Traffic sign detection and recognition topic using machine learningis one of the most popular topics of image processing and computer vision in recent times, as they play an important role in automatic driving system and traffic safety.Even though traffic sign detection has been studied for a long time and great achievement has been made with the rise of deep machine learning techniques, there are still many problems remaining to be unsolved.Considering difficult real-world traffic scenario, there are two main issues.Traffic signs have some constant characteristics that can be used for detection and classification, among them, color and shape are important attributes that can help drivers obtain road information. First of all, traffic sign boards are usually small size objects, which makes it more difficult to detect than large ones; Secondly, it is hard to distinguish false targets which looks a lot like real traffic signs in complicated street scenario without proper context information. Traffic sign detection and recognition system is majorly divided into two parts. First stage in the process is to localize traffic signs and second stage is to classifythe detected signs. Classification of traffic signs can be achieved by using convolutional neural networks. It can be grouped into two technologies, traffic-sign detection and traffic-sign recognition. Traffic sign classification has become a complicated task due to the rise of advancemethodologies, such as traffic encapsulation and encryption, which decrease the performance of classical traffic classification strategies. Automatic self-driving cars represents a promising technology for the future and it is expected to be safer than now, time saving and to considerenvironmental factors such as pollution control

:Traffic sign, Keras API, Analysis, LE-NET