Traffic Sign Board Recognition and Voice Alert System using Convolutional Neural Network

Abstract—To ensure a smooth and secure flow of traffic, road signs are essential. A major cause accidents is negligence in viewing the Traffic signboards interpreting them incorrectly. The proposed system helps recognizing sign sending voice alert through speaker to driver so that he/ she may take necessary decisions. trained using Convolutional Neural Network (CNN) which traffic image recognition classification. set classes defined on particular dataset make it more accurate. German Sign Benchmarks Dataset was used, contains approximately 43 categories 51,900 images signs. accuracy execution about 98.52 percent. Following detection by system, sent notifies driver. also section where vehicle alerted near proximity be aware what rules follow route. aim this safety vehicle's driver, passengers, pedestrians. Keywords—Convolutional Network, GTSRB Dataset, Object Detection, Classification Traffic, Signs, Voice Alert