Lane Detection Project

Overview

This project is inspired by the Teknofest 2025 Efficiency Challenge competition and implements one of the core ADAS (Advanced Driver Assistance Systems) functions: lane keeping. The system is designed to enable autonomous driving by processing live video from a camera mounted on a vehicle, detecting lane lines in real time, and providing steering guidance. Using OpenCV and other Python libraries, the project demonstrates how a vehicle can follow lanes autonomously, making it suitable for research, competitions, and educational purposes related to self-driving technologies.

---

Features

• Lane Detection: Detects lane lines using advanced image processing techniques.
• Curve Prediction: Measures curve radius and predicts steering direction (left, right, or straight).
• Perspective Adjustment: Adjustable perspective points for different video inputs.
• Live Camera Support: Supports video input from a webcam.
• Resolution Compatibility: All processing is optimized for 1280x720 resolution.
• Visual Indicators: Marks perspective points and detected lanes on the video.
• Error Handling: Handles empty lane detection gracefully without crashing.

---

Installation

Prerequisites

Ensure the following software and libraries are installed:

• Python 3.6 or higher
• OpenCV 4 or higher
• Numpy 1.20 or higher
• Matplotlib 3.3 or higher

Setup

1. Clone the repository:

   git clone https://github.com/your-repo/lane-detection.git

2. Install the required Python libraries:

   pip install opencv-python numpy matplotlib

---

How It Works

Main Workflow

1. Video Input: The system reads video input from a file or webcam.
2. Image Processing: Applies HLS filtering, grayscale conversion, thresholding, and edge detection.
3. Perspective Warp: Adjusts the perspective to a bird's-eye view for better lane detection.
4. Lane Detection: Uses sliding window and histogram techniques to detect lane lines.
5. Curve Measurement: Calculates the radius of curvature and predicts the curve direction.
6. Lane Visualization: Draws detected lanes and perspective points on the video.
7. Error Handling: Displays the raw frame if lane detection fails.

Key Functions

• readVideo(): Reads video input from a file or webcam.
• processImage(): Processes the image to isolate lane lines.
• perspectiveWarp(): Warps the perspective for better lane detection.
• plotHistogram(): Plots the histogram of white pixels.
• slide_window_search(): Detects lane lines using sliding window technique.
• measure_lane_curvature(): Measures the curve radius and direction.
• draw_lane_lines(): Visualizes detected lanes on the video.
• addText(): Adds curvature and deviation information to the video.

---

Usage

1. Place your video file (V1.mp4) in the project directory.
2. Run the script:

   python Lane_Detection.py

3. Press Enter to stop the video playback.

---

Customization

• Perspective Points: Modify the src array in the perspectiveWarp() function to adjust perspective points.
• Lane Colors: Change the lower_colour and upper_colour arrays in processImage() to detect different lane colors.

---

Troubleshooting

• Empty Lane Detection: If no lanes are detected, the system will display the raw frame with a warning message.
• Resolution Handling: The code automatically resizes any input to 1280x720 for optimal processing.

---

Acknowledgments

• Developed by Bilal Sinav
• Inspired by advanced driver assistance systems (ADAS) functions as followed in the Teknofest 2025 Efficiency Challenge competition for autonomous vehicles.
• Also inspired by Can Ozcivelek's lane detection project.

---

Contact

For questions or feedback, please contact Bilal Sinav.