Lane detection and traffic sign detection for autonomous driving systems

In the realm of autonomous driving, this study focuses on two critical tasks for scene understanding: lane detection and traffic sign detection. These tasks are essential to ensuring safe and reliable navigation in self-driving systems. A combination of traditional computer vision techniques and modern deep learning methods are used to achieve accurate detection and analysis of lanes and traffic signs in real-world driving scenarios. For lane detection, we utilize edge detection and the Hough Transform, while for traffic sign detection, we apply machine learning algorithms to recognize and classify various signs. This hybrid approach demonstrates the potential of integrating classical computer vision algorithms with deep learning models to enhance the perception capabilities of autonomous vehicles. The model achieves a balanced performance, with accuracy metrics ranging from 80-90% across both tasks, indicating a robust ability to detect lanes and signs effectively under varied conditions. This work serves as a foundation for developing more advanced scene understanding systems in autonomous driving technology. In the future, we aim to enhance the model’s adaptability through continuous learning and incorporating additional sensory data, such as LiDAR and radar. Additionally, expanding the dataset and testing the models in diverse environmental conditions, including different lighting and weather scenarios, will further improve the robustness and reliability.