Design and Development of Artificial Intelligence based Real-Time Traffic Sign Identification Scheme using Novel Learning Strategy

The identification of traffic signs, also known as traffic sign recognition or TSR, is one of the most important aspects of research that has to be done before fully autonomous driving systems can be developed. When it comes to the data that are taken in, autonomous driving systems have very specific criteria. The data must be trustworthy and the system must be able to do situational analysis in real time; there is no room for complicated transformations or complex image processing techniques. This endeavour is made considerably more challenging by the fact that it takes place in an urban environment, which features a greater number of relocating or distracting objects (such as traffic signs, ads, parked cars, people, and so on). In spite of the abundance of approaches that have been published, solutions are often only tested on highways, in rural areas, or at very slow speeds. The development of traffic sign detection scheme, which involves the creation of unmanned vehicles, has moved forward in a consistent manner. Finding a way to eliminate interference from various environmental circumstances is one of the most essential technological difficulties. This is necessary in order to ensure that traffic sign recognition and identification can be carried out in an accurate and time-effective manner. However, because of the limitations of colour and edge detection, conventional visual object recognition is almost entirely dependent on the extraction of visual features. In this paper, a novel traffic sign identification is introduced called Learning Strategy for Traffic Sign Detection (LSTSD), in which it follows the Artificial Intelligence (AI) logic to improve the resulting metric to perform accurate identification of road side traffic signs in an intelligent manner. This learning strategy LSTSD is cross-validated with the conventional learning method called Convolutional Neural Network (CNN) to prove the efficiency of the proposed scheme in clear manner.