Srikanth Pagadala

Finding Lane Lines on the Road

01 Dec 2016

Finding Lane Lines in a video clip is considered the “Hello World” of Computer Vision in Robotics.

It involves following steps:

  • Color Selection
  • Region Masking
  • Canny Edge Detection
  • Hough Tranform

OpenCV is our choice of tool in Python for computer vision. It provides implementation of all of the above algorithms.

This mini-project takes you step-by-step through the Lane Lines detection. Towards the end it also shows scenarios where such simple implementation will fail.

Source Code

Report

If you are in a hurry and don’t want to read the full report, just watch these three videos:

Finding White Lanes Lines

Finding Yellow Lanes Lines

Finding Lanes Lines Fail

Next: Deep Learning to Clone Driving Behavior