Skip to content

dhairyaostwal/Self-Driving-Car

Repository files navigation

Self Driving Car Lane Detection

Steps to run the project

Quick note:

Running python3 detector.py in terminal would show the video output

Running python3 streamVideo.py in terminal would show the stream output on localhost 5000 address generated

Jump to Output


1. Clone the project
2. Open the project inside terminal
3. Inside terminal type python3 detector.py
4. Inside terminal type python3 streamVideo.py

Transformations & Techniques used

✅Hough Transformation

➡️ Yields us infinite lines on edges detected by Canny Edge Detection Method

Applying on Soduku Image:

PHough-Soduku

✅Probabilistic Hough Transformation

➡️ Yields us finite lines on edges detected by Canny Edge Detection Method vs infinite as in case of Hough Transformation

Lets apply on a simple Soduku image:

PHough-Soduku

As you can see finite lines. Now let's apply on our Road Image :

PHoughTransform

✅ Building out Region of Interest aka ROI

Masked Image

Further edge detected

Edges detected

Removing corner edges which are unwanted requires edge detection first then masking

Perfect Edge Detection

Finally upon masking,

Masked Image

Video Dataset: Click here

✅ Output

Lane Detection Output

Streaming on localhost 5000