Efficient Road-Sign Detection Based on Machine Learning

Abstract

Road-signs which indicate the following traffic information can provide important assistance for driving safety in driving assistance system (DAS). In order to detect road-signs, we use a machine learning method in this paper. We first generate database using some image processing steps, then we use HOG for feature description. In the road marking identification step, we use SVM for this task. The designed method is applied using various video images from black box, and is verified to be robust and efficient.