Skip to content

Simple multi-lane detection using OpenCV in C++. The goal is to use it in the self-driving car simulator Carla.

Notifications You must be signed in to change notification settings

jhallier/LaneDetect

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

1 Commit
 
 
 
 
 
 
 
 

Repository files navigation

LaneDetect

LaneDetect is a simple road lane detection algorithm using OpenCV, which I originally wrote in Python and then transferred to C++ to get familiar with the C++ API of OpenCV.

Requirements

  • Linux (I recommend Ubuntu 16.04 or 18.04)
  • OpenCV Version >= 3.4 (sources download, installation instructions
  • Spline header library from here (will be automatically downloaded if you run cmake)

Build instructions

Clone this repository, e.g. git clone https://github.com/jhallier/LaneDetect

cd LaneDetect
mkdir build
cd build
cmake ..
make

Usage

Run

./LaneDetect [image filename]

or call the LaneDetect class functions from your own code.

ToDo

  1. Using different color spaces for improved lane detection
  2. Smoothing over a few frames
  3. Track lanes over time (assign IDs to lanes, distinguish own and other lanes)
  4. Improve robustness at crossroads etc.

Sample image

Image from Carla simulator Detection result

About

Simple multi-lane detection using OpenCV in C++. The goal is to use it in the self-driving car simulator Carla.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published