DocumentCode :
261043
Title :
Improved lane detection system using Hough transform with super-resolution reconstruction algorithm and multi-ROI
Author :
Jae-Hyun Cho ; Tsogtbaatar, Erdenetuya ; Seong-Hoon Kim ; Young-Min Jang ; Pham-Minh-Luan Nguyen ; Sang-Bock Cho
Author_Institution :
Univ. of Ulsan, Ulsan, South Korea
fYear :
2014
fDate :
15-18 Jan. 2014
Firstpage :
1
Lastpage :
4
Abstract :
Nowadays, due to the increasing need for using black boxes in vehicles, an evolving market is being created; that for example in South Korea, more than 1 million vehicles have been equipped with this device. In this paper, in order to improve the lane detection recognition via Hough transform, we improved the algorithm by set multi-ROI to reduce error rates and unrecognized part of the outside something as a lane. And we applied super-resolution reconstruction to correct the image. Through the proposed algorithm, proposed algorithm increase 0.6% of lane recognition rate and by setting multi-ROI irregular part of the road does not recognize so lane detection error rate was reduced significantly.
Keywords :
Hough transforms; image recognition; image reconstruction; image resolution; Hough transform; South Korea; black boxes; image super-resolution reconstruction algorithm; improved lane detection system; multiROI irregular part; Error analysis; Image reconstruction; Image resolution; Reconstruction algorithms; Roads; Transforms; Vehicles; Hough-transform; Super-Resolution Reconstruction; image processing; lane tracking; multi ROI;
fLanguage :
English
Publisher :
ieee
Conference_Titel :
Electronics, Information and Communications (ICEIC), 2014 International Conference on
Conference_Location :
Kota Kinabalu
Type :
conf
DOI :
10.1109/ELINFOCOM.2014.6914446
Filename :
6914446
Link To Document :
بازگشت