The detection of road lane markings has many practical applications, such as advanced driver assistance systems, as well as road maintenance. In this paper we propose an algorithm to detect and recognize lane markings from panoramic images. Our system consists of three steps. First after inverse perspective mapping, the potential lane markings are segmented by finding the high-intensity pixels of the image. Second, we extract the inner distance as a feature of each region of interest from the segmented image. Third, we classify features through a Support Vector Machine (SVM). The experiments show that the system is capable of recognizing 97% of the parts of long solid lines and 94% of the dashed lines and can be extended to recognize other road markings such as arrows.
Recognition of Lane Markings in Panoramic Images
2014
6 Seiten, Bilder, 4 Quellen
Conference paper
English
Detection and Recognition of Road Markings in Panoramic Images
British Library Conference Proceedings | 2015
|Detection and recognition of road markings in panoramic images [9407-7]
British Library Conference Proceedings | 2015
Ridgeness for detecting lane markings
Tema Archive | 2005
|Evaluation of Shared Lane Markings
NTIS | 2010
|