Road obstacles positional and dynamic features extraction combining object detection, stereo disparity maps and optical
- PDF / 4,288,752 Bytes
- 11 Pages / 595.276 x 790.866 pts Page_size
- 104 Downloads / 184 Views
ORIGINAL PAPER
Road obstacles positional and dynamic features extraction combining object detection, stereo disparity maps and optical flow data Thiago Rateke1,2
· Aldo von Wangenheim1,2
Received: 1 February 2020 / Revised: 31 August 2020 / Accepted: 8 September 2020 © Springer-Verlag GmbH Germany, part of Springer Nature 2020
Abstract One of the most relevant tasks in an intelligent vehicle navigation system is the detection of obstacles. It is important that a visual perception system for navigation purposes identifies obstacles, and it is also important that this system can extract essential information that may influence the vehicle’s behavior, whether it will be generating an alert for a human driver or guide an autonomous vehicle in order to be able to make its driving decisions. In this paper we present an approach for the identification of obstacles and extraction of class, position, depth and motion information from these objects that employs data gained exclusively from passive vision. We use a convolutional neural network for the obstacles detection, optical flow for the analysis of movement of the detected obstacles, both in relation to the direction and in relation to the intensity of the movement, and also stereo vision for the analysis of distance of obstacles in relation to the vehicle. We performed our experiments on two different datasets, and the results obtained showed a good efficacy from the use of depth and motion patterns to assess the obstacles’ potential threat status. Keywords Features extraction · Disparity map · Optical flow
1 Introduction Obstacles detection in autonomous and drive-assisted vehicles concerns the detection of any other objects, static or in movement, on or near the road. In an intelligent autonomous vehicle navigation scenario it is, along with path detection, This study was financed in part by the Coordenao de Aperfeioamento de Pessoal de Nvel Superior - Brasil (CAPES) - Finance Code 001. CAPES (Brazilian Federal Agency for Support and Evaluation of Graduate Education). It was also supported by the Brazilian National Institute for Digital Convergence (INCoD), a research unit of the Brazilian National Institutes for Science and Technology Program (INCT) of the Brazilian National Council for Science and Technology (CNPq).
B
Thiago Rateke [email protected] Aldo von Wangenheim [email protected]
1
Graduate Program in Computer Science (PPGCC), Department of Informatics and Statistics, Federal University of Santa Catarina, Florianópolis, SC, Brazil
2
Image Processing and Computer Graphics Lab (LAPIX), National Institute for Digital Convergence (INCoD), Florianópolis, Brazil
one of the most important tasks, because it involves not only the safety of the vehicle where the obstacles detection and recognition are performed, but also because it affects other participants in this scenario, such as other vehicles, pedestrians, cyclists and animals. Based upon information continuously gathered by the obstacle detection, the behavior of an autonomous vehicle must adjust i
Data Loading...
