A new threat degree analysis method of abandoned objects based on dynamic multifeature fusion in urban traffic. Issue 4 (3rd July 2016)

• Record Type: Journal Article
• Authors: Lan, Jinhui; Yu, Dongyang; Ran, Bin; Jiang, Yaoliang; Zhang, Zhaohui
• Is Part Of: Journal of intelligent transportation systems. Volume 20:Issue 4(2016)
• Page Start: 401
• Page End: 412
• Publication Date: 2016-07-03
• Subjects: Abandoned Object -- Feature Fusion -- Multivehicle Tracking -- Threat Degree Analysis; Intelligent transportation systems -- Periodicals; Transportation -- Technological innovations -- Periodicals; 388.312
• DOI: 10.1080/15472450.2015.1082910
• Languages: English
• ISSNs: 1547-2450

Abstract

ABSTRACT: Abandoned objects in a road can be a potential threat to urban traffic. In this article we present a new threat degree analysis method of abandoned objects, which is based on dynamic multifeature fusion to detect the abandoned objects and analyze the threat degree automatically. The image sequence is acquired with a visible sensor installed at overbridges, and is processed by a PC-based image-processing system. In this method, we first put forward a multivehicle tracking algorithm to acquire the vehicles' moving information as preprocessing. Then we extracted the dynamic features and built a security threat model upstream of the abandoned object to facilitate further analysis. Finally, we proposed a fusion decision method, based on dynamic multiple features, to analyze the threat degree. The experiments show that the developed method can determine the threat degree of abandoned objects well and truly, and it is appropriate in the intelligent traffic monitoring system.