Vehicle Detection Based on Multi-feature Clues and Dempster-Shafer Fusion Theory

Abstract

On-road vehicle detection and rear-end crash prevention are demanding subjects in both academia and automotive industry. The paper focuses on monocular vision-based vehicle detection under challenging lighting conditions, being still an open topic in the area of driver assistance systems. The paper proposes an effective vehicle detection method based on multiple features analysis and Dempster-Shafer-based fusion theory. We also utilize a new idea of Adaptive Global Haar-like (AGHaar) features as a promising method for feature classification and vehicle detection in both daylight and night conditions. Validation tests and experimental results show superior detection results for day, night, rainy, and challenging conditions compared to state-of-the-art solutions.

Towards explainable deep neural networks (xDNN)

“…We advocate and demonstrate that both, learning and reasoning can work together in a synergy and produce very impressive results. Indeed, the proposed xDNN method outperforms all published results [15], [1], [17] in terms of accuracy. Moreover, in terms of time for training, computational simplicity, low power and energy required it is also far ahead.…”

“…From the user perspective, the proposed approach is clearly understandable to human users. We tested it on some well-known benchmark data sets such as iRoads [15] and Caltech-256 [16] and xDNN outperforms the other methods including deep learning in terms of accuracy, time to train, moreover, offers a clearly explainable classifier. In fact, the result on the very hard Caltech-256 problem (which has 257 classes) represents a world record [1].…”

Global Haar-Like Features: A New Extension of Classic Haar Features for Efficient Face Detection in Noisy Images

“…The proposed technique is expected to be effective on various object detection fields, as we also applied the proposed technique as part of a vehicle detection system [14] with superior results. We highly recommend to use, and to do further study on the proposed global Haar-like features for different kinds of boosting or Haar-based object detectors, not limited to face detection.…”

Actively Semi-Supervised Deep Rule-based Classifier Applied to Adverse Driving Scenarios

“…The iROADS dataset [31] was considered in the analysis. The following dataset contains 4656 image frames recorded from moving vehicles on a diverse set of road scenes, recorded in day, night, under various weather and lighting conditions, as described below: • Tunnel -347 images The iRoads dataset was divided into 90% for training and 10% for validation purposes.…”