Abstract
Despite significant progress in vehicle detection over the last few decades, vehicle detection performance in heavy traffic is still inadequate. In this paper, we propose a new algorithm for vehicle detection in heavy traffic to improve detection performance. It uses two proposed segmentation methods, namely, the disparity map-based bird's-eye-view mapping segmentation method and the edge distance weighted conditional random field (CRF)-based segmentation method. Our experimental results show that the proposed algorithm outperforms conventional algorithms. The improvements in performance range from 10.8 % to 20.5 % increase in F-measure. DOI: http://dx.doi.org/10.5755/j01.eee.20.9.3734
Highlights
Forward vehicle detection is one of the most important technologies in an intelligent vehicle, which is closely related to autonomous driving, collision prevention, etc [1]
We focus on the improvement of vehicle detection performance in heavy traffic based on the proposed segmentation methods
The proposed vehicle detection algorithm was evaluated by application to heavy traffic images captured by our stereo vision system
Summary
Forward vehicle detection is one of the most important technologies in an intelligent vehicle, which is closely related to autonomous driving, collision prevention, etc [1]. We focus on the improvement of vehicle detection performance in heavy traffic based on the proposed segmentation methods. In most cases, extracting exact obstacle areas from images is a very important step for improving vehicle detection performance. Intensity and edges are other important cues for segmentation in small obstacle areas, namely results of disparity map-based obstacle detection and segmentation. DISPARITY MAP-BASED OBSTACLE DETECTION AND BIRD'S-EYE-VIEW MAPPING SEGMENTATION. The results segmented in the bird’s-eye-view are reconverted into the disparity map These disparity map-based segmentations improve detection performance in heavy traffic thanks to accurate identification of obstacle position and removal of unnecessary obstacles. The obstacles can be segmented more accurately and with disparity map-based bird's-eye-view mapping.
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
