Compound road environment recognition method using camera and laser range finder
Dongwook Seo, Laksono Kurnianggoro, Joko Hariyono, Danilo Cáceres Hernández, Kang-Hyun Jo · 2016
In this paper a road environment recognition method for intelligent vehicle systems is introduced. This is done in order to increase safety since road environment is a complex problem. We introduce the ongoing results of our obstacle classification based on sensor merging using camera and laser range finder sensors. Firstly, a laser-camera calibration is performed. Secondly, objects are extracted using a clustering method in the LRF, then labeled as a pedestrian, vehicle or a bigger object. Thirdly, a verification method is implemented using a combination of histogram of oriented gradients and support vector machines. The experimental results prove the effectiveness of the proposed method for real time applications.