Development of a real-time alert system of traffic sign detection and recognition

Choy, Ja Yeong (2021) Development of a real-time alert system of traffic sign detection and recognition. Project Report. Universiti Teknikal Malaysia Melaka, Melaka, Malaysia. (Submitted)

Text (Full Text) Development of a real-time alert system of traffic sign detection and recognition.pdf - Submitted Version Download (5MB)

Abstract

Nowadays, the road accident in Malaysia is increasing expeditiously. Therefore, the implementation of the road sign recognition system is necessary to reduce the number of road accidents. In this work, a road sign recognition system has been developed to ensure road safety and driver awareness. The system has low power consumption and can be easily installed in the car for driver awareness. Previous study on the development of Advanced Driving Assistance System (ADAS) shows that TensorFlow object recognition machine learning has higher accuracy for recognizing the road signs. Raspberry Pi 3 Model B+ is chosen as the microprocessor and microcontroller in this project due to its low power consumption and convenience. Meanwhile, the Raspberry Pi Camera NoIR Module V2 is used to record the real-time video and send to the Raspberry Pi 3 Model B+ for processing and analysis using TensorFlow machine learning. The result of the analysis that includes the type of the road sign and the percentage of accuracy when compared to its original sign image in its database, will be sent to the Raspberry Pi Display Module for display purpose. Also, to alert the driver, a speaker will be installed together with the Raspberry Pi 3 Model B+ to give an alert sound.

Actions (login required)

View Item