Real-time traffic sign detection and recognition using Raspberry Pi

Nowadays, the number of road accident in Malaysia is increasing expeditiously. One ways to reduce through development advanced driving assistance system (ADAS) by professional engineers. Several ADAS has been proposed taking into consideration delay tolerance and accuracy itself. In this work, a traffic sign recognition developed increase safety users installing inside car for driver’s awareness. TensorFlow algorithm considered work object machine learning due its high accuracy. The embedded Raspberry Pi 3 processing analysis detect from real-time video recording camera NoIR. This aims study accuracy, reliability using processor considering several scenarios related state environment condition signs. A testbed implementation conducted twenty different signs results show that more than 90% reliable with an acceptable delay.