Nighttime Vehicle Detection for Heavy Trucks
This paper presents a method for detecting vehicles at nighttime, particularly for an application in heavy trucks. Researchers suggested detecting vehicles at nighttime based on symmetry of taillights, or by training a classifier. The headlights of heavy trucks are very bright which causes that taillights of a vehicle in front appear as being asymmetrical. The bright headlight also defines disturbing information that affects the training of a classifier. We propose an improved threshold algorithm that can effectively remove most non-taillight regions and strengthen the shapes of taillight pairs. Positive and negative samples are extracted from these thresholded (pre-processed) images to train a vehicle classifier by using Haar-like features and AdaBoost. We then detect vehicles in these pre-processed images. Experiments are performed using two alternative methods. Results show that our method, i.e. combining threshold pre-processing and training a classifier, is more accurate and robust than the other two methods.
This work is supported by the Natural Science Found of Shandong NSFSD under No. ZR2013FM032.
- 1.Yen-Lin, C., Chuan-Yen, C.: Embedded vision-based nighttime driver assistance system. In: Proceedings of IEEE International Conference on Transactions of Computer Communication Control and Automation, pp. 199–200 (2010)Google Scholar
- 2.Guo, J.B., Wang, J.Q.: Vehicle detection in nighttime based on monocular vision. Automot. Eng. 36, 574–585 (2014). (in Chinese)Google Scholar
- 4.Klette, R.: Vision-based driver assistance systems. Webster, J. (ed.), Wiley Encyclopedia Electrical Electronics Engineering (2015). doi: 10.1002/047134608X.W8272
- 5.Kumar, U.: Vehicle detection in monocular night-time grey-level videos. In: Proceedings of IEEE International Conference on Image Vision Computing New Zealand, pp. 214–219 (2013)Google Scholar
- 6.Liu, B., Zhou, H.Q., Wei, M.X.: Vehicle detection in nighttime based on color and motion information. J. Image Graph. 10, 187–191 (2005). (in Chinese)Google Scholar
- 7.OpenCV. www.opencv.org.cn
- 9.Rakate, G.R., Borhade, S.R.: Advanced pedestrian detection system using combination of Haar-like feature, AdaBoost algorithm and edgelet-shapelet. In: Proceedings of IEEE International Conference on Computational Intelligence Computing Research, pp. 1–5 (2012)Google Scholar
- 10.Rezaei, M., Terauchi, M., Klette, R.: Robust vehicle detection and distance estimation under challenging lighting conditions. IEEE Trans. Intell. Transp. Syst. 16(99), 1–21 (2015)Google Scholar
- 11.Paul, V., Michael, J.: Rapid object detection using a boosted cascade of simple features. In: Proceedings of IEEE Conference on Computer Vision Pattern Recognition, pp. 511–518 (2001)Google Scholar
- 12.Xue, P., Tong, G., Jing, Q.: Face detection in video based on AdaBoost algorithm and eye location. Video Eng. 35, 114–117 (2011). (in Chinese)Google Scholar
- 13.Zhu, C.: Face detection and tracking based on OpenCV. Comput. Eng. Appl. 48, 157–158 (2012). (in Chinese)Google Scholar