Published Paper Details:

With the increasing prevalence of intelligent transportation systems and the demand for safer roads, the detection and classification of road traffic signs have become critical in computer vision and deep learning research. This study presents an efficient deep learning system for reliable detection and precise classification of road traffic signs in real-world scenarios. Leveraging convolutional neural networks (CNNs), the system automatically extracts hierarchical features from raw image data, optimizing both computational efficiency and performance. We introduce techniques to improve the model's robustness against various lighting conditions, occlusions, and sign shapes. By utilizing lightweight neural network architectures and applying optimization strategies such as model pruning and quantization, the algorithm achieves high accuracy with low computational overhead, making it suitable for resource-constrained environments. Evaluations on benchmark datasets and real-world traffic scenarios demonstrate superior accuracy and adaptability to different sensor modalities, including camera-based and lidar-based systems. This research contributes to advancing efficient and accurate road traffic sign detection and classification, providing a promising solution for enhancing road safety and supporting intelligent transportation systems.