An opencv-based implementation of a self-driving car module responsible for detecting lane lines under different lighting conditions, pavement textures and curves.
Implementation includes the following:
- Camera calibration and distortion correction.
- Applying Gaussian Blur.
- Edge Detection using Sobel.
- Color/gradient thresholding.
- Perspective transform.
- Detecting lane lines using peaks of histogram and sliding window.
- Drawing lanes on the original image
- Determining the lane curvature.
pip install numpy
pip install matplotlib
pip install opencv-python