Optimizing Traffic Sign Detection with MobileNetV2: A Lightweight Deep Learning Approach

The ability to recognize traffic signs is a prerequisite for both autonomous driving and intelligent transportation systems, and it is essential to the realization of sustainable cities and communities. By lowering the number of traffic accidents and fatalities, efficient and accurate traffic sign recognition improves road safety and urban mobility, promoting good health and well-being. This project uses deep learning methods, namely Convolutional Neural Networks (CNNs) and MobileNetV2 architecture, to classify Indian traffic signs into 85 distinct categories using a dataset from Kaggle. The collection of traffic signs features a range of signs that are commonly seen on Indian highways and that promote infrastructure, industry, and innovation through the application of transportation technology. The integration of MobileNetV2 in the CNN model offers high accuracy and possible implementation of real-time traffic deployment. Furthermore, the CNN model will support Climate Action; it can aid in the superior intelligent management in traffic matters while curbing car emissions through efficient flow of traffic. Deep learning models make a massive improvement of traffic sign recognition efficiency and accuracy since it is significant in the evolution of better road safety and navigation of self-driving vehicles. This technical innovation advances the sector of the autonomous car besides making cities more resilient, safe, inclusive, and sustainable. All things considered, this study brings out the importance of deep learning to the creation of intelligent transportation systems that support and promote safer and more sustainable urban lives.