Computer Vision News - May 2016

Project Traffic Signs Detection and Identification Every month, Computer Vision News reviews a successful project. Our main purpose is to show how varied image processing applications can be and how the different techniques allow to solve technical challenges and physical difficulties. This month we review a traffic signs detection and identification software developed by RSIP Vision’s engineers to enhance road security. Do you have a project in computer vision and image processing? Contact our consultants . COMPUTER VISION NEWS Background In this specific job, one of RSIP Vision’s ADAS projects, we want to collect information found upon speed limit signs in Europe. The goal of the project is to enable an automatized system, which can be based on the simple camera and Central Processing Unit of a smartphone, to perform the tasks of identifying traffic signs and communicating them to another device. The benefits of such a system can be multiple: it can support drivers to respect speed limitations imposed upon their route, by giving them relevant information in real time; it can also transmit to a central database any changes detected on a specific stretch of road at a specific point in time; this data can be used at a subsequent time to inform other drivers. Speed limit signs in Europe are round and feature a red circle along the border; within that circle lays a black number made up of two or three digits and indicating the maximum speed allowed in kilometers per hour (the United Kingdom being the only European country that still defines speed limits in miles per hour). One might think that the simplest way to identify the speed limit would begin with the detection of the red circle. But any software engineer with significant experience in image processing and color processing would reply that the task is not as simple as it sounds. Light conditions during the different hours of the day and the night, as well as shadows, weather condition and visibility are all so changeable that identifying a color on the basis of RGB information coming from the source would prove extremely challenging. 10