DocumentCode
3577965
Title
Robust vehicle detection and tracking method for Blind Spot Detection System by using vision sensors
Author
Seunghwan Baek ; Heungseob Kim ; Kwangsuck Boo
Author_Institution
High Safety Vehicle Core Technol. Res. Center, Inje Univ., Gimhae, South Korea
fYear
2014
Firstpage
729
Lastpage
735
Abstract
This paper presents a method to detect the present vehicles from side and rear for BSDS(Blind Spot Detection System) with vision system. Because the real image acquired during car driving has a lot of information to exam the target vehicle, background image, and the noises such as lighting and shading, it is hard to extract only the target vehicle for the background image with satisfied robustness. In this paper, the target vehicle is detected by repetitive image processing such as sobel and morphological operations and a Kalman filter is also designed to cancel the background image and prevent the misreading of the target image. Compared to previous researches, the proposed method can get an image processing with much improved speed and robustness. Various experiments were performed on the highway driving situations to evaluate the performance of the proposed algorithm.
Keywords
Kalman filters; automobiles; computer vision; image sensors; object detection; object tracking; traffic engineering computing; BSDS; Kalman filter; background image; blind spot detection system; car driving; highway driving situations; image processing; morphological operations; repetitive image processing; robust vehicle detection; robust vehicle tracking method; satisfied robustness; sobel operations; target vehicle; vision sensors; Computational modeling; Image resolution; Kalman filters; Robustness; Vehicles; Blind Spot Detection System(BSDS); Collision Prevention; Kalman Filter; Lane Change Assist(LCA); Vehicle Detection;
fLanguage
English
Publisher
ieee
Conference_Titel
Complex Systems (WCCS), 2014 Second World Conference on
Print_ISBN
978-1-4799-4648-8
Type
conf
DOI
10.1109/ICoCS.2014.7060984
Filename
7060984
Link To Document