Sun R.,Hefei University of Technology |
Chen J.,Academy of Science and Advanced Technology |
Gao J.,Hefei University of Technology
Dianzi Yu Xinxi Xuebao/Journal of Electronics and Information Technology | Year: 2013
Pedestrian detection is a key ability for a variety of important applications, such as robotics, driver assistance systems and surveillance. This paper presents a fast pedestrian detection based on saliency detection and Histogram of Oriented Gradient - Non-negative Matrix Factorization (HOG-NMF) features. The regions of interest are extracted using the frequency tuned saliency detection and threshold based on entropy. A novel HOG-NMF features that reduce significantly the length of feature vector are proposed. Classification method using intersection kernel SVM offers significant improvements in accuracy over linear SVM with the same computational complexity. Experiments on INRIA dataset show that the proposed method reduces significantly runtime compared with HOG/linear SVM and HOG/RBF-SVM, achieves the satisfactory accuracy.
Chen X.,Academy of Science and Advanced Technology |
Sun R.,Academy of Science and Advanced Technology |
Chen J.,Academy of Science and Advanced Technology
Qiche Gongcheng/Automotive Engineering | Year: 2013
A monocular vision-based lane departure warning system (LDWS) is proposed for being used in driver assistance system (DAS) to help driver prevent and avoid accidents. The system is implemented based on field programmable gate array (FPGA) with the key technologies adopted covering adaptive exposure technique for improving the quality of captured images, parallel Hough transform for robustly realizing real-time lane mark detection and simple and effective warning strategy based on the intersection angle between running path and lane marks. The results of experiments demonstrate the robustness of the system in various road scenes and the high detection speed and application significance of the system proposed.