A Real Time Traffic Sign Classification Model Based on Convolutional Neural Networks

This research paper introduces a hybrid CNN-LeN et-5 model for real-time traffic sign detection. The model combines the efficiency of a lightweight CNN architecture with the proven effectiveness of the LeNet-5 model. By incorporating two convolutional layers, two max-pooling layers, and two fully connected layers, the proposed solution classifies traffic signs from 30x30x3 RGB images. The hybrid model is well-suited for versatile use cases, including real-time detection and alerting in vehicles. To overcome challenges such as adverse weather conditions and high-speed scenarios, different CNN architectures and image pre-processing techniques are explored. With its lightweight nature, the model can be deployed on resource-constrained devices, expanding applications to mobile and
$\text{IoT}$
devices. This research paper presents an efficient and adaptable solution, leveraging the strengths of CNN-LeNet-5 hybrid models for enhanced road safety through real-time traffic sign detection.