Road region extraction based on motion information and seeded region growing for foreground detection

Hongwu Qin; Jasni Mohamad Zain; Xiuqin Ma; Tao Hai

doi:10.1007/978-3-642-14292-5_14

Road region extraction based on motion information and seeded region growing for foreground detection

Hongwu Qin
, Jasni Mohamad Zain
, Xiuqin Ma
, Tao Hai

Universiti Malaysia Pahang Al-Sultan Abdullah

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

Abstract

This paper proposes a road region extraction method based on the motion information of foreground objects and seeded region growing (SRG) algorithm. By learning on a training set of a scene over a period of time, we get the trajectory of moving object, then use SRG algorithm in which the trajectory is used as seed to extract road region. As a result, instead of detecting foreground objects in a conventional pixel by pixel manner, detection can be mainly performed on or near the pixels of road region so as to facilitate and accelerate foreground detection. In addition, the regions outside road region most of the time do not need to be transmitted in visual communication. Experimental results represent the accuracy and usefulness of our proposed method.

Original language	English
Title of host publication	Networked Digital Technologies - Second International Conference, NDT 2010, Proceedings
Pages	128-134
Number of pages	7
Edition	PART 1
DOIs	https://doi.org/10.1007/978-3-642-14292-5_14
State	Published - 2010
Externally published	Yes
Event	2nd International Conference on 'Networked Digital Technologies', NDT 2010 - Prague, Czech Republic Duration: 7 Jul 2010 → 9 Jul 2010

Publication series

Name	Communications in Computer and Information Science
Number	PART 1
Volume	87 CCIS
ISSN (Print)	1865-0929

Conference

Conference	2nd International Conference on 'Networked Digital Technologies', NDT 2010
Country/Territory	Czech Republic
City	Prague
Period	7/07/10 → 9/07/10

Keywords

Foreground detection
region extraction
seeded region growing
visual surveillance

Access to Document

10.1007/978-3-642-14292-5_14

Cite this

Qin, H., Zain, J. M., Ma, X., & Hai, T. (2010). Road region extraction based on motion information and seeded region growing for foreground detection. In Networked Digital Technologies - Second International Conference, NDT 2010, Proceedings (PART 1 ed., pp. 128-134). (Communications in Computer and Information Science; Vol. 87 CCIS, No. PART 1). https://doi.org/10.1007/978-3-642-14292-5_14

@inproceedings{a3013ca071354fdfba82331506b3d16c,

title = "Road region extraction based on motion information and seeded region growing for foreground detection",

abstract = "This paper proposes a road region extraction method based on the motion information of foreground objects and seeded region growing (SRG) algorithm. By learning on a training set of a scene over a period of time, we get the trajectory of moving object, then use SRG algorithm in which the trajectory is used as seed to extract road region. As a result, instead of detecting foreground objects in a conventional pixel by pixel manner, detection can be mainly performed on or near the pixels of road region so as to facilitate and accelerate foreground detection. In addition, the regions outside road region most of the time do not need to be transmitted in visual communication. Experimental results represent the accuracy and usefulness of our proposed method.",

keywords = "Foreground detection, region extraction, seeded region growing, visual surveillance",

author = "Hongwu Qin and Zain, \{Jasni Mohamad\} and Xiuqin Ma and Tao Hai",

year = "2010",

doi = "10.1007/978-3-642-14292-5\_14",

language = "English",

isbn = "3642142915",

series = "Communications in Computer and Information Science",

number = "PART 1",

pages = "128--134",

booktitle = "Networked Digital Technologies - Second International Conference, NDT 2010, Proceedings",

edition = "PART 1",

note = "2nd International Conference on 'Networked Digital Technologies', NDT 2010 ; Conference date: 07-07-2010 Through 09-07-2010",

}

Qin, H, Zain, JM, Ma, X & Hai, T 2010, Road region extraction based on motion information and seeded region growing for foreground detection. in Networked Digital Technologies - Second International Conference, NDT 2010, Proceedings. PART 1 edn, Communications in Computer and Information Science, no. PART 1, vol. 87 CCIS, pp. 128-134, 2nd International Conference on 'Networked Digital Technologies', NDT 2010, Prague, Czech Republic, 7/07/10. https://doi.org/10.1007/978-3-642-14292-5_14

Road region extraction based on motion information and seeded region growing for foreground detection. / Qin, Hongwu; Zain, Jasni Mohamad; Ma, Xiuqin et al.
Networked Digital Technologies - Second International Conference, NDT 2010, Proceedings. PART 1. ed. 2010. p. 128-134 (Communications in Computer and Information Science; Vol. 87 CCIS, No. PART 1).

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

TY - GEN

T1 - Road region extraction based on motion information and seeded region growing for foreground detection

AU - Qin, Hongwu

AU - Zain, Jasni Mohamad

AU - Ma, Xiuqin

AU - Hai, Tao

PY - 2010

Y1 - 2010

N2 - This paper proposes a road region extraction method based on the motion information of foreground objects and seeded region growing (SRG) algorithm. By learning on a training set of a scene over a period of time, we get the trajectory of moving object, then use SRG algorithm in which the trajectory is used as seed to extract road region. As a result, instead of detecting foreground objects in a conventional pixel by pixel manner, detection can be mainly performed on or near the pixels of road region so as to facilitate and accelerate foreground detection. In addition, the regions outside road region most of the time do not need to be transmitted in visual communication. Experimental results represent the accuracy and usefulness of our proposed method.

AB - This paper proposes a road region extraction method based on the motion information of foreground objects and seeded region growing (SRG) algorithm. By learning on a training set of a scene over a period of time, we get the trajectory of moving object, then use SRG algorithm in which the trajectory is used as seed to extract road region. As a result, instead of detecting foreground objects in a conventional pixel by pixel manner, detection can be mainly performed on or near the pixels of road region so as to facilitate and accelerate foreground detection. In addition, the regions outside road region most of the time do not need to be transmitted in visual communication. Experimental results represent the accuracy and usefulness of our proposed method.

KW - Foreground detection

KW - region extraction

KW - seeded region growing

KW - visual surveillance

UR - https://www.scopus.com/pages/publications/77955742628

U2 - 10.1007/978-3-642-14292-5_14

DO - 10.1007/978-3-642-14292-5_14

M3 - Conference contribution

AN - SCOPUS:77955742628

SN - 3642142915

SN - 9783642142918

T3 - Communications in Computer and Information Science

SP - 128

EP - 134

BT - Networked Digital Technologies - Second International Conference, NDT 2010, Proceedings

T2 - 2nd International Conference on 'Networked Digital Technologies', NDT 2010

Y2 - 7 July 2010 through 9 July 2010

ER -

Road region extraction based on motion information and seeded region growing for foreground detection

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Fingerprint

Cite this