This paper proposes a method for vehicle detection which can gather the traffic information including the number of vehicles, the dimension of vehicles, and types from the image sequences of traffic surveillance video. To collect the traffic information which provides the reference for transportation institute to manage efficiently the traffic flow is the major purpose of this paper. The proposed method first obtains the moving objects from the background subtraction gotten the foreground images and then uses spatial filtering, morphological processing and a series of image processing technologies to detect the vehicles. All the detection processing will focus on the region of interest (ROI) that should reduce the computation of great complexity and improve the speed of system process. Once the vehicles are detected, the system can obtain the vehicle border frame based on the feature retrieving method to start the tracking and find the vehicles correlation from the consecutive images to gain a useful traffic behavior. Based on the experimental result, we convince that the proposed method can provide accurate traffic information and be helpful to the traffic management efficiently.
|頁（從 - 到）||1301-1305|
|期刊||ICIC Express Letters|
|出版狀態||Published - 2014 五月|
All Science Journal Classification (ASJC) codes
- Control and Systems Engineering
- Computer Science(all)