Detection of Road Signs Using Tensor Flow and Neural Networks

Almost all of the jobs we conduct in today's world have been eased by automation. Vehicles frequently disregard signs posted on the side of the road because they only want to concentrate on driving, which is dangerous for both them and other drivers. The driver should be alerted of this concern in a way that doesn't force them to shift their attention away from the road. In this situation, traffic sign detection and recognition (TSDR) is crucial since it warns the driver of impending signals. Because of this, not only are roads safer, but drivers also experience more comfort while navigating unfamiliar or challenging routes. A common problem is the inability to read the sign. With the use of this software, driver assistance systems (ADAS) will make it simpler for drivers to comprehend traffic signals. We offer a method for detecting and identifying traffic signs that uses image processing to find signs and an ensemble of Convolutional Neural Networks (CNNs) to identify signs. CNNs have a high recognition rate, making them suitable for a variety of computer vision applications. CNNTSR (Traffic Sign Recognition), a crucial element of modern driving assistance systems that increases driver comfort and safety, uses TensorFlow. CNNTSR is implemented using TensorFlow (Traffic Sign Recognition). This article looks at a piece of technology that helps drivers read traffic signs and steer clear of collisions. The feature extractor and the classifier are two factors that affect TSR accuracy.. Although there are several methods, the majority of current algorithms do both feature information extraction tasks using CNN (Convolutional Neural Network). We develop the identification of traffic signs using CNN and TensorFlow. 43 different types of traffic signs will be used in the dataset for training the CNN. 95 percent of the results will be accurate.