Ask Your Question
3

lane tracking - how to group hough lines

asked 2014-06-27 11:02:23 -0500

sy456 gravatar image

updated 2015-01-16 02:28:45 -0500

berak gravatar image

Hi,

I use hough transform for straight lane detection but I got lots of false positives even in clear lane markings. My approach is like this:

  1. Select ROI for near field
  2. Eliminate noise (various options like gaussian, median filter...)
  3. Detect edges and binarize image
  4. Detect Hough lines
  5. Eliminate lines considering the angle of the lines (different angles for left and right side of image)
  6. Group Hough lines and draw bounding box around them
  7. Track bounding boxes with Kalman filter

I am stuck with 5th and 6th step. Some of the lines, which are very close to each other, are drawn on the lane markings. How can I distinguish these lines and the others that are not on the lane markings? I also wonder if hough transform is a really good idea for straight lines. Maybe I need to use RANSAC for fitting. Any suggestions?

This is very simple image from Caltech Lanes Dataset. However, I don't work on single images but image sequences under changing lighting conditions.

image description

Regards

edit retag flag offensive close merge delete

Comments

Can you post the image you are working?.

Haris gravatar imageHaris ( 2014-06-27 12:11:10 -0500 )edit

I am currently working on the same project actually i want to know how you did step 6 which is grouping hough lines and drawing window to apply RANSAC line fitting on it ?

Sarah gravatar imageSarah ( 2017-10-07 13:04:47 -0500 )edit

1 answer

Sort by ยป oldest newest most voted
2

answered 2014-10-06 03:42:20 -0500

zweistein666 gravatar image

I'm currently working on the same subject and this http://hompi.sogang.ac.kr/fxlab/paper/45.pdf">publication might help you in all the steps.

Regarding step 5:

  1. first you should filter out all the almost horizontal and vertical lines.
  2. differentiate the left and right side of the image, so on one side u filter out the lines with one orientation and on the other side with the other. If y=ax+b than left side a>0 and on the right a<0...
  3. I would suggest to do line unification in case you detect more the one line close by each other
edit flag offensive delete link more

Comments

Dont you think that a bird eye view would help you more in detecting and tracking the lane?

zweistein666 gravatar imagezweistein666 ( 2014-10-07 06:54:39 -0500 )edit

Yes it will :)

StevenPuttemans gravatar imageStevenPuttemans ( 2015-01-17 06:21:38 -0500 )edit

Question Tools

Stats

Asked: 2014-06-27 11:02:23 -0500

Seen: 3,788 times

Last updated: Jan 16 '15