Moving object detection via TV-L1 optical flow in fall-down videos

Nur Ayuni Mohamed, Mohd Asyraf Zulkifley

Research output: Contribution to journalArticle

1 Citation (Scopus)

Abstract

There is a growing demand for surveillance systems that can detect fall-down events because of the increased number of surveillance cameras being installed in many public indoor and outdoor locations. Fall-down event detection has been vigorously and extensively researched for safety purposes, particularly to monitor elderly peoples, patients, and toddlers. This computer vision detector has become more affordable with the development of high-speed computer networks and low-cost video cameras. This paper proposes moving object detection method based on human motion analysis for human fall-down events. The method comprises of three parts, which are preprocessing part to reduce image noises, motion detection part by using TV-L1 optical flow algorithm, and performance measure part. The last part will analyze the results of the object detection part in term of the bounding boxes, which are compared with the given ground truth. The proposed method is tested on Fall Down Detection (FDD) dataset and compared with Gunnar-Farneback optical flow by measuring intersection over union (IoU) of the output with respect to the ground truth bounding box. The experimental results show that the proposed method achieves an average IoU of 0.92524.

Original languageEnglish
Pages (from-to)839-846
Number of pages8
JournalBulletin of Electrical Engineering and Informatics
Volume8
Issue number3
DOIs
Publication statusPublished - 1 Sep 2019

Fingerprint

Moving Object Detection
Optical flows
Optical Flow
Video cameras
Computer networks
unions
ground truth
Surveillance
Computer vision
surveillance
Union
Intersection
Camera
Cameras
intersections
boxes
Detectors
Motion Detection
Elderly People
Motion Analysis

Keywords

  • Background subtraction
  • Fall-down detection
  • Motion detection
  • Object detection
  • Optical flow

ASJC Scopus subject areas

  • Computer Science (miscellaneous)
  • Control and Systems Engineering
  • Information Systems
  • Instrumentation
  • Hardware and Architecture
  • Computer Networks and Communications
  • Control and Optimization
  • Electrical and Electronic Engineering

Cite this

Moving object detection via TV-L1 optical flow in fall-down videos. / Mohamed, Nur Ayuni; Zulkifley, Mohd Asyraf.

In: Bulletin of Electrical Engineering and Informatics, Vol. 8, No. 3, 01.09.2019, p. 839-846.

Research output: Contribution to journalArticle

@article{d553b42113af4929a2ca82142ffd420b,
title = "Moving object detection via TV-L1 optical flow in fall-down videos",
abstract = "There is a growing demand for surveillance systems that can detect fall-down events because of the increased number of surveillance cameras being installed in many public indoor and outdoor locations. Fall-down event detection has been vigorously and extensively researched for safety purposes, particularly to monitor elderly peoples, patients, and toddlers. This computer vision detector has become more affordable with the development of high-speed computer networks and low-cost video cameras. This paper proposes moving object detection method based on human motion analysis for human fall-down events. The method comprises of three parts, which are preprocessing part to reduce image noises, motion detection part by using TV-L1 optical flow algorithm, and performance measure part. The last part will analyze the results of the object detection part in term of the bounding boxes, which are compared with the given ground truth. The proposed method is tested on Fall Down Detection (FDD) dataset and compared with Gunnar-Farneback optical flow by measuring intersection over union (IoU) of the output with respect to the ground truth bounding box. The experimental results show that the proposed method achieves an average IoU of 0.92524.",
keywords = "Background subtraction, Fall-down detection, Motion detection, Object detection, Optical flow",
author = "Mohamed, {Nur Ayuni} and Zulkifley, {Mohd Asyraf}",
year = "2019",
month = "9",
day = "1",
doi = "10.11591/eei.v8i3.1346",
language = "English",
volume = "8",
pages = "839--846",
journal = "Bulletin of Electrical Engineering and Informatics",
issn = "2089-3191",
publisher = "Institute of Advanced Engineering and Science (IAES)",
number = "3",

}

TY - JOUR

T1 - Moving object detection via TV-L1 optical flow in fall-down videos

AU - Mohamed, Nur Ayuni

AU - Zulkifley, Mohd Asyraf

PY - 2019/9/1

Y1 - 2019/9/1

N2 - There is a growing demand for surveillance systems that can detect fall-down events because of the increased number of surveillance cameras being installed in many public indoor and outdoor locations. Fall-down event detection has been vigorously and extensively researched for safety purposes, particularly to monitor elderly peoples, patients, and toddlers. This computer vision detector has become more affordable with the development of high-speed computer networks and low-cost video cameras. This paper proposes moving object detection method based on human motion analysis for human fall-down events. The method comprises of three parts, which are preprocessing part to reduce image noises, motion detection part by using TV-L1 optical flow algorithm, and performance measure part. The last part will analyze the results of the object detection part in term of the bounding boxes, which are compared with the given ground truth. The proposed method is tested on Fall Down Detection (FDD) dataset and compared with Gunnar-Farneback optical flow by measuring intersection over union (IoU) of the output with respect to the ground truth bounding box. The experimental results show that the proposed method achieves an average IoU of 0.92524.

AB - There is a growing demand for surveillance systems that can detect fall-down events because of the increased number of surveillance cameras being installed in many public indoor and outdoor locations. Fall-down event detection has been vigorously and extensively researched for safety purposes, particularly to monitor elderly peoples, patients, and toddlers. This computer vision detector has become more affordable with the development of high-speed computer networks and low-cost video cameras. This paper proposes moving object detection method based on human motion analysis for human fall-down events. The method comprises of three parts, which are preprocessing part to reduce image noises, motion detection part by using TV-L1 optical flow algorithm, and performance measure part. The last part will analyze the results of the object detection part in term of the bounding boxes, which are compared with the given ground truth. The proposed method is tested on Fall Down Detection (FDD) dataset and compared with Gunnar-Farneback optical flow by measuring intersection over union (IoU) of the output with respect to the ground truth bounding box. The experimental results show that the proposed method achieves an average IoU of 0.92524.

KW - Background subtraction

KW - Fall-down detection

KW - Motion detection

KW - Object detection

KW - Optical flow

UR - http://www.scopus.com/inward/record.url?scp=85071356495&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85071356495&partnerID=8YFLogxK

U2 - 10.11591/eei.v8i3.1346

DO - 10.11591/eei.v8i3.1346

M3 - Article

AN - SCOPUS:85071356495

VL - 8

SP - 839

EP - 846

JO - Bulletin of Electrical Engineering and Informatics

JF - Bulletin of Electrical Engineering and Informatics

SN - 2089-3191

IS - 3

ER -