Comparative Study Between Canny-Edge Versus CNN based Lane Detection for High-Definition Map Generation
Nadana Ayzah Azis, Fauzand Mestakindo Erizal, Yul Yunazwin Nazaruddin

Instrumentation and Control Research Group
Department of Engineering Physics,
Institut Teknologi Bandung

Abstract

In order to pursue high-accuracy localization for Autonomous Vehicles (AVs) in semi-open scenarios, a HD map is conducted to assist the sensor fusion in the localization system. Lane/road mark becomes necessary aspect to build an accurate HD map. Lane detection is to detect lanes on the road and provide the accurate location and shape of each lane. It severs as one of the key techniques to enable modern assisted and autonomous driving systems. Lane detection is difficult problem because of varying road condition that one can encounter during driving. This study examines a comparison of the canny-edge based lane detection and deep neural-network based lane detection in order to build an accurate HD Map. Simulations are carried out using datasets available in the community.

Keywords: lane detection, canny-edge detection, deep neural-network, HD map, autonomous vehicles

Topic: Transportation Safety