PROJECT TITLE:

Robust Vehicle Detection and Distance Estimation Under Challenging Lighting Conditions

ABSTRACT:

Avoiding high computational prices and calibration problems concerned in stereo-vision-based algorithms, this paper proposes real-time monocular-vision-based techniques for simultaneous vehicle detection and inter-vehicle distance estimation, in which the performance and robustness of the system remain competitive, even for highly challenging benchmark datasets. This paper develops a collision warning system by detecting vehicles ahead and, by identifying safety distances to assist a distracted driver, prior to occurrence of an imminent crash. We tend to introduce adaptive world Haar-like features for vehicle detection, tail-light segmentation, virtual symmetry detection, intervehicle distance estimation, and an efficient single-sensor multifeature fusion technique to boost the accuracy and robustness of our algorithm. The proposed algorithm is able to detect vehicles ahead at each day or night and conjointly for short- and long-range distances. Experimental results below numerous weather and lighting conditions (together with sunny, rainy, foggy, or snowy) show that the proposed algorithm outperforms state-of-the-art algorithms.