Semi-supervised Training Set Adaption to Unknown Countries for Traffic Sign Classifiers
Traffic signs in Western European countries share many similarities but also can vary in colour, size, and depicted symbols. Statistical pattern classification methods are used for the automatic recognition of traffic signs in state-of-the-art driver assistance systems. Training a classifier separately for each country requires a huge amount of training data labelled by human annotators. In order to reduce these efforts, a self-learning approach extends the recognition capability of an initial German classifier to other European countries. After the most informative samples have been selected by the confidence band method from a given pool of unlabelled traffic signs, the classifier assigns labels to them. Furthermore, the performance of the self-learning classifier is improved by incorporating synthetically generated samples into the self-learning process. The achieved classification rates are comparable to those of classifiers trained with fully labelled samples.
KeywordsPattern recognition self-training sample selection confidence bands
Unable to display preview. Download preview PDF.
- 1.Chapelle, O., Schölkopf, B., Zien, A. (eds.): Semi-Supervised Learning. Adaptive Computation and Machine Learning. The MIT Press (2006)Google Scholar
- 2.Culotta, A., McCallum, A.: Confidence Estimation for Information Extraction. In: Proc. of Human Language Technology Conference and North American Chapter of the Association for Computational Linguistics (HLT-NAACL), pp. 109–112 (2004)Google Scholar
- 3.Fu, M.Y., Huang, Y.S.: A survey of traffic sign recognition. In: Proc. of the International Conference on Wavelet Analysis and Pattern Recognition (ICWAPR), pp. 119–124 (2010)Google Scholar
- 4.Hillebrand, M., Wöhler, C., Krüger, L., Kreßel, U., Kummert, F.: Self-learning with confidence bands. In: Proc. of the 20th Workshop Computational Intelligence, pp. 302–313 (2010)Google Scholar
- 5.Hoessler, H., Wöhler, C., Lindner, F., Kreßel, U.: Classifier training based on synthetically generated samples. In: Proc. of the 5th International Conference on Computer Vision Systems (ICCV) (2007)Google Scholar
- 6.Jeon, J.H., Liu, Y.: Semi-supervised Learning for Automatic Prosodic Event Detection Using Co-training Algorithm. In: Proc. of the 47th Annual Meeting of the ACL and the 4th IJCNLP of the AFNLP, pp. 540–548 (2009)Google Scholar
- 7.Martos, A., Krüger, L., Wöhler, C.: Towards Real Time Camera Self Calibration: Significance and Active Selection. In: Proc. of the 4th Int. Symp. on 3D Data Processing, Visualization and Transmission (3DPVT) (2010)Google Scholar
- 8.Schürmann, J.: Pattern Classification: A Unified View of Statistical and Neural Approaches. John Wiley & Sons (1996)Google Scholar
- 9.Settles, B.: Active Learning Literature Survey. Computer Sciences Technical Report 1648. University of Wisconsin–Madison (2010)Google Scholar
- 11.Xu, L., Crammer, K., Schuurmans, D.: Robust Support Vector Machine Training via Convex Outlier Ablation. In: Proc. of the 21st National Conference on Artificial Intelligence (AAAI), pp. 536–542 (2006)Google Scholar
- 12.Zhu, X., Goldberg, A.B.: Introduction to Semi-Supervised Learning. Synthesis Lectures on Artificial Intelligence and Machine Learning. Morgan & Claypool Publishers (2009)Google Scholar