Applying Real-Time Object Shapes Detection To Automotive Traffic Roads Signs

Abstract

This paper deals with a topical issue for researchers, namely the detection of objects in real-time. This is an important topic in the field of computer vision, as the detection and recognition of objects from images require very powerful algorithms. There are different ways to detect objects, for example, based on either the color or shape of the objects. The proposed approach consists in the realization of an algorithm to detect an object in the shape of a triangle, square, and rectangle, applied on traffic road signs for the automotive sector. Many detection methods can be adapted to progress efficiently and robustly, combining these different studies to find the best strategy. Besides, the use of numerous images in the test phase can be used to increase the accuracy of object detection. This approach could be interesting, can be established in the industrial sector, for the conception or development of embedded systems such as Advanced Driver Assistance Systems (ADAS) integrated into autonomous vehicles. The objective of this approach is to help systems based on artificial intelligence for road safety, to have a high degree of accuracy with negligible errors for detecting traffic road signs.