CNN-Based Traffic Sign Recognition with Voice Alerts.

Traffic sign recognition is a crucial element of intelligent transportation systems aimed at improving road safety and driver awareness.In real-world driving conditions, traffic signs are often missed due to distractions, environmental factors, unfamiliar routes, or language barriers.To address these challenges, this project presents a real-time traffic sign recognition system with multilingual voice alerts using deep learning techniques.The proposed system processes images obtained from a live camera feed or uploaded files, which are pre processed through resizing and normalization before being classified using a Convolutional Neural Network (CNN).Once a traffic sign is recognized, its corresponding description is mapped to multilingual labels supporting English, Hindi, and Kannada, enabling region-specific accessibility.The system also incorporates a manual speed input mechanism, where the user-entered vehicle speed is compared with detected speed-limit signs to identify potential violations.To enhance usability and reduce driver distraction, Text-to-Speech (TTS) technology is integrated to generate real-time voice alerts.The system is implemented using Python, TensorFlow, OpenCV, and PyQt5, and demonstrates reliable real-time performance.Overall, the proposed solution provides a practical, cost-effective, and scalable approach to traffic sign recognition with potential applications in driver assistance and intelligent transportation systems.