that their average accuracies and processing times for each dataset are similar but their accuracies at frame level are different. A fusion technique is proposed and used to increase accuracy by combining the result of the algorithms at frame level. The mean deviation error and the mean absolute error for the two proposed algorithms are less than 0.1 and 3.1, respectively, for the Mall dataset and less than 0.07 and 1.9, respectively, for UCSD dataset.

AB - Closed circuit television cameras (CCTV) are widely used in monitoring. Most rely on human operators and controllers for detection and information gathering, e.g. the whether there is a person in the monitored area and how many people may be present. This paper presents an adaptive crowd counting system based on two algorithms that estimate the density of each pixel in each frame and use it as a basis for counting people. One algorithm uses scale-invariant feature transform (SIFT) features and clustering to represent pixels of frames (SIFT algorithm) and the other uses features from accelerated segment test (FAST) corner points with SIFT features (SIFT-FAST algorithm). Each algorithm is designed using a novel combination of pixel-wise, motion-region, grid map, background segmentation using Gaussian mixture model (GMM) and edge detection based on Canny algorithm. The Mall and University of California (UCSD) datasets have been used to evaluate the proposed algorithms. Results show that their average accuracies and processing times for each dataset are similar but their accuracies at frame level are different. A fusion technique is proposed and used to increase accuracy by combining the result of the algorithms at frame level. The mean deviation error and the mean absolute error for the two proposed algorithms are less than 0.1 and 3.1, respectively, for the Mall dataset and less than 0.07 and 1.9, respectively, for UCSD dataset.

KW - Crowd counting systems

KW - Monitoring

KW - CCTV cameras

KW - Background segmentation

U2 - 10.1007/s11042-016-4156-x

DO - 10.1007/s11042-016-4156-x

M3 - Article

VL - 76

SP - 1

JO - Multimedia Tools and Applications

JF - Multimedia Tools and Applications

SN - 1380-7501

IS - 22

ER -