Robust vehicle detection and distance estimation under challenging lighting conditions
Download as DOCX, PDF1 like222 views
This paper presents a real-time monocular-vision-based system for vehicle detection and distance estimation, aimed at avoiding the issues of stereo-vision algorithms. It develops a collision warning system using adaptive features for vehicle detection and distance estimation, maintaining high performance in challenging conditions. Experimental results demonstrate that the proposed algorithm surpasses existing state-of-the-art methods under various weather and lighting situations.
1 of 1
Download to read offline
More Related Content
Ad
Similar to Robust vehicle detection and distance estimation under challenging lighting conditions (20)
Ad
More from I3E Technologies (20)
Robust vehicle detection and distance estimation under challenging lighting conditions
- 1. ROBUST VEHICLE DETECTION AND DISTANCE ESTIMATION UNDER CHALLENGING LIGHTING CONDITIONS ABSTRACT Avoiding high computational costs and calibrationissues involved in stereo-vision-based algorithms, this paper proposesreal-time monocular-vision-based techniques for simultaneous vehicle detection and inter-vehicle distance estimation, inwhich the performance and robustness of the system remaincompetitive, even for highly challenging benchmark datasets. Thispaper develops a collision warning system by detecting vehiclesahead and, by identifying safety distances to assist a distracteddriver, prior to occurrence of an imminent crash. We introduce adaptive global Haar-like features for vehicle detection, tail-lightsegmentation, virtual symmetry detection, intervehicle distance estimation, as well as an efficient single-sensor multifeature fusiontechnique to enhance the accuracy and robustness of our algorithm.The proposed algorithm is able to detect vehicles ahead atboth day or night and also for short- and long-range distances. Experimentalresults under various weather and lighting conditions(including sunny, rainy, foggy, or snowy) show that the proposedalgorithm outperforms state-of-the-art algorithms.