Road lane tracking based on monocular vision
Lane tracking and detection is a complex problem due to the versatility of road conditions in which the road stream of images is analyzed. In this paper, an algorithm was developed to obtain a robust real-time lane tracking under a sudden appearance of obstacles such as vehicles and shadows. The met...
Main Authors: | , , |
---|---|
Format: | Article |
Language: | English English |
Published: |
Research India Publications
2015
|
Subjects: | |
Online Access: | http://irep.iium.edu.my/43041/ http://irep.iium.edu.my/43041/ http://irep.iium.edu.my/43041/1/Lane_tracking.pdf http://irep.iium.edu.my/43041/4/43041_Road%20lane%20tracking%20based%20on%20monocular%20vision_SCOPUS.pdf |
Summary: | Lane tracking and detection is a complex problem due to the versatility of road conditions in which the road stream of images is analyzed. In this paper, an algorithm was developed to obtain a robust real-time lane tracking under a sudden appearance of obstacles such as vehicles and shadows. The method starts with applying Gaussian filter to grey scale images. Then, ROI is defined by finding the lowest mean of all rows in the image. Canny edge detector followed by Hough Transform are applied to the ROI to acquire lines. The information of lines in a sequence of frames are stored to predict the lane position. This method was tested on 1700 frames in different situations and proved its robustness. |
---|