Real-time UHD background modelling with mixed selection block updates

Axel Beaugendre, Satoshi Goto, Takeshi Yoshimura

Research output: Contribution to journalArticle

Abstract

The vast majority of foreground detection methods require heavy hardware optimization to process in real-time standard definition videos. Indeed, those methods process the whole frame for the detection but also for the background modelling part which makes them resourceguzzlers (time, memory, etc.) unable to be applied to Ultra High Definition (UHD) videos. This paper presents a real-time background modelling method called Mixed Block Background Modelling (MBBM). It is a spatio-temporal approach which updates the background model by carefully selecting block by a linear and pseudo-random orders and update the corresponding model's block parts. The two block selection orders make sure that every block will be updated. For foreground detection purposes, the method is combined with a foreground detection designed for UHD videos such as the Adaptive Block-Propagative Background Subtraction method. Experimental results show that the proposed MBBM can process 50 min. of 4K UHD videos in less than 6 hours. while other methods are estimated to take from 8 days to more than 21 years. Compared to 10 stateof-the-art foreground detection methods, the proposed MBBM shows the best quality results with an average global quality score of 0.597 (1 being the maximum) on a dataset of 4K UHDTV sequences containing various situation like illumination variation. Finally, the processing time per pixel of the MBBM is the lowest of all compared methods with an average of 3.18×10-8 s.

Original languageEnglish
Pages (from-to)581-591
Number of pages11
JournalIEICE Transactions on Fundamentals of Electronics, Communications and Computer Sciences
VolumeE100A
Issue number2
DOIs
Publication statusPublished - 2017 Feb 1

Fingerprint

Background Modeling
Update
Real-time
Lighting
Pixels
Hardware
Data storage equipment
Processing
Order Selection
Background Subtraction
Modeling Method
Illumination
Lowest
Pixel
Optimization
Experimental Results

Keywords

  • 4K
  • Background modelling
  • Detection
  • MBBM
  • UHD

ASJC Scopus subject areas

  • Signal Processing
  • Computer Graphics and Computer-Aided Design
  • Applied Mathematics
  • Electrical and Electronic Engineering

Cite this

Real-time UHD background modelling with mixed selection block updates. / Beaugendre, Axel; Goto, Satoshi; Yoshimura, Takeshi.

In: IEICE Transactions on Fundamentals of Electronics, Communications and Computer Sciences, Vol. E100A, No. 2, 01.02.2017, p. 581-591.

Research output: Contribution to journalArticle

@article{45c8a1e2502c4c4996a603b927b1ee13,
title = "Real-time UHD background modelling with mixed selection block updates",
abstract = "The vast majority of foreground detection methods require heavy hardware optimization to process in real-time standard definition videos. Indeed, those methods process the whole frame for the detection but also for the background modelling part which makes them resourceguzzlers (time, memory, etc.) unable to be applied to Ultra High Definition (UHD) videos. This paper presents a real-time background modelling method called Mixed Block Background Modelling (MBBM). It is a spatio-temporal approach which updates the background model by carefully selecting block by a linear and pseudo-random orders and update the corresponding model's block parts. The two block selection orders make sure that every block will be updated. For foreground detection purposes, the method is combined with a foreground detection designed for UHD videos such as the Adaptive Block-Propagative Background Subtraction method. Experimental results show that the proposed MBBM can process 50 min. of 4K UHD videos in less than 6 hours. while other methods are estimated to take from 8 days to more than 21 years. Compared to 10 stateof-the-art foreground detection methods, the proposed MBBM shows the best quality results with an average global quality score of 0.597 (1 being the maximum) on a dataset of 4K UHDTV sequences containing various situation like illumination variation. Finally, the processing time per pixel of the MBBM is the lowest of all compared methods with an average of 3.18×10-8 s.",
keywords = "4K, Background modelling, Detection, MBBM, UHD",
author = "Axel Beaugendre and Satoshi Goto and Takeshi Yoshimura",
year = "2017",
month = "2",
day = "1",
doi = "10.1587/transfun.E100.A.581",
language = "English",
volume = "E100A",
pages = "581--591",
journal = "IEICE Transactions on Fundamentals of Electronics, Communications and Computer Sciences",
issn = "0916-8508",
publisher = "Maruzen Co., Ltd/Maruzen Kabushikikaisha",
number = "2",

}

TY - JOUR

T1 - Real-time UHD background modelling with mixed selection block updates

AU - Beaugendre, Axel

AU - Goto, Satoshi

AU - Yoshimura, Takeshi

PY - 2017/2/1

Y1 - 2017/2/1

N2 - The vast majority of foreground detection methods require heavy hardware optimization to process in real-time standard definition videos. Indeed, those methods process the whole frame for the detection but also for the background modelling part which makes them resourceguzzlers (time, memory, etc.) unable to be applied to Ultra High Definition (UHD) videos. This paper presents a real-time background modelling method called Mixed Block Background Modelling (MBBM). It is a spatio-temporal approach which updates the background model by carefully selecting block by a linear and pseudo-random orders and update the corresponding model's block parts. The two block selection orders make sure that every block will be updated. For foreground detection purposes, the method is combined with a foreground detection designed for UHD videos such as the Adaptive Block-Propagative Background Subtraction method. Experimental results show that the proposed MBBM can process 50 min. of 4K UHD videos in less than 6 hours. while other methods are estimated to take from 8 days to more than 21 years. Compared to 10 stateof-the-art foreground detection methods, the proposed MBBM shows the best quality results with an average global quality score of 0.597 (1 being the maximum) on a dataset of 4K UHDTV sequences containing various situation like illumination variation. Finally, the processing time per pixel of the MBBM is the lowest of all compared methods with an average of 3.18×10-8 s.

AB - The vast majority of foreground detection methods require heavy hardware optimization to process in real-time standard definition videos. Indeed, those methods process the whole frame for the detection but also for the background modelling part which makes them resourceguzzlers (time, memory, etc.) unable to be applied to Ultra High Definition (UHD) videos. This paper presents a real-time background modelling method called Mixed Block Background Modelling (MBBM). It is a spatio-temporal approach which updates the background model by carefully selecting block by a linear and pseudo-random orders and update the corresponding model's block parts. The two block selection orders make sure that every block will be updated. For foreground detection purposes, the method is combined with a foreground detection designed for UHD videos such as the Adaptive Block-Propagative Background Subtraction method. Experimental results show that the proposed MBBM can process 50 min. of 4K UHD videos in less than 6 hours. while other methods are estimated to take from 8 days to more than 21 years. Compared to 10 stateof-the-art foreground detection methods, the proposed MBBM shows the best quality results with an average global quality score of 0.597 (1 being the maximum) on a dataset of 4K UHDTV sequences containing various situation like illumination variation. Finally, the processing time per pixel of the MBBM is the lowest of all compared methods with an average of 3.18×10-8 s.

KW - 4K

KW - Background modelling

KW - Detection

KW - MBBM

KW - UHD

UR - http://www.scopus.com/inward/record.url?scp=85011798933&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=85011798933&partnerID=8YFLogxK

U2 - 10.1587/transfun.E100.A.581

DO - 10.1587/transfun.E100.A.581

M3 - Article

AN - SCOPUS:85011798933

VL - E100A

SP - 581

EP - 591

JO - IEICE Transactions on Fundamentals of Electronics, Communications and Computer Sciences

JF - IEICE Transactions on Fundamentals of Electronics, Communications and Computer Sciences

SN - 0916-8508

IS - 2

ER -