Traffic sign detection using Deep learning techniques

Abstract

In this fast-growing world, the trend of self-driving cars and advanced driving systems is increasing rapidly. For humans, identifying a traffic sign is an easy task. However, for self-driving cars, accurate and fast road sign location and classification remains an incredible challenge. So, detection of traffic sign in autonomous vehicles has been catching the attention of the computer vision community incessantly for several decades. With the intensive development of artificial intelligence and computer vision technologies, Deep Learning appears to be one of the most effective solutions for complex detection tasks such as traffic sign detection, which has high demands on accuracy and response speed in multi-object detection. Our proposed work involves in developing a model based on Convolution Neural Network (CNN) in Deep Learning that can be utilized to characterize the traffic signs with great exactness. The performance of our model will be measured by using different metrics like accuracy, F1 score.