I recently completed the Self-Driving Cars Specialization by the University of Toronto. While the course provided a comprehensive overview of autonomous vehicle technology, I wanted to delve deeper into the computer vision aspect.
To gain hands-on experience, I developed a lane line detection program. This project involved:
- Image Processing: Applying filters and transformations to isolate lane markings from the surrounding environment.
- Edge Detection: Utilizing algorithms to identify the boundaries of the lane lines.
- Hough Transform: Converting edge points into lines representing the lane markings.
- Curve Fitting: Approximating the lane lines with mathematical curves for smoother representation.
- Visualization: Overlaying the detected lane lines on the original image for verification. This project allowed me to solidify my understanding of computer vision principles and gain practical experience in a core component of self-driving car technology.
Key Technologies Used
- Python
- OpenCV
Feel free to check out my code and give feedback on GitHub. Although I’ve been lazy and haven’t actually written a README, there are a few comments on the code.