Deep CNN based Road Sign Recognition for Vehicles using Raspberry Pi 3B

A key component of a system of intelligent transportation (SIT) is the method for recognizing road signs. Having the ability to recognise traffic signs precisely a nd effectively can boost driving safety. It primarily aims to detect and organize signs based on their circular shape. A pre-processed image is first used to highlight important details. Also the Hough Transform is used for region recognition and segmentation. On the basis of deep learning, a traffic sign identification method is proposed in this paper. The recognized street signs include then categorized using deep learning. This research proposes a detection of road signs and identification technique based on image processing that combines Convolutional Neural Network (CNN) with road sign classification. Due to its high recognition rate, CNN may be used to solve a number of computer vision jobs. Tensor Flow is used to implement CNN. The proposed model recognizes the circular sign with the greatest accuracy in the German data sets.