/lane-and-vehicle-detection

Lanes detection using image processing techniques and Real-time vehicle detection using YOLOv3 and NMS

Primary LanguageHTML

Table of contents

General info

The project divided into two phases, For the first phase:

  • The pipeline be able to detect the lanes, highlight them using a yellow color, and colour the area between them in blue.
  • Display the estimatation of the vehicle position away from the center of the lane and the radius of curvature of the road as well.

For the second phase, we locate and identify the cars on the road using YOLOv3 and select the right bounding box using Non-Max Suppression.

System Pipeline

The seven stages of our pipeline

Run

To run normal mode

chmod u+x run.sh
./run.sh <input_video>

To run in deubg mode

chmod u+x run.sh
./run.sh <input_video> deubg
Output in Normal Mode Output in Debug Mode